THE GEOCHEMISTRY OF PYRITIC SHALE WEATHERING

WITHIN AN ACTIVE LANDSLIDE

by

Alwyn Vear

Submitted in fulfilment of the requirements

for the Degree of Doctor of Philosophy

Departments of Geography and Geology,

University of Sheffield.

November, 1981.
 SUMMARY

Intense chemical weathering accompanies physical instability at

the site of a repeatedly active landslide in North Derbyshire. In order

to describe and quantify the chemical weathering, a programme of water

sampling and analysis was devised. A sequence of reactions are proposed,

based on theresults of this work, to account for the observed concentra-

tions of chemical species in the drainage waters.

It is thought that pyrite oxidation, accelerated by the presence

of catalytic bacteria, is responsible for the considerable acidity of these

waters. Additional reactions involving carbonates and silicates occur

at strictly comparable rates and consume over 99% of the acidity prior

to the water's emergence in a number of ochre-precipitating springs.

Analysis of the solid reactants and products confirms the suggested

sequence of events and suggests a number of ways in which chemical weathering

might be related to slope stability. Clay minerals appear to be little

affected by weathering, and the growth of precipitate minerals such as

gypsum in joints and on bedding planes might be a more important mechanism

in shale breakdown. After this initial rapid physical disintegration,

chemical weathering, at the surface, proceeds relatively slowly.

Chemical processes build up stresses within the rocks and possibly

help to maintain any inherent planes of weakness. Ultimately landslide

movement is triggered by increases in porewater pressures brought about

by fluctuations in local ground water levels.

 ACKNOWLEDGEMENTS

I wish to thank Dr. Charles Curtis for his continued encouragement,

enthusiasm and invaluable criticism throughout.

Professor R.S. Waters gave considerable support, and the Natural

Environment Research Council provided the necessary funds.

Special thanks also to Margaret, Rod, Villy and Pat - and many

others too numerous to mention - of the Geography and Geology Departments,

University of Sheffield.
 CONTENTS

1. INTRODUCTION 1

1.1 Objectives

1.1.1 Physical Background 1

1.1.2 Chemical Background 2
1.1.3 Initial Objectives 3

1.2 Literature Review 3

1.2.1 Landslides 3
1.2.2 Chemical Weathering: General 5
1.2.3 Chemical Weathering: Sediments 8
1.2.4 Weathering and Slope Stability 11

1.3 Experimental Design 12

1.3.1 Geochemistry 12
1.3.2 Engineering Properties 15

2. THE STUDY AREA 18

2.1 Location 18

2.2 Geology 18

2.2.1 Edale Shales 18

2.2.2 Mam Tor Beds 22
2.3 History (Mining) 24

3. GEOMORPHOLOGY OF THE LANDSL IDE 27

3.1 Landslide Classification 27

3.2 Description of the Mam Tor Landslide 28

3.3 Landslide Management 32

4. WATER SAMPLING AND ANALYSIS 35

4.1 Sampling Design and Procedure 35

4.2 Sample Storage 37

 4.3 Sample Analysis 38

4.3.1 Temperature, pH and Conductivity 38

4.3.2 Cations 40
4.3.3 Anions 40
4.3.4 Trace Metals 40

4.4 Results 41

4.4.1 Temperature, pH and Conductivity 41

4.4.2 Cations 43
4.4.3 Anions 44
4.4.4 Trace Metals 46

5. INTERPRETATION OF WATER CHEMISTRY DATA 47

5.1 Environmental Responses 47

5.1.1 Rainfall 47
5.1.2 Temperature 49

5.2 Cluster Analysis 50

5.3 Charge Balance 53
5.4 Origin of Drainage Waters 55

5.4.1 Deep and Surface Components 55

5.4.2 Dilution Gauging 57
5.4.3 Additional Inputs 59

6. DEVELOPMENT OF A WEATHERING MODEL 62

6.1 Weathering Reactions 62

6.1.1 pyrite Oxidation 62

6.1.2 Carbonate Dissolution 66
6.1.3 Clay Mineral Transformations 67
6.1.4 A Quantitative Weathering Model 68
6.1.5 The Sulphate Problem 71

6.2 Wider Implications of the Model 72

6.2.1 Catalysts 72
6.2.2 Acid Mine Drainage 75

6.2.2.1 Formation 74
6.2.2.2 Abatement 75

6.2.3 Comparison with Mam Tor Drainage Waters 77

 6.3 Environmental Considerations 79

6.3.1 Phytosociological Responses 79

6.3.2 Biological-Chemical Interactions 82

6.4 Conclusions 83

7. MINERAL ANALYSIS 85

7.1 Samp lirig 85

7.1.1 "Little Mam Tor" 86

7.1.2 Retrogressive Units 86
7.1. 3 Other Surface Materials 87
7.1.4 Borehole Cores 88

7.2 Particle Size Separation 89

7.2.1 Disaggregation 89
7.2.2 Sedimentation 90

7.2.2.1 Theory 90
7.2.2.2 Practice 92

7.3 X-Ray Diffraction 93

7.3.1 Principles 93
7.3.2 Mineral Identification in Whole Rock
Powder Samples 95
7.3.3 Identification and Estimation of Clay
Minerals 96
7.3.4 Other Materials 99

7.4 XRD Results 100

7.4.1 Whole Rock 100

7.4.2 Clay Minerals 102
7.4.3 Other Materials 106

7.4.3.1 Carbonate Coatings 106

7.4.3.2 "Allophane" 106
7.4.3.3 Ochre 107

7.5 Thin Sections 108

 7.6 The Mechamisms of Formation of Three Precipitate
Minerals 109

7.6.1 Gypsum 109

7.6.2 Jarosite 111
7.6.3 Iron Hydroxide 113

8. CHEMICAL THERMODYNAMICS 116

8.1 Temperature Considerations 117

8.2 Mineral Stability and Eh-pH Diagrams 120

8.2.1 The Nernst and Debye-Huckel Equations 120

8.2.2 Stability of Water 122
8.2.3 The Iron Oxides 123
8.2.4 Iron Hydroxides 125
8.2.5 Relationships Among Ions 126
8.2.6 Dissolved Sulphur Species 128
8.2.7 The Composite Diagram 129

8.3 Further Consideration of the Eh-pH Diagrams 130

8.4 Gypsum Solubility 133

9. CHEMICAL WEATHERING AND SLOPE STABILITY 136

9.1 Dissolution and Precipitation 136

9.2 The Mechanism of Shale Breakdown 138
9.3 Porewater Pressure and Failure Surfaces 140
9.4 Future Management 141

10. FINAL CONCLUSIONS 144

11. REFERENCES 148

APPENDICES

A. WET CHEMISTRY ANALYTICAL PROCEDURES 167

Ai Determination of Calci~m 167

A ii Determination of Magnesium 168
A iii Determination of Sodium 169
A iv Determination of Potassium 170
 A v Determination of Total Iron 171
A vi Determination of Ferrous Iron 172
A vii Determination of Aluminium 173
Aviii Determination of Chloride 174
A ix Determination of Sulphate 175
A x Determination of Trace Metals 177

B. RESULTS OF THE WATER ANALYSES

0
B i Temperature ( C) 180
B ii pH and Rainfall (- log [Hj and mm respectively) 181
B iii Conductivity (VS cm-I) 182
2+ -1
B iv Calcium, Ca (Vg m1 ) 183
. 2+ -1
B v Magnes1um, Mg (Vg m1 ) 184
+ -1
H vi Sodium, Na (~g ml ) 185
+ -1
B vii Potassium K (~g ml ) 180
-1
Bviii Total Iron, Fe (~g m1 ) 187
B xi Aluminium, A13+ (~g m1-l) 188
B x Chloride, Cl - -1
(Vg ml ) 189
2- -1
B xi Sulphate, S04 {Vg m1 ) 190
-1
B xii Trace Metals (~g ml ) 191

C. MINERAL ANALYSIS

C i Sample Description and Treatment 192

C ii Mineralogy of the Samples 197
C iii Comparison of Pure Mineral Samples, After Brown
(1961), with Selected Samples From Mam Tor 198
 LIST OF FIGURES Preceeding
Page

Figure 1 Geology and Landslides of the Edale-Mam Tor Area. 19

Figure 2 Schematic Section Through Shafts and Soughs of 26

Odin Mine.

Figure 3 A Landslide Classification Scheme. 28

Figure 4 The Major Features of a Rotational Slump 28

Figure 5 Location and Major Geomorphological Features of

the Landslide at Mam Tor, Derbyshire. 28

Figure 6 Longitudinal Section through the Landslide and

Retrogressive Units. 28

Figure 7 Total Dissolved Solids - Conductivity Relationship. 43

Figure 8 Cluster Analysis Hierarchy. 51

Figure 9 Precipitation Totals for the Study Period. 53

Figure 10 Variations in Temperature. 53

Figure 11 Variations in pH. 53

Figure 12 Variations in Conductivity. 53

Figure 13 Variations in Calcium Content. 53

Figure 14 Variations in Magnesium Content. 53

Figure 15 Variations in Sodium Content. 53

Figure 16 Variations in Potassium Content. 53

Figure 17 Variations in Total Iron Content. 53

Figure 18 Variations in Chloride Content. 53

Figure 19 Variations in Sulphate Content. 53

Figure 20 The Derivation of the Seven Groups of Waters. 57

Figure 21, Principles of Dilution Gauging. 58

Figure 22 Sequence of Chemical Reactions Affecting the 71

Composition of Water at Site 7.

Figure 23 Soil Profile Above Retrogressive Units. 87

Figure 24 Location of Samples in Borehole 7 and 8 89

Figure 25 A Method for Separating <2~m clays. 92

Figure 26 Examples of Powder Diffraction Traces. 100

 Preceeding
Page

Figure 27 Effects of Glycolating and Heating a

< 0.5 11m Sample (BH89) 102

Figure 28 Effects of Glycolating and Heating a

< 0.5 11mSample (SP2) 102

Figure 29 Comparison of Four Goethite Samples 108

Figure 30 Eh-pH Diagram for Water and Iron Oxides 123

Figure 31 Eh-pH Diagram for Iron Hydroxides 125

Figure 32 Composite Eh-pH Diagram Illustrating the 'Solubility'

of Ferric and Ferrous Ions. Total Activity of Iron =
10-3 molar 128

Figure 33 Eh-pH Diagram for Dissolved Sulphur Species. Total

Activity of Sulphur = 10-1 molar 129

Figure 34 Composite Eh-pH diagram for the system S-Fe-H20

[S] = 10-1 m, .!.Fe] = 10-3 m 129

Figure 35 Method of Construction of Field Boundaries and the

Plotting of Eh-pH Values of Mam Tor Waters. See
Table 22 for Description of Samples 1-4 129

Figure 36 Eh-pH Diagram for Iron Hydroxides and Sulphides,

Including Jarosite. After Van Breemen, 1972 130

Figure 37 Eh-pH Diagram Showing the Field of Natural Waters.

After Baas-Becking et al. 1960 133

Figure 38 Pyrite oxidation model. 144

 LIST OF PLATES
Preceeding
Page
Frontispiece From: Peak Park News, Autumn 1977

Plate 1 A General View of the Mam Tor Landslide. 2

Plate 2a Carbonate Concretion from the Retrogressive 20

Units.
2b Photomicrograph of 'Fresh' Edale Shale. 20

Plate 3a Photomicrograph of 'Weathered' Edale Shale. 20

3b Detail from Thin-Section of Sample G2. 20
Plate 4a Euhedral Gypsum in Thin-Section. 20
4b Gypsum Crystals in Hand Specimen. 20
Plate .5 Large Gypsum Crystal on Weathered Shale. 21

Plate 6a Edale Shale Scree Material. 23

6b General View of the Turbidite Succession. 23

Plate 7a Sole Marks on the Base of a Sandstone Unit. 23

7b Mam Tor, Showing Geological Boundary. 23

Plate 8 'Little Mam Tor'. 25

Plate 9 Retrogressive Units. 25

Plate lOa Odin Mine and Gank Mouth Cave. 25

lOb Crushing Circle and Knowles Engine Shaft. 25
Plate lla The Odin Sitch. 25
llb White Precipitate in the Odin Sitch, 25

Plate 12 Aerial Photograph of the Mam Tor Landslide. 29

Plate 13a The Main Failure Surface. 30

13b A Portion of the Slickensided Failure Surface. 30

Plate 14a Seepage at Site 16. 32

14b Seepage at Site 3. 32

Plate 15a Encroachment at B1acketley Barn. 32

l5b Tilting of Fences Due to Movement at the Tip. 32

Plate 16 Downward Movement of the Road on the Upper

Section. 32
 Preceeding
Page '

Plate 17 Upward Movement of the Road on the Lower Section. 32

Plate 18 Further Examples of Road Damage Due to Movement. 32

Plate 19 Severe Road Damage and 'Tarmac Stratigraphy'. 32

Plate 20 Gypsum Precipitation on Plants in the Seepages. 36

Plate 21a Joint-Fill Material on the Retrogressive Units. 87

21b A Closer View 87

Plate 22a Sedimentation Tank 92

22b Thin-Section of Weathered Eda1e Shale. x 5 Negative
92
View.
148
Plate 23 The End:
 LIST OF TABLES
Facing Page

Table 1 Comparative Dimensions of Mam Tor and

21 Edale Landslides. 28

Table 2 The Effects of Sample Storage and Acidification

on Water Quality. 38

Table 3 Summary of Water Temperature Data. 42

Table 4 The Changing Proportions of Ferrous:Ferric

Iron on Sample Storage. 44
Table 5 Basic Statistics for Water Variables. 45
Table 6 Correlation Matrix for Water Variables. 46

Table 7 Median Monthly Bulk Precipitation Concentrations,

1959-64 (After Stevenson, 1968) Compared with
Values for Two Samples From Mam Tor. 47

Table 8 Characteristics of the Seven Cluster Analysis

Groups. 51

Table 9 Charge Balances Calculated for Group-Averaged Data. 54

Table 10 A Quantitative Mixing Model for 'Group 6' Waters. 55

Table 11 Comparison of Five Acid Mine Drainage Waters. 77

Table 12 Species List For Selected Water Sampling Sites. 79

Table 13 Operating Conditions for the X-Ray Diffractometer. 95

Table 14 Peak Areas and Proportions of Clay Minerals in the

<0.5 llm Samples. 103

Table 15 Results of the Gypsum Precipitation Experiment. 111

Table 16 Values of Standard Free Energies of Formation. 117

Table 17 Values of Parameter ai in the Debye-Huckel Equation. 121

Table 18 Molarities and Activities of Ions in the Site 7 Water;

Based on the Value of I = 9.34 x 10-2 122

Table 19 Reactions Relating Ionic Activities to Hematite and

Magnetite. 128

Table 20 Reactions Between Diss~lved Sulphur Species for

[Total Sulphur] = 10- • 129

Table 21 Reactions Between Ions, Dissolved Sulphur Species and

Solids. 130
Table 22 pH and Redox Measurements on Four Mam Tor Samples. 131

Table 23 Volume Changes Involved in the Three Principle

Reactions 136
 1

1. INTRODUCTION

1.1 Objectives

1.1.1 Physical Background

Mam Tor, in North Derbyshire, is notorious both locally and nationally

as the site of an extremely active landslide (Plate 1). Episodes of move-

ment possibly date back some ten or eleven thousand years to the end of the

last glaciation in this country.

The precise dating of ancient mass-movement is difficult unless organic

material can be found beneath the landslide and subjected to radiometric

dating. No such material has been found at Mam Tor, but the landslide

does rest on solifluction deposits dated as late Pleistocene, for this

region, by McArthur (1970) and Said (1969), which places a maximum age

on the landslide of 10,000 years B.P. The presence of an Iron Age hill

fort on the summit, which appears to have been dissected by the landslip

scar, would suggest a minimum age of at least 2.S to 3 thousand years.

Recent archaeological excavations, however, have uncovered Bronze Age

artifacts from within the 16 acre site and it is thought that occupation

may go back still further (Coombs, 1971). Also it is not known for certain

that the fort was not originally constructed using the landslide scar as a

natural defensive flank; in which case the initiation of landsliding is

pushed back still further into prehistory.

In 1802, a major highway was constructed across the landslide (now

the A625) by the Manchester to Sheffield Turnpike Company to by-pass the

older route through Winnats Pass which has a gradient of one-in-five. The

Peak Park Joint Planning Board (1977) observe that since the road was

~pened in 1802: "Records show that the slip has been more or less con-

tinually active and repairs have been carried out annually since its
PLATE 1: A general view of the Mam Tor Landslide
looking west from the village of Castleton.
 2

opening which in turn have caused closure of the road on an almost regular

basis." More recently the route has been closed for major repairs in 1912,

1933, 1946, 1952 and 1966 (Reed, 1977). In the latter case the rate of

landslide movement, and of certain sections of the road, exceeded 0.5 m

day-l (Brown, 1977) and the road was closed for two weeks. Following

similar movements in March, 1977, the A625 was reduced to a single lane

controlled by traffic lights and closed to heavy traffic which, it was

believed at the time, contributed to the repeated failure of certain

sections. After two years of weight restrictions, however, the road was

severely damaged once again, and in January 1979 the route was permanently

closed to all through traffic when parts of the road were displaced vertically

by more than one metre.

An alternative route is still being sought (see Dunn, 1979).

1.1.2 Chemical Background

Prior to the commencement of this research it was noticed by Curtis

(1977, personal communication) ,and no doubt by others, that in addition to

this repeated physical instability Mam Tor is also the site of considerable

chemical activity. The evidence for this lies in the fact that certain

streams draining the landslide precipitate ochre upon mixing with atmospheric

oxygen or with other streams. Further up on the major slipped units of the

landslide, where bedrock is exposed due to fracturing, gypsum (CaS04) can

be seen in hand specimen as a major constituent of the rocks. Clearly this

mineral could not have been an original component of the sediment (Middle

Carboniferous mudstone) as no evaporite deposits of this age are known in the

area.

1.1.3 Initial Objectives

The above observations led, at the outset, to the formulation of

the two principal objectives of this research. Firstly, the subsequent work
 3

should attempt to describe and possibly even quantify the geochemical

reactions taking place within the landslide and giving rise to these

somewhat unusual products. Secondly, since both intense physical and

chemical activity are occurring in the same small area, the research

should investigate the possibility of links between chemical weathering

and certain aspects of the geomorphology.

Prior to a discussion of the experimental procedures followed and

results obtained, a review of the existing body of literature concerning

certain related topics will help to place this research in its wider

scientific context.

1.2 Literature Review

1.2.1 Landslides

Mass movements in general, and landslides in particular, have been

studied for many years. All too often such studies are essentially des-

criptive and have been conducted after a major disaster has occurred; such

as that at Aberfan in 1966 (see Bishop et al., 1969). Certain workers,

however, have studied the actual mechanisms which cause landslide movement

and even investigated the possibilities of predicting landslides in any

giVen situation (Hutchinson, 1968; Terzaghi, 1950).

Much of the research on landslides comes from the discipline of civil

engineering, where damage to structures (or, more precisely, the prevention

of Such damage) is the prime concern. An early, and somewhat c1assic,work

dealing with the geological and geomorphological aspects of landslides is

presented by Sharpe (1938, reprinted 1960).

More recently Piteau (1970) looks at certain geological controls on

landsliding and stresses the importance of discontinuities in bedrock such

as bedding and, in particular, jointing. Piteau notes that weathering

Products often accumulate in joints and the stability of the landslide

 4

mass as a whole is then dependent on the shear strength of this gouge

material. The weakest link of all may be at the gouge/bedrock interface.

Similarly, Nonveiller (1967) in a study of the Zalesina and Vajont

landslides in .Yugoslavia and Italy, respectively, concludesthat: "The

investigation of stability conditions of rock slopes should be based on

a study of the systems of jointing and their relation to possible failure

surfaces as well as on comprehensive investigation of the strength para-

meters of the decomposed material filling the joints".

Most workers are agreed that the ultimate cause of slope failure is an

increase in porewater pressure in the bedrock. In an excellent and up-to-

date review of work on landslides, Cooke and Doornkamp (1974)not~ that

"there is seldom only one cause of slope instability". They envisage a

sequence of events resulting in landsliding only when "The forces creating

movement exceed those resisting it (shear strength of materials)". It is

important, therefore, to distinguish between factors contributing to in-

creased shear stress (disturbing forces) and those contributing to low

shear strength(resisting properties). Cooke and Doornkamp list four

factors which produce shear stress:

1. The removal of lateral or underlying support

(especially by weathering)

2. Increased disturbing forces.

3. Transi tory earth stresses (e.g. earthquakes)

4. Increased internal pressure (particularly porewater

pressure).

Three factors which leadto decreased shear resistance are given as:

1. Properties of materials (degree of fracturing,

jointing, etc.).

2. Weathering changes (particularly water absorption

by clay minerals).

3. Porewater pressure increase.

 5

The ultimate stability of a slope is thus given by the Safety

Factor:

Fs = EFopces Resisting8lope Failure (1)

EDisturbing Fopces
Clearly, if Fs ~ 1, the slope is stable (or conditionally stable), and when

Fs < 1 it is unstable and failure is imndnent. Terzaghi (op.cit.) believes

that by performing engineering tests on materials the value of such a

Stability Factor can be calculated and the stability (or otherwise) of a

given slope determined.

In his paper of 1950, Terzaghi also discusses the lubricating effect

of water on slip planes; a mechanism generally thought to be responsible

for the initiation of movement. He shows, however, that many common minerals,

such as calcite and quartz, actually 'increase their resistance to sliding if

they become wet. In addition, the author points out that only a very thin

film of any lubricant is needed to produce the full lubricating effect, and

the required amount is nearly always present in bedrock anyway, even in

relatively arid areas. It would seem, therefore, that lubrication is not

nearly so important a control as changes in porewater pressure; a fact

which is demonstrated using a model landslide by Curtis and Williams (1978).

Terzaghi ends his discussion of the mechanisms, responsible for 1and-

slide movement with a plea for more co-operation between engineers and

geologists, since their 'two disciplines approach a common problem from

different origins. With the continued growth of engineering geology as a

subject in its own right since the 1950's, some common ground.seems finally

to have been established.

1.2.2 Chemical Weathering - General

Weathering: "Changes in the degree of consolidation and in composition

which take place in the earth's crust within the sphere of infZuence of
atmospheric and hydrosphePic agencies." (Jackson and Sherman, 1953).
 6

A considerable volume of literature concerning chemical weathering

of rocks has been produced both prior and subsequent to Goldich's (1938)

classic paper, describing a weathering sequence of common rock-forming

minerals ('stability series') which is. essentially the reverse of Bowen's

reaction series (Bowen, 1928). This article, however, in common with the

vast majority of others, is concerned with the weathering of igneous and

metamorphic rocks. Comparatively few studies until recently have paid

attention to sediments. Of these, most are devoted to carbonate rocks and in

particular to the development of karst features. Loughnan (1962) believes

that Goldich's stability series is in agreement with a classification of

silicates based on the number of silicon - oxygen bonds. He considers

mineral structure to be one of two factors which influence weathering,

the other being the mobilities of the bonding cations. The solubility

of these cations is said to be a function of five inter-related environ-

mental factors: pH, Eh, leaching potential, time and fixation.

Perhaps the most common approach to chemical weathering - because of

the ease with which laboratory simulation can be conducted - is the study

of interactions between individual minerals and aqueous solutions. Wilson

(1975) reviews such studies in relation to soil formation and fertility,

with special reference to the influence of crystal structure on the

conversion of micas to kaolinite (see also Andersson and Wiklander, 1975).

Similarly Correns (1961; 1963) uses individual minerals, first ground to

a fine powder in order to increase the active surface area, and then placed

in a continuous-flow apparatus. He concludes that: "Dissolution is

affected by the speed with which the solution moves through the soil or

weathering product".

A more realistic attempt at simulating the natural weathering environ-

ment is made by Barnhisel and Rotromel (1974) in their study of the weathering

of clay minerals in a coal mine spoil bank. This is done by subjecting the
 7

material to dilute acid solutions for periods of up to six months. The

mode of attack of the acid is found to be at the edges of the clay minerals

whereupon AI, Fe, K and Si ions are released into solution. The authors

conclude that: "Even though the drastic acid conditions in strip-mine

spoils may dissolve the clay minerals and release potentially toxic ions
3+
such as Al ,the relative mineralogic composition as a result of this

chemical weathering should not change significantly." This presumably is

because the rates of dissolution of the original minerals (micas) and weathered

products (kaolinite) are essentially the same.

In recent years work on geochemical weathering has tended to move away

from descriptive studies and attempts.at simulating the weathering environ-

ment (which often left much to be desired) towards a more systematic and

mechanistic approach. Detailed input/output analyses on a well-defined

(therefore, usually small) catchment area, such as. those described by Bricker

et al. (1968) and Cleaves et al. (1970) for the Hubbard Brook experimental

catchment, are typical of the systematic approach. Laboratory weathering of

the same rocks which underlie the catchment (in this case pelitic schists)

accompany field based experiments. Agreement between the two sets of data

is generally very good. The results of these studies have shown that

silicate-water reactions are much more rapid than previously thought,

although the calculation of erosion rates for certain valley features does

not take into account drastic climatic changes which have occurred over the

millennia.

Thermodynamic approaches to chemical weathering are aimed at describing

reactions using standard free energy data. Krauskopf (1967, Chapter 8)

defines the free energy change for a given reaction as being the sum of

the free energies of formation of all the reaction products minus the

sum of the free energies of the reactants (in their standard states).

This may be represented by:

 8

o f (products) - !J.G
0 (
= !J.G f reactants)
(2)

o o
where !J.G ris the standard free energy of reaction and !J.G fthe free energy
o
of formation. If !J.G ris negative (due to the conventional way of writing

thermodynamic equations) the reaction will be spontaneous since it requires

no external energy for initiation. The likely extent of a reaction and even

the 'rate' or intensity can be estimated from further considerations of

free energies and 'activation energies'.

In such a thermochemical approach to weathering Curtis (1976) considers

the "special position" of oxidation reactions. An example quoted is the

oxidation of pyrite in the presence of water:

(3)

The free energy of reaction (from Equation 1) is -17.68 k cal g.atom-1

which Curtis notes as being "of the same order of magnitude as hydrocarbon

combustion" (approximately -20 k cal g.atom-l).

1.2.3 Chemical Weathering: Sediments

Very few studies have been conducted concerning the weathering of

natural sediments, other than carbonates. Mudrocks in particular have

received little attention. When considered at all the weathering environ-

ments described, or the experimental conditions set-up, are invariably far

removed from those of the 'natural' situation.

One such work is the comparative study of Laguros, et al. (1974) of

natural and simulated shale weathering. The ultrasonic disaggregation

technique they employ can only be of use in predicting the durability of

shales in engineering tests, and the authors concede that: "it is not a

predictive test where the mineralogical characteristics are of significance".

Spears and Taylor (1972) investigate the weathering of shales in coal

mine spoil heaps. In their study they suggest that the mechanism of weathering
 9

in the natural, rather than simulated, environment is different from that

proposed by Barnhisel and Rotromel (op.cit.). Spears and Taylor believe

that after an initial relatively rapid breakdown along bedding and lamina-

tions the material remains largely unaffected by chemical weathering. In

another paper, the same authors note that the rates of chemical weathering

are controlled by this physical disintegration which increases the surface

area of particles (Taylor and Spears, 1970). They consider the expansion

of mixed-layer clays to be a more likely mechanism in breakdown than actual

'dissolution' as suggested by Barnhisel and Rotromel. It is reasonable

to believe that the disintegration, particularly of fine-grained and finely

laminated rocks can result from clay mineral expansion - an essentially

physical process. Another mechanism is the growth of secondary minerals

between the laminations and in. joints. This action combines both chemical

and physical processes of crystal growth and expansion leading ultimately

to the build-up of stresses in the rocks. The actual growth of crystals from

saturated pore solutions can occur against pressures of up to 47 bars, and

the expansion of these crystals on hydration against 63 bars (Goudie et al.,

1970; Cooke and Smalley, 1968). Since the tensile strength of rocks is

commonly between 20 and 200 bars Goudie et al. believe disintegration as

a result of these processes to be highly likely, especially close to the

surface where the overburden is slight. In one of the few reported

weathering experiments to include mudrocks these authors find that strong

salt solutions are not necessarily needed to cause splitting. During the

course of their wetting-drying/heating-cooling procedure the only fresh

shale samples to split were the controls which had been treated only with

distilled water.

Volume increases accompanying the formation of weathering products

often occur in sulphide-bearing shales. As early as 1920 damage to certain

structures was recognised as being caused by'sulphide mineral expansion

 10

(Anon, 1960) and much has since been written concerning the problem of

'heave'. This phenomenon seems to be particularly prevalent in Penn-

sylvania where pyritic shales underlie the foundations of many buildings

in the Pittsburgh area (Dougherty and Barsotti, 1972).

Numerous possible mechanisms for causing these shales to heave have

been suggested, but Coveney and Parizek (1977) believe that "gypsum for~

ation is the prime cause". They observe that in parts of Kansas City's

underground development sites the floor has risen in places as much as

6-8 inches over a period of several years since excavation. They further

state that: "Major mineral constituents of the weathered or the fresh

Hushpuckney Shale are similar but the minor mineral constituents are

quite different. Most notable is the presence of abundant gypsum which

occurs in fractures or partings in the shale". Dougherty and Barsotti

also attribute heaving to crystal growth on bedding planes of fissile

shales which force the bedding apart and cause overlying rocks to buckle.

If there is sufficient overburden (10 feet seems adequate) heaving is not

detected. When the authors tried to simulate such crystal growth in the

laboratory, however, they failed to precipitate gypsum on bedding planes.

It is felt that this is a function of the equipment used (the shale samples

were not completely confined) since gypsum did form in other parts of the

apparatus. When the samples were successively wetted and dried with dilute

H2S04 an uplift pressure of 500 pounds per square foot was noted.

Coveney and Parizek (op.cit.) observe that the sulphur bearing minerals

in weathered rock occupy nearly twice the volume of the primary sulphides in

fresh rock (due to the different densities of gypsum (SG = 2.36) and pyrite

(SG = 5.02». Fassiska et al. (1974) study changes in the molecular packing

arrangements during the pyrite-gypsum transformation. They note "an approxi-

mate volume increase of 350%" which was increased further by hydration and

the addition of bulky water molecules.

 11

These volume expansions and resultant pressure increases agree well

with the cases of salt weathering quoted by Goudie et al. (1970) and are

not dissimilar to the process of physical breakdown proposed by Spears

and Taylor (1972). Many more examples of shale heaving due, in the first

instance, to chemical weathering of sulphides can,be found in the literature

of engineering geology. The interested reader is referred to Grattan-

Bellew and Eden, 1975; Quigley and Vogan, 1970.

1.2.4 Weathering and Slope Stability

Extensive exposures of Tertiary mudstones occur throughout Japan.

The close association of these rocks with landslides has prompted research

into possible links between bedrock weathering and slope stability.

Geologists commonly believe that the association is simply due to

the fact that changes in the position of the water table induce changes

in porewater pressures, lowering the overall strength of the material and

leading ultimately to failure.

This connection has indeed been shown to exist, although the mechanistic

steps are more complex.

Nakano (1967) demonstrates that certain shales slake more readily if

repeated wetting and drying takes place (this is somewhat similar to the

findings of Goudie et al., op.cit.). Such a process operates continually

in the zone o f fluctuating groundwater level.

In a study of 185 landslides in the Nagano Prefecture, Nakano shows

that a strong correlation exists between frequency of landslides and the

occurrence of mudstones which slake on wetting. He believes that heavy

rainfall (and particularly snow melt) and increased pore pressures simply

act as a trigger mechanism on sliding surfaces which have already developed

due to weathering of the shales.

 12

Similarly Yoshioka (1975) believes that the formation of clay minerals

by weathering (of a tuff layer between more consolidated rocks) leads to

a lowering of cohesive strength through expansion and weakening of the

constituent minerals and leads ultimately to 1ands1iding. In a study of

the same area (the Kamenose landslide) Kitano et al. (1967) describe how

the solute content of the groundwater gives an indication of the degree

of weathering of the bedrock and the likelihood of landslides occurring.

They do not, however, consider the weathering reactions producing the

solutes.

Little direct work has been done, outside Japan (where the standard

of research often leaves a great deal to be desired) on the links between

weathering and slope stability and examples from the existing literature

often appear to be contradictory.

For example, in a consideration of factors causing landslides, Cooke

and Doornkamp (op.cit.) suggest that the free flow of water through joints

"prevents the build-up of locally excessive water pressures, and thus

decreases the likelihood of instability". If such water is chemically

reactive, however, secondary minerals may precipitate.in the joints (as

described by Piteau, op.cit.) and cause increased stress. The subsequent

dissolution of these minerals (salt, gypsum, dolomite, etc.) at a later

date or in another part of the landslide may even decrease the shear

resistance and, consequently, the safety factor.

It is hoped that the work described in the following sections might

help to resolve some of these differences of opinion and present a generally

acceptable view of links between chemical weathering and slope instability.

1.3 Experimental Design

1.3.1 Geochemistry

The chemically-charged streams draining the Mam Tor landslide hold the

key to weathering reactions taking place at depth - the products of such

 13

reactions inevitably being carried away in solution. By sampling and

subsequently analysing the drainage waters on a regular basis the pro-

portions of the various dissolved species can be .determined. Reactions

between groundwater and bedrock can then be suggested to account for the

observed composition. Therefore, a detailed sampling strategy was developed,

including control sites from the surrounding area, which is described inmore

detail in Section 4.

In order to fulfil the first objective of this research (Section 1.1)

it was considered necessary also to monitor the precipitation input to the

landslide and stream discharge from it. If the quantities of dissolved

solids at each stage can be determined then a solute budget-type study can

be undertaken and the overall rates of chemical weathering calculated (see

Bricker, et al. and Cleaves et al., op.cit.).

Precipitation totals are easily monitored using a standard Meteorological

Office rain gauge. Such an instrument was installed at Mam Tor prior to

the commencement of the main sampling programme. Recordings were checked

against those obtained daily at the Meteorological Office Station No.106852

at the nearby village of Hope (GR. 4168.3822).

The quantitative assessment of solute inputs to a catchment from rain-

fall, however, is more difficult to achieve than a simple volume estimate.

Specially designed rainfall samplers constructed from chemicallyinert plastics

must be used to intercept·and store the rainfall until chemical analyses can

be performed. Contamination of samples held in such apparatus is reported

by many workers (Allen et al., 1968; Gore, 1968) and measures taken to prevent

birds perching on the rim of the collecting funnel, or insects falling in,

generally prove unsatisfactory.

In addition, there is no secure site for such an obtrusive structure as

a rainfall sampler (out of the reach of inquisitive cattle, sheep and tourists)
 14

anywhere on the Mam Tor landslide. In view of these problems it was

decided that continuous monitoring of rainfall quality would not be

conducted. Spatial variations in the chemistry of rainfall in Britain

are in any case slight (Stevenson, 1968) and the quantitites of solutes

likely to be added are considered small in comparison to the amounts

contained in the Mam Tor streamwaters (which must be highly concentrated

in order to precipitate ochre and gypsum, as noted in Section 1.1.2).

Average compositions of precipitation in the Pennines are well documented

in the literature (see Allen et al., 1968; Crisp, 1966; Rippon, 1980).

Stream discharge is most easily monitored using some standard gauging

device such as a flume or weir. These structures restrict the natural

flow of a stream, confining it to a narrow channel or to emerge through a

restricted opening. The 'head' of water behind the structure (in the case

of a weir) is proportional to the discharge, measured in litres per second

or cubic metres per second - "CUMECS". British Standard No. 3680 gives

details of the construction of flumes and weirs (B.S.I. 1965).

Ideally the discharge at several streams on Mam Tor should be monitored

and small, easily removable 'V'-notch weirs were designed for this purpose.

However, the National Trust (who own over 90% of the land at Mam Tor) would

not grant permission for the excavation of small approach channels or the

installation of the weirs, and the work could not go ahead.

Since the majority of the 'streams' are in fact just shallow seepages

none of the other standard methods of discharge measurement which use a

velocity-area technique could be employed. The only alternative remaining

was to estimate discharge by a 'dilution gauging' method. Details of this

technique are given in Section 5.4.2, but it should be noted at this stage

that dilution gauging could not be conducted at many of the stream sites,

particularly during the period of streamwater analysis. The problem lies

in the principle of the technique, which requires the injection of a

 15

saturated salt solution into the streamwater. Clearly if this same

water is being sampled for chemical analysis such a procedure might

seriously affect the results. Also, since many of the streams precipi-

tate secondary minerals the addition of such a concentrated solution

might lead to supersaturation and cause.further deposits to form. In

view of such possibilities discharge measurements were only made after the

main period of study had been completed, and then at only one of the

sampling sites (see Section 5.4).

Consequently, a quantitative solute budget study such as that

conducted by Cleaves (op.cit.) could notbeundertaken at Mam Tor. It

should also be noted that a further difficulty would have been encountered

in delimiting the catchment area, since the Mam Tor landslide forms only

a small part of a much larger drainage basin. What is more Cleaves et al.

state that sources of water, other than precipitation falling within the

catchment, must be small and preferably known (determined during base-

flow conditions). At Mam Tor, however, it would seem that there is a

considerable supply of groundwater from outside the immediate area. This

possibility is considered in detail in Section 5.4.3.

1.3.2 Engineering Properties

In order to fulfil the second objective of this research - to in-

,
vestigate links between chemical weathering and slope stability - it is

necessary to examine the solid material which is being weathered. Ideally

this examination should include both mineralogic investigations of 'fresh'

and 'weathered' bedrock and analysis of the relative strength of these

materials.

The mineralogy of the rocks (principally the Eda1e Shales) was

determined using X-ray diffraction techniques. Other materials, such

as carbonates and the ochre and gypsum precipitates, were also analysed

in this way to verify their identification.

 16

The engineering properties of rocks are most commonly assessed by

performing triaxial or shear-box tests. Triaxial testing, however,

requires that the material be largely homogeneous (such as in the case

of a granular intrusive igneous rock) so that there is no inherent 'weak'

direction. Finely laminated mudstones which are frequently jointed, inter-

bedded with diagenetic carbonates, and even highly fractured due to land-

slide activity such as are present at Mam Tor, are clearly not amenable

to this type of investigation.

Shear-box testing is more usually applied to weaker rocks and soils.

This type of strength test describes the tendancy of materials to shear

under stresses operating ·in opposite directions on different parts of

the mass. Shear-testing would appear to be particularly useful in a

landslide context where the rocks are moving on discrete rupture surfaces.

The material making up the Mam Tor landslide, however, is of such a highly

fragmented and inhomogeneous nature-that the testing of any single small

sample will not give results representative of themass as a whole.

In such cases Nonveiller (1967) recommends that: "Tests should

be carried out on large samples of rock, the size of which must be so

large that the influence of defects on the test result may be regarded

as giving a statistical average, characteristic of the domain which is

investigated. This leads to the necessity to test very large samples

in cases where exceptionally large volumes of rock are involved, as in

the study of stability of rock slopes". Facilities for testing the

strength of very large rock samples (over 1 m3) were not available in

this study. After consulting engineering geologists familiar with the

problems of strength testing (J. Cripps, 1979; C. Gribble, 1980 - personal

communications) it became apparent that no suitable alternative for materials

of this type exists. Consequently engineering tests were not carried out

on the Mam Tor rocks.

 17

A further consideration which was taken into account is the lack

of any suitable supply of unweathered Eda1e Shale for comparison.

Material from within the landslide, even at considerable depth, is

badly weathered. The only place where fresh rock can easily be obtained

is from a cement-works quarry at Hope, 4 km east of Mam Tor, where the

beds being exploited are somewhat lower in the succession.

 18

2. THE STUDY AREA

2.1 Location

Mam Tor is known locally as the 'Shivering Mountain', presumably because

of its repeated movement. The hill is located close to the Derbyshire village

of Castleton, itself some 24 km west of Sheffield (see inset, Figure 5) and

the summit at 1695 feet is encircled by a double ditch and mound which marks

the position of an ancient hill fort.

On the east side of Mam Tor a landslide extends on to the floor of the

Hope Valley at its westernmost end. To the north lies the grits tone ridge

of Back Tor-Lose Hill separating it from the parallel-running Vale of Edale.

To the south steep slopes of reef limestone mark the northernmost limit of

the Derbyshire limestone massif. The stream draining eastwards out of the

Hope Valley joins with the River Noe flowing out of Edale to become a

tributary of the Derwent.

As note d previously, the A625 Sheffield to Manchester (via Chapel-en-

Ie-Frith) road runs along the Hope Valley from Castleton and twice crosses

the landslide on its ascent of Mam Tor before continuing westwards along

Rushup Edge.

The area described above is located towards the centre of Sheet 110 of

the Ordnance Survey's 1:50,000 Series (First Edition).

2.2 Geology

2.2.1 Edale Shales

Sheet SKl8 of the Geological Survey's 1:25,000 series shows the Mam

Tor landslide to be one of many in this area, which are usually located

where Kinderscout Grit or Shale Grit overlie steep shale slopes. At Mam

Tor it is the Mam Tor Beds and underlying Edale Shales that are affected

by slipping (see Figure I,).

FIGURE 1: Geology and Landslides of the Edale-
Mam Tor Area. (Source: GeolQgical
Survey Sheet SK18, 1:25,000 Series)
 Fig.1 Geology
and Landslides
of the
Edale - Mam Tor Area

KIOderscout Grit Grlndslow Shales ,

~ ~ ~
Inndshde:
[§ Shale Grit
~
Edale Shales arrow shows
direction of
movement

El Mom Tor SondstonestBl.L] Bee Low limestones

Section 1

'lQ1. of Edala

- '" - Mam~ bond In Edole Shales Rap = Apron reef

•
 19

As the geological section in Figure 1 shows, the Edale Shales rest

unconformably on the Dinantian limestones and vary in thickness considerably

as a result. In the Vale of Edale to the north, these beds are in excess

of 200 m thick. In the Hope Valley at Castleton successively higher strata

overlap the lowest beds (Cravenoaeras Zeion) as the edge of the limestone

shelf is approached. AT Mam Tor the Edale Shales consist mainly of mud-

stones, but occasionally siltstones and diagenetic carbonates occur. The

latter may take the form of continuous beds or discrete concretions of

dolomitic composition (see X-ray analysis in Appendix C). The concretions

vary in size considerably from a few centimetres to two metres across (Plate

2a) and many show calcite-lined septaria.

These carbonates, on a fresh surface, are very dark - almost black -

and extremely hard. A weathered surface skin, however, exists on both

bands and concretions which is very variable in character. In places this

rind is a light grey-brown colour and can easily be removed with a finger-

nail, having a soft and chalky texture. Elsewhere a very hard crust has

formed which is characteristically purple or red-brown, suggesting the

presence of hematite or other iron oxides. Gypsum crystals frequently

encrust the surface of these carbonates, which are easily distinguished

at a distance from the surrounding shales by an overall red-and yellow-

brown colouration.

The shales are equally variable in appearance from place to place and

depending upon the degree of weathering. Fresh rock, from Hope Cement Works

or (rarely) from borehole cores, appears dark grey and largely homogeneous.

This rock is extremely fine-grained and splits along relatively few

bedding planes, which seem to contain a higher proportion of mica or car-

bonaceous material. In thin section this observation is confirmed (Plate

2b) and small «0.25 mm) subhedral grains of an opaque mineral, possibly

pyrite, can also be seen.

PLATE 2a: A carbonate concretion from the
retrogressive unit above Site 7.

PLATE 2b: Photomicrograph of 'fresh' Edale

Shale. Sample BH88, x 20.
PLATE 3a: Photomicrograph of 'weathered' Edale Shale.
Sample G2, x 20. The white areas are gypsum.

PLATE 3b: Detail from thin section of sample G2, x 25.

PLATE 4a: Euhedral gypsum crystals in thin section of
sample G2, x 25.

PLATE 4b: Hand specimen of Edale Shale showing

euhedral gypsum crystals.
 20

Weathered Edale Shale has a completely different appearance. In

the field bedding and laminations can be seen at 1 or 2 mm intervals

throughout the exposure, and at least two joint sets are noticed. These

laminations and joints appear to be widened in places and filled with

euhedral crystals of gypsum. The presence of gypsum is confirmed in

resin-impregnated thin-sections of weathered shale such as those shown

in Plates 3 and 4a. These crystals vary in size from less than 1 mm to

several cm and occur both as small rosettes and single large tabular forms

(Plates 4b and 5). They are frequently discoloured on the surface by a

powdery red-brown or sometimes yellow overgrowth. A powdery yellow mineral

which resembles sulphur is closely associated with the gypsum on joints

and bedding surfaces.

Cracks running at about 600 to bedding are also present (see Plates

21 and 22b) and the result is that most exposures aresurrounded by material

which has broken along these cracks, bedding planes and joints into small,

angular fragments less than 5 cm across. Indeed the shales in this highly

weathered state offer little resistance even to slight finger pressure

and disintegrate easily. The resulting blocks accumulate as scree below

the exposures where continued breakdown further reduces their size.

Typical Edale Shale scree material is shown in Plate 6a and can be seen

to comprise platy particles less than 1 cm across and 1 mm or less thick.

The fragments of carbonate, once they have acquired their protective

weathered rind, remain relatively large. When the scree material is

disturbed a sulphurous smell is noticed.

The Edale Shales are deep basinal mudstones and as such contain an

abundant marine fauna comprising both thick and thin-shelled goniatites,

Peatinoid bivalves, brachiopods and trilobites. Diagenetic pyrite occurs

throughout the succession as small, disseminated grains (Amin, 1979).

PLATE 5: Single large gypsum crystal (and numerous
smaller ones) on a bedding plane in
weathered shale.
 21

Pyrite is known to be a common constituent of many sedimentary

rocks, particularly mudstones, where it is almost always associated

with organic matter~-

Thermodynamic calculations show that pyrite is only stable in the

absence of oxygen and in the presence of dissolved sulphide (Garrels and

Christ, 1964). Such conditions are found in modern sediments where

organic matter first accumulates and subsequently decomposes as a result

of bacterial action. Kaplan et al. (1963) however, have shown that not

enough organically derived sulphur exists in sediments to produce the

observed quantitites of pyrite. Consequently Berner (1970) believes that

high pyrite contents are a result of "Continuous diffusion of dissolved

sulfate into sediments where it is reduced to H2S and reacted with iron

minerals". The first-formed phase in this process is seldom pyrite and

laboratory experiments show that non-crystalline FeS, mackinawite (Fel.SS)

and greigite (Fe S4) are an intermediate step (Berner, 1964; Rickard, 1969).
3
Berner (1970) notes that most of the pyrite is formed in the top 10 cm

of the accumulating sediment in the following sequence of events:

1. Organic matter and iron are deposited.

2. Dissolved oxygen is removed from the sediment by aerobic meta-

bolism and anaerobic bacterial sulphate reduction occurs. H2S

is also formed.

3. H2S reacts with the iron minerals to form the monosulphides.

4. Sulphate diffusion from the overlying water occurs. Sulphate

reduction and Fe to FeS conversion continues.

S. Some H2S is oxidised to elemental sulphur.

6. This sulphur combines with the monosulphide to form pyrite

(FeS+S = FeS ) which crystallises

2
as minute 'framboidal micro-

spheres'.
 22

Berner concludes that "The time for complete transformation of

FeS to FeS2, in the presence of abundant H2S and elemental sulfur, is in

the order of several years".

For more detailed accounts of the geology of the Edale Shales the

reader is referred to Hudson and Cotton (1945) and Stevenson and Gaunt

(1971). The ajoining limestone massif is also adequately described else-

where (see Eden et al., 1964; Stevenson, 1967).

2.2.2 Mam Tor Beds

The Eda1e Shales pass upwards, with no conspicuous change in lithology,

into the Mam Tor Beds which are at their thickest - 140 m - in the vicinity

of Mam Tor itself. These beds were first described in detail by Jackson

(1927) who referred to them as the "Mam Tor Sandstones". A11en (1960)

and Mason (1961) have also documented the succession in considerable

detail. It is apparent from these works that the beds are typical

distal turbidites and true sandstones are often not well developed. As

a consequence Stevenson and Gaunt (1971) suggest that the name Mam Tor

Beds is in fact more appropriate.

On passing out of the Eda1e Shales and up into the turbidites the

only detectable change is that one or two more silty horizons are en-

countered. Here the shales yield ReticuZoaeras reticuZatum which places

the base of the Mam Tor Beds in the RIc zone. Above these lower, dominantly

sha1y horizons, the cyclicity typical of turbidites is apparent and alter-

nating shales, siltstones and fine-grained sandstones become gradually

coarser towards the top of the exposure (Plate 6b).

White mica flakes are conspicuous both in the shales and sandstones,

and the former fracture along bedding planes containing a greater proportion

of coarse flakes. The sandstones often show graded bedding, in addition

to penecontemporaneous deformation structures such as slumpball structures,

PLATE 6a: Typical Edale Shale scree material.

PLATE 6b: General view of the turbidite succession,

Mam Tor Beds, main scarp face. Note the
increasing proportion of shale towards the
base.
PLATE 7a: Sole marks on the underside of a
coarse sandstone unit.

PLATE 7b: The scarp face of Mam Tor showing the

geological boundary. Mam Tor Beds
above (a), Edale Shales below (b).
PLATE 7a: Sole marks on the underside of a
coarse sandstone unit.

PLATE 7b: The scarp face of Mam Tor showing the

geological boundary. Mam Tor Beds
above (a), Edale Shales below (b).
 23

corrugated bedding, convolute and crumpled bedding (Mason op.cit.).

The undersides of many sandstones exhibit excellent groove casts, flute

casts, load casts, trail casts and burrows (see Plate 7a). These indi-

cate a general derivation from the north-east.

Each cyclothem of the Mam Tor Beds from sandstone through siltstone

to shale can vary in thickness from 7 cm to 3 m, yet even the thinnest

are remarkably laterally persistent. None are seen to wedge out across

the entire width of the scarp face; a distance of some 300 m. These

turbidites are interpreted by Allen (op.cit.) as having been laid down

on the lower slopes of a Namurian delta, and as such represent the first

coarse clastic material to be deposited in the southern part of the Central

Pennine Basin (Walker, 1966). It is interesting to note that further north

the stratigraphically-equivalent beds are older, marking the advance of the

delta into the southern part of the Basin (Reading, 1964).

In a geochemical investigation of Namurian rocks from the Pennine

Basin, Amin (1979) notes that pyrite is only a very minor constituent of

the Mam Tor Beds, occurring largely in the non-marine shale horizons in

association with plant remains and siderite. This is in marked contrast

with the Eda1e Shales which contain abundant pyrite.

Both the Edale Shales and Mam Tor Beds dip gently north-westwards at
o
about 5 in the vicinity of the landslide. This 'dip into slope' situation

commonly accompanies deep-seated, concave failures of the type discussed in

Section 3.1. The angle of dip steepens. considerably as the limestone to

the south is approached.

Shale Grit, the next highest formation in the succession, is not

present at Mam Tor, but can be seen in exposure both to the east and west

at Lose Hill and Rushup Edge, respectively, where this hard, resistant

rock forms conspicuous tors. More detailed descriptions of the succession

above the Edale Shales can be found in Allen (1960), Collinson and Walker

(1967) and Walker (1967).

 24

Plate 7b shows the face of Mam Tor as viewed from the east. The

dashed line indicates the position of the geological boundary between

the Mam Tor Sandstones (A) and Edale Shales (B). Sandstone units in the

turbidites, being more resistant to erosion than the interbedded shales,

form conspicuous ledges on the main scarp face of Mam Tor and these can

be easily identified from the photograph. The Edale Shales are best

exposed in the minor scarp face of 'Little Mam Tor' (Plate 8), but can

also be seen in the retrogressive units (Plate 9). The foot and toe areas

of the landslide contain both lithologies. This material is so highly dis-

rupted by centuries of movement, however, that relationships between the

two cannot be established and the whole area is a jumble of sandstone and

carbonate blocks set in highly fragmented black shales.

2.3 History (Mining)

Some consequences of economic mineral exploitation can be seen on

and around the Mam Tor landslide. A worked-out mineral vein, the Odin

Rake, parallels the south side of the landslide where it moves against

the reef limestones of Treak Cliff (Plate lOa). The most important

mineral that the Odin Rake contained was galena (PbS), although sphalerite

(ZnS) was a common associate. Today traces of these minerals can be found

in the fine tailings from Odin Mine although their weathered carbonate

forms, cerussite and smithsonite, are more common. Isotope dates place

the injection of the hydrothermal liquids from which these minerals crystal-

lised at about 180 million years ago during the Jurassic (Ford and Rieuwerts,

1975). The famous variety of fluorspar called Blue John (from the French

'b1eue-jaune') has also been mined at Treak Cliff for centuries and is

often found in close association with the sulphide minerals.

Lead mining has a long history in Derbyshire and galena has probably

been worked since Roman times. Odin Mine at Mam Tor is the earliest named
PLATE 8: Little Ham Tor showing sample locations.
PLATE 9: The area of retrogressive units showing
sample locations. Spring No.7 is
located just beyond the two figures.
PLATE lOa: Odin Mine and Cank Mouth Cave. The near-
vertical cleft on the right is the worked
out Odin vein. The drain at top centre
carries water from Site 11 to the Odin
Sitch, which flows from left to right
through the trees.

PLATE lOb: The crushing circle and Knowles Engine Shaft

(surrounded by a fence at top left). The ridge
in the foreground is the right flank of the
land slide.
PLATE lla: The artificial channel of the Odin Sitch.

PLATE lIb: The Odin Sitch 20 m further downstream,

below the mixing site. Note the white
precipitate coating the stream bed.
 25

mine in the area, being first mentioned in print around 1280 A.D. (Ford

and Rieuwerts, 1976). The effects of this mine on the landscape of Mam

Tor have been considerable, particularly on the south side of the toe.

This was for centuries the location of the mine's waste hillocks which

have continually been built up, levelled, re-worked for gangue minerals

and ultimately landscaped as a picnic site. Much of the waste was finally

removed for ballast during road maintenance operations in the 1940's and

1950's.

A solitary shaft reaching to the old workings (the now flooded

Knowles Engine Shaft) can still be seen close to the lead ore crushing

circle, erected in 1823, and shown in Plate lOb. Many of these relics

of the lead mining industry can be seen from the air (Plate 12).

A further consequence of mining operations was the diversion of the

stream known as Odin Sitch. This stream flowing off the limestone would

naturally have taken the easiest course down the gorge which marks the

former position·of the Odin Rake, here following the line of a fault.

To avoid flooding of the workings this stream was diverted during the

18th century to flow north and then east around the mine and on to the

landslide (Plate 11). This channel is still the most prominent feature

of the drainage at Mam Tor today. Some consequences of re-routing these

karst waters onto the landslide will be discussed in Section 7.4.3.2.

It is possible that concentrations of ions in streams draining the

south flank of the landslide may be partly indicative of waste hillock

mineralogy. It is not possible, though, that weathering of lead and

zinc sulphide minerals associated with the Odin Rake can contribute

sulphate ions to streams on the landslide. This is because any remaining

mineralisation is located well below the Edale Shales, deep in the under-

lying limestones. In addition, a complex system of underground drainage

channels (known as 'soughs') draws water down from the old workings,
FIGURE 2: Schematic section through the shafts and
soughs of Odin Mine. (After Ford and
Rieuwerts, 1976).
 o
to
.
t

Vl
:J:
Cl
::>
-
....
d
s:
Vl
u.J
e Z
Vl
z o ~
e Z Z
et C5
e
&.l.-
e
01
Cl
'-
o
Cl .-
u..
cu
cu
lit

oI
U
~
en
•
et
lit

- -
(II C
u
d .~
u

-
0. 01
'- lit
d
u
lit III o
c: ....
Cl
(3 ~
'0 .
E\ .... ~
c
.... .. .12'
.... .......]
d
'-
'-
t!f?, Vl o
e
C-;
! 41
&.I.-

C
:I: : ~
C
W

\
'..~
i
i
Qj
--GI
d-
ue
-
Qj

: Vl-
\
\
•
 26

several hundred feet below the surface, to finally emerge well to the

east of the landslide close to Castleton village. Figure 2 is a dia-

grammatic representation of the underground soughs in relation to the

landslide.
 27

3. GEOMORPHOLOGY OF THE LANDSLIDE

3.1 Landslide Classification

Engineers and geomorphologists alike have produced many landslide

classifications. Two of the most widely used, and most relevant here,

are those of Hutchinson (1968) and Varnes (1958). According to Hutchinson

"rotational landslips occur principally in slopes largely fo.rmed of, or

underlain by, a fairly thick and relatively homogeneous deposit of clay

or shale". This is certainly the case at Mam Tor, which according to

Hutchinson would comprise an upper zone of 'rotational slip' and a lower

zone of 'translational slide'. Similarly Varnes' model (Figure 4)

considers an upper area of 'rotational slumping' in bedrock giving way

to a lower area of 'earth flow'. Figure 3 is a general outline of Varnes'

classification scheme.

Although Varnes' model more closely describes the Ham Tor landslide

his terminology concerning the toe region is somewhat out-dated. The term

'translational slide' is preferred in this study.

By comparing Figures 4 and 5 it can be seen just how closely Varnes'

diagram fits the Mam Tor slide; many features are identical in both figures.

For example, there is a distinct main scarp, the main slipped unit shows

reversed slope and increased dip of the strata (up to 550 in~aces), minor

scarps are clearly developed (such as that of 'Little Mam Tor'), there

are numerous transverse ridges on the foot of the slide, and the toe is

very clearly defined. Several of these geomorphological features can be

seen on Plates 1, 8 and 12.

The retrogressive units on the south side of the main unit at Mam

Tor are not shown on the diagram of Varnes. It would seem that these

features result from secondary flank failures which occurred some time

after the main unit itself rotated away from the face, leaving the side
FIGURE 3: Landslide classification scheme
(After Varnes, 1958)
 tI)
q) q)
)
0
»
r-4
00 ~ r-4
~ CO Ul ~
Cd4-4
~ 0
~ ~
'0
».~
A.
.~
'0
0..
~
Q)
-c
)
0
» ~)
J.I ~ 0
Q)

Q) ~
tI) Q) CO CO 0 r-4 Q) Xr-4
~ Q) >~ ~ ~ Ul ~ ~Ul
CO~ ~
.....S ~CO Q):> < ~ IUI~I~I~
-
X 0 ()
04-4 ~ <U
J.I 0 tI) N ~ \0 00
0- N I I I I
~ 0 r-4 0
,...., 0 0 0
~ 11-1 r-4 r-4 r-4 r-4

-- ---- ---
()
~
""' - U

- -
Q) 0
~ ~ I I
~
1-- 1- '" ~ ~
< U~< ~< .~
r-4 ~
0 ~
Q)
til
:>
0
'" S
~
r-4 ~ 11-1
Q) U r-4 0
tI)

.~
r-4
-cOO
p. ..... I=l
S r-4 .~ .~
00
I=l
tI)
.. 11-1
..c:
~
tI)
q)
0 ;::It!)'O "t:I .!:<: ~ A-
Ul r-4 CO CO () CO ) »
Ul~<U Q) Q) 0 ~ 0
,...., ~
r-4 o 0-
COr-4Ul
U ~
.~
'0 ~
r-4t1l
A.
~
-c ~
---+
11-1
'0
Q)
Q)
,...., C::~ til q) ,...., ;::I ~
,...., ,....,
CO
11-1
.~
~
0
~ CO
r-4
CO ~ .....
tI)
~
CO
X
.~
~
Ul ~ ..c::
~
.~
r-4 CO I=l Q)
~CO~
~ q)
.0 ~
Q)
..c::

-- «~ -
0 0""" CO Q) CO ~
til ~p.,~ O~ ......
~ U

--
r-4 __ 'I 11-1

.~
~
CO "'d
~
Q)
~
,....,
.~ « 1-
< '-"" ~ '-"
0
<U
<U CO til ~
~
CO .~
'0
~ Q) ~
,...., ~
~ r-4 0 .g ~ ~

-- --- --
0
11-1 tI) '\:l c:: ~
0
0 c:: ~ CO
,....,
~I_ ~ 0 Cd )r-4 Q)
Q) I () til CO o·~ I=l
p. < UU < c:: ~ r-4t1l 0
» ~ '\:l
Q)
~r-4 ~ J'<.!
~ 11-1
E-4
~
~
0
aJ'<.!

-c
tI)
tI)
-+-
.~
~
.~
tI) .. tI)

~'\:l4-4
0 ~
,...., 0_

~
~
en ~ <U
COO
en~
.0
Cl
Q)
.o~'O
Q) CO ::I
AtIl~
.~~
o
~ ;::I
<U

-.g -- --
CO 00
Q)
'\:l
...... ~.,-4
',-4 J'<.!
~ < < ~
30
0...,-4
11
-
0 tI) Q)
J.I ~ Q) o Q)
"'d r-4 s: () tI)
Q) tIl~
~
r-4
co,....,
()
0
<U
!
CO
~
CO
,...., ~
CO
»:>
<U

- ~-
r-4 S:P::) '\:l ~ Cd r-4 .c 0
r-4 0 .,-4
,...., ~ 11-1
Cl! .,.j ~ ~ ..c: (fJ~
11-1 ~ CO (fJ ~ ,...., ~ .,-4
~ Cl! s: ~() u ,...., ~ '\:l
u ~ CO 0 Cl! ~ Q)
0 0,...., 0 ~ CO ) ~.c
~ ~
~p..
Q)
bO
J'<.!
"'d
.,.j
0
,....,
11-1
~'C
Q) ()
~ ~ -c :> tI)
~ o
Cl!
...:I
0
~ ~ :f ~"'d
q)

"'d

~o e
Q)
>-
r-4 ~
~ ~II-I
0
~
~
CO A
t' Q)
~
',-4 q) .,-4 ell
(fJ'O ~
~ r-4 .,-4 r-4 bO"'d
s: CO
~,....,» CO Q) (fJ
',-4 .,-4 s: 13 <U
~ J.I 0 ~ ~ o ~ '\:l
Q) Q).,-4 CO Q) ..... 0 .,-4
:> ~ ~ <U ~~14-1 ,....,
0 Cl! o ~ o q) (fJ
~ ~ S 00 ~ S'\:l -c
~
11-1
0
<
. ~
. CIl
...:I
ell X
0- <U
,....,
» (fJ (fJ
E-4
(fJ q) ~
r-4
,...., '\:l
.,-4
~
,...., 0
C,)
,....,
~
Cl!

. til

H
. H
J'<.!

. ~
.
H H
H H
H
FIGURE 4: The major features of a rotational slump
(After Varnes, 1958)
The Major Features of a Rotational Slump

[After O.J. Varnes.195a]

Main Scarp The exposed surface of the

main shear plane

Minor Scarp The exposed surface of a

subsidiary shear plane within the stump

Foot The intersection of thE' main shear

plane and original ground sur+ace

The lowest margin of moved material

Toe
FIGURE 5: Location and major geomorphological
features of the landslide at Mam Tor,
Derbyshire.
(A.B.C.D. refer to lines of section
on Figure 2).
 Location and Major MANCHESTER
Geomorphological Features .Glo~sop ~
of the Landslide at Sheffield

1t
(ha Ccstteton
~'lm Tor, Derbyshire en-le-Fr
~ N
Buxton
STUDYAREA
~
MAM TOR 61,695'

-- - Major fault

Major active shear

Boundary of landslide

Trunk Road

-... Spring and stream

• Farm Buildings

• Sampling sites

~ Free face

1:_:1 Seepage area

J'

o 300 m

---t-N

•
FIGURE 6: Longitudinal section through the land-
slide and retrogressive units. Plane
of section is west (A) to east (B).
Broken line indicates dip of strata.
w
o
..J
VI
o
:z
~
..J

:z
o
....
u
w
VI
..J
c.(
z
6
....
=>
l:)
z
o
..J

Ci
u
tCU e
>-
Cl
'U:J
Ccr

I:
ClCU
-11'1
Cl~
1:0
OU
,...11'1
C
.2
•
 Q)
..-f
0
M
. g 0 11"1
lIS \0 ...:t
"'0
~
..-f
..-f
QC)
"

'0
(I)
Q)
Q)
00
C; 0
I
...:t
. 0
I I

11"1
,..._ 0-
I

IlS ..-f
o~
..-f
~ ..-f
Cl)
'0
c;
lIS
,..;j

Q)
..-f
lIS Q)
'0 ..-f
~ lIS

.
"'0
-Ie
..-f
~ 0
N C; 0 0 ..-f ...:t
'0
lIS
Q)
..-f ...:t
M g M
C; ~
lIS «
~
0
E-I

~
~
4-f
0 ~
0
Cl)
E-I
.
-
11"1
C;
0 alIS ...:t 0 0 In
o~
,..._ M 11"1 11"1 0
Cl)
C; ,..._ ~ 0- 11"1 ..-f
0-
~ ..-f
0..1
Cl ..
Q)
> ~
~
-
0..1
.u IX:!
lIS
~ ~ 13
lIS Q) '-"
.u
~ e-
-
4-f
0 <
U '-"
a
'-"
a
'-"
lIS
CJ
til
.c
.u c:I
00
.c
.u
0..1
11!
c; ~
-
'0
Q) 0..1

lIS
,..;j ~ 4-f
0
.c
'" a .u
II:!
Q)
~
=
o~
>< o~

lIS
.~
i>
Q)
< ~ ~ ::r:
 28

wall unsupported. The smaller cross-section in Figure 6 is drawn through

these retrogressive units, which can be seen to have an identical structure

to the main landslide but are on a much smaller scale.

3.2 Description of the Mam Tor Landslide

In terms of dimensions, Mam Tor is one of the largest landslides in

the area covered by Sheet SK18, and is certainly the most elongate (due

to the translational slide component beyond the foot). The figures

given in Table 1 help to illustrate this point. The total volume of

material which has moved is probably around 25 million cubic metres

(Brown, 1977).

The most conspicuous feature of this landslide, which can easily

be seen on Plate 1 and from the air (Plate 12), is the main scarp face.

This rises to a height of 105 m above the main slipped unit. Horizontal

banding on the face, which can be seen in more detail on Plate 6b,

can also easily be discerned on the aerial photograph. This scarp

appears generally light-toned due to the lack of covering vegetation,

which is due to highly active erosion rather than a lack of suitable

niches.

As a result of this continuous removal of material from the face

thick 'scree' deposits have accumulated at its base. This is not true

'scree' as usually defined, since it contains a large proportion of very

fine material which binds the mass together. In this way it falls some-

where between scree and colluvium. Whatever terminology is preferred

this area is also easily distinguished from the air and frequently displays

gully and levee-like features on its surface, particularly after heavy

rain.

Beneath this 'scree' lies buried the main surface of rupture on

which the landslide is moving. After particularly rapid movements, such

PLATE 12: Aerial photograph of the Mam Tor land-
slide. Contact scale 1:7,000.
(Numbers refer to location of boreholes
7 and 8).
 29

as those in 1977, the uppermost part of this surface is exposed as the

main unit moves downslope, taking the 'scree' with it. After a time

material falling from the face begins to accumulate and the surface

becomes covered again. Plate 13a shows the uppermost portion of this

surface with 'scree' material in the foreground. Slickensides indicate

the direction of landslide movement and the polished appearance is due to

the alignment of clay ~nera1s under pressure, parallel with the rupture

surface (Plate 13b).

Another bare rock/scree surface can be seen on Plate 12 just below

the upper section of road. This is the minor scarp face known as 'Little

Mam Tor' and is the result of activity on a second major shear plane which

roughly parallels the upper section of road (see Figure 5). Here the in-

clined banding in the face (Plate 8) is due to the presence of carbonate

horizons in the Edale Shale.

The appearance of this minor scarp face is not unlike the retro-

gressive units, which can also be picked out as light tonal areas on Plate

12, above the top road on the main unit.

Lying roughly between the two road sections is the 'foot zone' of the

landslide. Movement upslope is greater than on the 'toe' and as a result

material comprising the foot becomes rucked-up and a complex arrangement

of ridges and hollows can be seen from the air.

Below the bottom section of road the toe of the landslide can be seen

to spread out onto the solifluction-mantled foots lopes of the Hope Valley.

The fact that no solifluction material is found on top of the landslide

material means that the landslide must have been activated after the last

period of glaciation had ended.

On the toe, movement of material en masse along failure surfaces

gradually dies out and is replaced by much slower translational slide.

Roughly, the lowest point on the failure surface marking the right flank
PLATE 13a: The main failure surface exposed at
the base of the scarp face.

PLATE 13b: A portion of the slickensided failure

surface.
 30

of the landslide can be seen forming the slight ridge in the foreground

of Plate lOb. As a result of slowing movement downslope this region is

also characterised by a network of ridges and hollows. In many of these

hollows exist seepage areas where waters upwell from the landslide material

and precipitate ochre.

The vegetation of these boggy seepages is very different from the

surrounding grass and bracken, and as a result they can be easily discerned

as dark-tone patches on the air-photo. The position of the seepages must

be constantly changing as the toe material creeps downslope since several

isolated patches of ochre can be found which attest tothe former positions

of iron-precipitating springs. Plate 14 shows the characteristic appearance

of these seepages.

The actual rate of movement on this lowest part of the landslide is

minute in comparison with the often rapid rates recorded higher up. Tilted

and displaced trees, fences and even telegraph poles indicate that movement

is nevertheless still taking place. This is particularly evident at the

very tip of the landslide and at one locality a farm building, Blacketley

Barn, has periodically to be excavated to prevent it from being totally

engulfed (see Plates 15a and b).

Easily distinguished, also, are the flanks of the landslide. In the

upper part this is due to a line of tears in the turf due to the shearing

movement, and in certain places sections of walls have been displaced con-

siderable distances relative to each other on either side of the shear.

These flank shears are most spectacularly developed where the road crosses

from stable ground on one side onto the landslide on the other.

On the upper road section the moving material has been displaced down-

wards relative to the stable ground on either side (Plate 16). On the lower

portion of road the landslide material has been pushed up (Plate 17). This

in itself is conclusive evidence that the slipped mass is moving on a concave,

 31

spoon-shaped surface. Further down on the landslide the flanks are

defined where the toe debris rests with convexity on the valley floor.

Severe road damage has also occurred along the transverse shear

plane described above as being responsible for the scarp of Little Mam

Tor. In places the road surface has been constructed directly on top

of this highly active shear and the carriageway is split in two during

periods of activity (Plate l8a) •. Vast quantities of limestone ballast

have been laid down along this shear in short-term attempts at re-grading

and, therefore, stabilising the road. These are seen as the lightest

patches on the aerial photograph. Further results of movement on this

shear plane can be seen in Plates 18b and 19.

One feature which is common to all highly active landslides, and Mam

Tor is certainly no exception, is the lack of an integrated surface drainage

network. The artificial drainage ditches on the main unit never contain

water, even after periods of intense rainfall. This is due to the fact

that they become sheared after movement of the landslide and thereafter

precipitation soaks rapidly into the highly permeable covering of scree

and subsequently into the network of failure surfaces, and cracks, below.

Some of this deeper drainage may also find its way into the underground

drainage soughs;- originally intended to drain the workings of the Odin

Mine, and never re-emerge in the toe seepages as much of the shallower

drainage does. Downstream from these seepages the water does begin to

flow in well defined channels which link with each other and with streams

draining the surrounding area beyond the landslide toe.

The only large drainage channel, which permanently carries water, on

the entire landslide is the man-modified stream known as Odin Sitch. This

water eventually finds its way into the area of lead-mine spoil near Knowle-

gates Farm. This is also the point of emergence of Knowlegates Sough draining

the workings of Odin Mine.

PLATE 14a: Seepage at Site 16.

PLATE 14b: Seepage at Site 3.

PLATE lSa: Encroachment at Blacketley Barn.

PLATE lSb: Tilting of fences due to movement at

the toe.
PLATE 16: Downward movement of the road on the
upper section.
PLATE 17: Upward movement of material in the
foot zone.
PLATE 18: Further examples of road damage.
PLATE 19: Severe road damage and tarmac stratigraphy.
 32

3.3 Landslide Management

During studies of the large scale and rapid movements of the Mam

Tor landslide, which were referred to in Section 1.1, it was noted that

the maximum rate of movement (in the order of 0.5 m per day) is usually

maintained at most for two or three days (Brown, 1977). Subsequently

the rate diminishes gradually and returns to the 'normal' minimum of

less than 1 mm per day. This abrupt transition from slow creep to rapid

failure and a return to steady creep corresponds closely with the obser-

vations of Terzaghi (1950) concerning 'threshold behaviour'.

The resulting disruptions of the road surface present highway

engineers with perplexing problems. Should a long-term but permanent

solution be found, or is it sufficient to simply 'patch up' the damage

each time it is incurred? The latter approach to management has been

chosen in the past at Mam Tor and maintenance operations have been

carried out on. the worst-affected sections of the road nearly every year

since 1802 (when the present road was first opened). These 'maintenance

operations' consist simply of laying down fresh ~urfacing material (and

hard core if necessary) over the cracks and ridges and as a consequence

an impressive stratigraphy of tarmac has built up over the years. At

least two metres of this material can be seen below the photographer's

feet in Plate 19.

Many more permanent solutions to landsliding have been described

including the installation of giant retaining structures, the regrading

of slopes, or even the artificial acceleration of movement to produce a

more stable configuration (see Forbes, 1947; Cooke and Doornkamp, 1974).

All of these are too costly and unsightly (in a National Park) to be

practicable at Mam Tor, and the latter scheme might even make the situation

worse by removing support for the main face which would itself then become
 33

unstable. Royster (1979) advocates the complete removal of material

in cases of slipping such as this, although the 1ands1id~he describes

are somewhat smaller than Mam Tor.

One method of management which was seriously considered by Derbyshire

County Council was deep drainage of the entire area. As Cooke and Doornkamp

(op.cit.) note, with regard specifically to rotational slips "The back

tilting at the head of the mobile mass forms hollows in which water can

collect. This tends to drain down the rupture surface and maintain

permanently high pore pressures. By this means it may be self-perpetuating

as _,-an
area of instability".

As a precursor to drainage several boreholes were put down (under

the direction of A.W. Skempton and R.J. Chandler of Imperial College,

London) through the landslide in 1978, in order to determine the position

of the water table and the location of any failure surfaces. The sub-

sequent report from the team of consultants is still not available. One

can only assume that the results were inconclusive.

The proposed scheme was to consist of a network of four-foot diameter

drains bored horizontally into the slope. The consultants suggested that

this would reduce infiltration and run-on, drawing water away from the

slip planes. Such deep drainage has proved successful in stabilising

other landslides (Forbes, 1947; Early and Skempton, 1972), where the major

input of water is from precipitation falling within the bounds of the

moving mass. At Mam Tor, however, there seems to be an additional source

of water, irrespective even of the position of the local water table. The

possibility of groundwater upwelling from a fault is discussed in detail

in Section 5.4.3. The point made here is simply that such a drainage

scheme would not necessarily render the slide more stable. Indeed it

may even make the situation worse by improving throughflow and aeration

and increasing the rate of chemical weathering within the landslide material.
 34

Due to an immediate lack of funds and pressure from local inhabitants

the plans for drainage have been shelved. Rather than trying to keep

the existing road open (which was becoming increasing more costly and

difficult) the County Council have proposed a route for a new road which

will by-pass Mam Tor on the more stable limestone plateau to the south

(Peak Park Planning Board, 1977). The date of commencement for building

the road has not been definitely agreed. In the meantime the A625

remains open for access to farm premises, but use as a through-route is

impossible due to the extensiveness of the recent damage.

 35

4. WATER SAMPLING AND ANALYSIS

4.1 Sampling Design and Procedure

As noted in Section 1.1.2, certain springs on the Mam Tor landslide,

notably those of the seepage areas on the toe, precipitate ochre. The

presence of these accumulations of iron hydroxide suggests that the emergent

waters contain a high proportion of dissolved ferric iron. Also, in dry

weather, the stems of plants in the seepages become coated with gypsum

crystals in fine fibrous habit (Plate 20). This suggests that the stream

waters are also concentrated in calcium and sulphate ions. Capillary rise

of the solutions up the plant stems and subsequent evaporation leads to

crystallisation of the gypsum (CaS0 ).

4
In view of these observations it was decided to sample, on a regular

basis, any waters draining the landslide area with a veiw to determining

their overall chemical composition. A pilot study was conducted in the

Spring of 1978 when water from six streams was collected and removed to

the laboratory for analysis. Detectable quantities of several cations and

anions were found and there was considerable variation between samples.

A more thorough sampling network was established on the basis of this

pilot study and the main sampling programme commenced in October 1978,

lasting for twelve months.

Originally every distinct site of flowing water on the landslide was

sampled, in addition to one surface runoff site located on Rushup Edge to

the west, and two sites in the village of Castleton to the east. The

latter two sites were located where streams emerge from the limestone

cave system of the area to the south and west of the village. These acted

as 'controls', representing the 'normal' karst waters of the region. The

number of sites originally totalled twenty-eight.

PLATE 20: Gypsum precipitation on plants in the
seepages
 36

This total was reduced ,however , during the course of the year and

data are presented here for fifteen sites only. A number of the original

twenty-eight sites dried up after thefirst few weeks of analysis, or else

froze for long periods during the winter. Indeed the weather was so severe

during January and February, 1979, that Mam Tor was cut-off and impossible

to reach for periods of up to three weeks. During this time the rain gauge

was buried by snow for seven weeks in all, and data from the Hope Meteoro-

logical Office station is used in Appendix Bii. Other sites are obviously

linked hydrologically with ones nearby (although this is not apparent from

the surface drainage pattern), giving identical analytical results.

Sampling at such sites was also discontinued after a time, resulting in a

total reduction of thirteen sampling sites. The fifteen remaining sites

are shown on Figure 5.

In large channels, different velocities in different parts of the flow

can lead to cross-channel variations in chemical composition, and bulk

sampling is often necessary. Since the flow of water at most of these

fifteen sites,however, is in the form of shallow seepage or turbulent flow

in a narrow channel (ensuring adequate mixing) it was decided that single

depth-integrated samples from the centre of flow would be sufficiently

representative of the stream as a whole (Rainwater and Thatcher, 1960).

New or acid-cleaned 250 mI. polythene bottles with air-tight caps

were used to collect the water samples. These were filled completely, by

immersion in the stream, so that no air space was left in the bottle above

the solution. Each bottle and its cap was rinsed two or three times in the

water downstream from the actual sampling site before taking the sample.

Careful handling of the bottles at all times until capped was necessary to

prevent contamination, especially by salts from the fingers. Each sample

bottle was clearly marked with a site number and a letter corresponding to

the date of collection; this was necessary to avoid confusion in the laboratory
 37

when handling large numbers of otherwise identical bottles. Initially

samples were taken from each of the sites at seven-day intervals. In

order to standardise this procedure the samples were taken every Tuesday

morning between the hours of 8.00 a.m. and 10.30 a.m. (l~ hours was needed

to visit all. the sites). Such standardisation is often important when

taking samples of natural waters since chemical changes (most easily

detected by monitoring fluctuations in pH) take place throughout the day,

particularly during warm sunny days where plants and other organisms are

living in the water (Golterman et al.,1978).

As stated in Section 1.3.1 rainfall quality was not regularly determined.

Two samples were taken, however, during storm events on 21st August, 1979,

and 2nd October, 1979. Full chemical analyses were performed on these two

samples and the results are given in Table 7.

4.2 Sample Storage

During the pilot study it was noticed that samples from certain streams

produced an orange deposit on the insides of the sample bottles even if

left unopened for only a matter of hours. If left overnight the amount of

precipitate was considerable. Since the whole aim of collection is to

transfer the sample from stream to place of analysis with minimal chemical

alteration this precipitation is clearly undesirable.

It is reasonable to assume that iron in solution is responsible

since the orange precipitate forms only in samples from those streams

which deposit ochre in the field. Many chemical preservatives for natural

water samples have been suggested (see Zobell and Brown, 1944) but acidifi-

cation is the most common method of preventing the precipitation of iron

and other metals. Since the amounts of chloride and sulphate ions were

to be determined at a later date the use of HCl or H2S04 for acidification

was precluded. HN03 was, therefore, used instead, 0.5 mI. being added to
 TABLE 2: The Effects of Sample Storage and Acidification
on Water Quality

Solute Content
Sample Date of Days After
+
Number Analysis Sampling Ca2+ Mg
2+ Na K+ Fet

lB 10.10.78 0 111 30 50 3.1 8.2

1B 17.10.78 7 114 31 49 3.2 0.6
lB 24.10.78 14 111 30 61 3.2 0.3
2B 10.10.78 0 188 41 37 5.7 23.3
2B 17.10.78 7 190 42 34 6.0 22.1
2B 24.10.78 14 191 41 43 6.0 22.0

7B 10.10.78 0 408 201 13 5.1 89.5

7B 17.10.78 7 427 197 13 5.6 31.5
7B 24.10.78 14 422 206 17 5.6 26.1

8B 10.10.78 0 24 14 10 2.7 1.1

8B 17.10.78 7 24 15 5 2.8 0.7
8B 24.10.78 14 21 13 14 2.7 0.1

11B 10.10.78 0 223 122 11 4.1 43.3

11B 17.10.78 7 260 124 11 4.3 31.0
11B 24.10.78 14 252 114 15 4.3 23.3

13B 10.10.78 0 100 31 26 2.0 5.9

13B 17.10.78 7 113 47 30 2.0 3.1
BB 24.10.78 14 95 37 33 2.0 2.2

IBa 10.10.78 0 109 29 51 3.1 8.2

IBa 17.10.78 7 137 31 64 3.2 8.2
IBa 24.10.78 14 120 31 61 3.2 8.4

2Ba 10.10.78 0 186 40 37 5.5 23.0

2Ba 17.10.78 7 190 41 34 6.0 22.7
2Ba 24.10.78 14 190 40 40 5.8 22.5

10.10.78 0 410 206 13 5.2 90.0

7Ba
7Ba 17.10.78 7 416 197 8 5.5 93.0
24.10.78 14 419 200 12 5.4 92.1
7Ba
7Ba 02.10.79* 357 426 209 15 5.4 88.4

8Ba 10.10.78 0 20 12 9 2.6 1.4

8Ba 17.10.78 7 20 14 8 2.7 0.9
8Ba 24.10.78 14 19 14 11 2.8 0.2

11Ba 10.10.78 0 250 137 11 4.0 49.7

11Ba 17.10.78 7 256 121 7 4.3 49.1
11Ba 24.10.78 14 252 131 11 4.6 49.9
02.10.79* 357 255 126 12 4.0 48.0
llBa

l3Ba 10.10.78 0 99 30 26 1.9 6.0

13Ba 17.10.78 7 99 31 23 2.0 5.8
13Ba 24.10.78 14 102 39 30 2.0 5.7

a Samples acidified prior to analysis

* Samples analysed approximately 12 months after collection
 38

each sample on arrival back at the laboratory. This reduced the pH of all

samples to the range 1.7-2.0. Table 2 shows the effects of acidifying

samples. The most noticeable improvement in storage efficiency is observed

for total iron content, which is remarkably constant over twelve months,

when the sample is acidified but decreases rapidly in the control.

Coarse particulate and organic matter was removed from the samples

prior to acidifying by passing each solution through a 'Whatman' 541

(hardened, ashless) filter-paper. This procedure undoubtedly causes some

extremely fine particulate iron and aluminium hydroxides to be recorded

as solutes, but the amounts are small compared with the truly ionic species

present.

Samples were kept for a maximum of fourteen days prior to analysis

although storage for several months was found to have little effect on

composition (Table 2). During the two weeks of storage the samples were

kept at room temperature out of direct sunlight (see Slack and Fisher, 1975).

4.3 Sample Analysis

Analysis of the water samples was conducted in two phases. Firstly

the temperature, pH, and electrical conductivity of the waters was measured

in the field at the time of sample collection. After the samples had been

removed to the laboratory, filtered and acidified as described in Section

4.2, they were analysed for ten major ions and a number of trace metals.

Details of standard conditions and procedures for these chemical analyses

are given in Appendix A, but a brief outline follows here.

4.3.1 Temperature, pH,and Electrical Conductivity

Stream water temperature at the point of sampling was measured· by

immersing a mercury thermometer in the flow. The thermometer was left to

equilibrate while the 250 mI. sample was being taken as describ~d in Section

4.1. Temperature was recorded to the nearest D.loC and entered into a field
 39

notebook along with readings of pH and conductivity and any relevant

notes concerning flow conditions (i.e. 'very low', 'heavy', etc.).

pH (the negative log of the hydrogen ion concentration) was measured

in the field using a portable battery-operated pH meter -Walden Precision

Instruments model WPA C6. Since the pH of most samples was in the range

3.0-6.0 a buffer solution of pH 4.0 was used to check the instrument. The

accuracy of the measurements was in the order of ± 0.1 units, and all values

were correct for the temperature of the water. pH is readily converted to

H+ activity, [H+] , by taking the antilog and changing the sign.

Electrical conductance is a measure of the ability of a conductor, in

this case a dilute aqueous solution, to convey electricity. The conductance

of water is measured between two parallel platinum or carbon plates and is

directly proportional to plate area and inversely proportional to plate

2
separation. Results are expressed as if for plates of 1 cm separated by

1 cm. When standardised in this way the measure becomes independent of the

measuring system and is a property particular to the water sample. The units

o f th'1S measure 0f con d uctlVlty

" are~,n-l cm -lIS
,or more common y, cm-l ('5'

for Siemens). The older name of 'specific conductance' with units of ~mho
-1
cm is now obsolete (Colterman and Clymo, op.cit.). The instrument used

to measure conductivity in this study, by immersion of the electrode in the

stream, was manufactured by Electronic Switchgear (model MC 1, mark V).

This particular model incorporates a temperature compensation dial (conduc-

o
tivity readings are quoted at 25 C). By taking measurements of conductivity

both in the field and on return to the laboratory it was discovered that

differences of less than 5% occurred between the two, and very often there

was no difference at all. Therefore, it was decided to take measurements

immediately upon return, thereby eliminating the need to carry heavy field

equipment from site to site.

 40

4 .3.2 .
Cations: A13+ , Fe 3+ , Fe 2+ , C a 2+ , Mg 2+ , Na + , K + ,

The quantities of six cations (the only ones present in significant

amounts in the pilot study samples) were determined each week by a variety
2+ M 2+ + +
of methods. Ca , g , Na and K were measured by atomic absorption
2+
spectrophotometry, total Fe(Fe + Fe 3+) by a colorImetrIC
.. met h od, and H +
3+
indirectly from pH as described in Section 4.3.1. At the outset Al was

not thought to be present but later was measured from time to time by atomic

absorption spectrophotometry. Samples which were high in total Fe (those

2
precipitating ochre) were occasionally analysed for Fe + separately, by

titration. This had to be done prior to acidification since this process

oxidises iron to the ferric (Fe3+) form. In this way the proportions of

ferric to ferrous ions could be determined.

Principles and applications of atomic absorption techniques will not

be discussed here. The text by Christian and Feldman (1970) is recommended

to the interested reader; Chapter 4 on 'interferences' is considered par-

ticularly useful.

4.3.3 Anions: RC0 , Cl ,

3
so~-
The only anions detected during the pilot study were RC0 , Cl and
3
2-
S04. Bicarbonate ions were, however, only found in the waters from the

Peak Cavern resurgence in Castleton. Since these waters are of no direct

consequence to the work at Mam Tor the bicarbonate titration was discon-
- 2-
tinued and Cl and S04 only were measured on all samples, by silver nitrate

titration and barium chloride turbidimetry, respectively.

4.3.4 Trace Metals: Cd, Cr, Cu, Mn, Ni, Pb, V, Zn

Black shales commonly contain heavy metals in trace amounts (Vine

and Tourtelot, 1970; Holland, 1979). Since the waters at Mam Tor are

• draining an area of fragmented black mudstones it seemed likely that the

waters ndght also contain certain trace metals and that these could eventually
 41

co-precipitate in the seepages with the ochre since iron hydroxide is a

well-known 'scavenger' of other metals (see Section 7.6.3). As a result,

eight samples were selected to cover the range of site types from ochre

seepages to surface runoff and these were analysed for Cd, Cr, Cu, Mn,

Ni, Pb, V, and Zn by atomic absorption using a high temperature carbon

arc furnace attachment.

4.4 Results

A complete table of results obtained from the chemical analyses is

given in Appendix B. There follows here a brief description of the results

overall, which leads on to a more detailed discussion of their significance

in Chapter 5.

4.4.1 Temperature, pH and Electrical Conductivity

At the commencement of the main study in October, 1978, the temperatures

of all the streams, springs and seepages were in the range 8-90C. Tempera-

tures rose slightly during this month, which was remarkably mild, but then

fell consistently until the end of the year. By January the temperature
o
of most streams had reached or closely approached 0 C. There were two

notable exceptions.

Waters draining the limestone region and emerging at Peak Cavern stayed
o
relatively warm and the temperature never fell below 5.6 C. At sampling site

No.7, below the retrogressive units on the south side of the landslide, the

total fall in temperature throughout the period October to February was only

O.90C. The lowest temperature recorded at this site was 8.70C.

Throughout the Spring of 1979 stream temperatures began to rise and

by May had reached their October level. The highest temperatures of the

year were recorded during a very hot spell of weather in June when some

sites became dry and one (Site No.4) was not to flow again during the

I SN~~i~~:;
period of study. The highest temperature of all, recorded at Site No.3a,

l._. :\)"\:,",~l
0!:'''' 0 ....·11
-.-- t
 ~ ....
IlS
M 0 · N

~
N
· 00· \0

\0·
-
....
1--- -- -- -- -- -- -- ....-- -- -- - --
~ ....
~· ·
\0
....0 · 0\ 00....·
-- t-- --
~
-- -- --- -- -- --- - -
0
....
0 ~ · ....·
ClO

N
\0

·
ClO
N

N
•

....
ClO
....·
-0
"· 0\
•
0\

·
....
M
" M

....
· ....· ·
0 ClO 0
....
" 0 ~ • ~
"
· · "·
N 00
·
N
....
-0
M ~
..... 00 M

..... ~
· .....·
N
M
..... ..... \0
"· ~ ·
"
.... ~
....· ....· ·
0\ ClO

·
0\
.... \0 00 M

~ ....
ClO ·0 · "·
N
N
0\
•
\0

- --- --- - --- -- - -

1---
~ ~
"·
.....·
N

" · ·ClO 0 0\ 0
-- -- -- --- - --- --- -- --
t---

~
· ....· ....·
0\ N N
\0 •N N 0 M

....
~ 0 · "·
\0
"· 0\

~ ·
..... \0

M ~
M ~ · 0

.....·
~ ClO · ClO

N ·
~
N ·
.....
....·
N

" "·
ClO
"·
~

....
0\

0 · ....·
0\
M
"·
"
M

~ ·
~ ~
...
::s
...::s ~
...
o&.J o&.J
::s e:: e::
§ ...~
IlS
§ ...IlS
o&.J
ca
... ........
"C

ca
0 111
~
..... e ~ Qj "C
o&.J;:&::
111

~-
e::
~-
Qj

.....o&.J .....e:: (rS 'Q (ru- ca fit'G c:: .....

>CIJ
e0
til ;:&::
~O IlS CIJ 0
;I:!~- ~
CIJ 0
111
o&.J
til A ....
...
\
 42

From February until June, as the temperature at most sites was rising

steadily, that of the water at Site 7 rose by only 1.3 degrees to 10.00C.

The Peak Cavern water remained cooler and reached only 9.50e during August.

In this same month the temperature at Site 7 attained its maximum value

Temperatures fluctuated considerably during the unsettled summer

but all lay within the range 8-l00e once again by October, 1979.

Table 3 is a summary of the data given in Appendix Bi, and clearly

shows the constancy of Sites 7 and 101 (Peak Cavern).

The vast majority of pH readings fell in the range 3.0-3.7, a fact

which can clearly be seen on Figure 11. Surface runoff sites and waters

draining the limestone area have slightly higher pH values, approaching

neutrality. Variations in pH seem in no way related to time of year. On

two occasions during the winter months air temperatures were so low that

the pH 4 buffer solution froze and the field meter ceased to operate.

During May, 1979, the conductivity meter which had been used through-

out the study was so severly damaged that repair was impossible. A second

Ldent ica l model was acquired and used for the remainder of the sampling

period. It is not known if the two meters would have given identical

readings for the same sample. There was no systematic variation in con-

ductivity according to the season, but levels were generally higher at the

end of the twelve months of sampling than they were at the start. It is

hoped that this observation is not the result of having changed meters.

A considerable range in conductivities can be noted in Appendix Biii

between the two extremes of Site 7 and Site 100. The mean values at these
-1
two sites are 2265 and 92 ~S cm respectively.

Differences in conductivity between samples reflect differences in the

concentration of ions in solution. Thus, conductivity is a rapidly obtained

measure of the total ionia aonaentration or aativity of a water sample.

FIGURE 7: Total dissolved solids-conductivity
relationship.
•

10000

,.~-:..... ~
....>- "<"':;1
~

7
....
~
o
z
-.
o
u
..

10 1000 10000
10 100
TOTAL DISSOLVEr SOLIOS [ug ml-')
 43

Lind (1970) shows that there is a log-log relationship between conductivity

and ionic strength, both for natural water samples and laboratory-prepared

salt solutions. A similar relationship seems to exist for the Mam Tor

samples, shown in Figure 7, between conductivity and total dissolved

solids (TDS). One group of points, however, lies off the 'best fit' line.

This group (circled) consists of samples which are known to contain the ion

BC~;, although this was not included in the chemical analyses (see Section

4.3.3). If bicarbonate had been measured, although the conductivity would

remain the same, the TDS content would be higher and the points would lie

closer to the line.

4.4.2 Cations

Calcium in solution is often assumed to indicate alkaline conditions.

The site with the highest pH in this study is Site No.lOl issuing from

Peak Cavern. Yet the data in Appendix Biv clearly indicate that this site
2
has one of the lowest Ca + concentrations. Much higher concentrations of

this ion are to be found in the acidic (pH 3-3.5) streams and, in particular,

Site 7 where values in excess of 400 Vg ml-l are frequently recorded.

Similarly Mg2+ concentrations are highest in the Site 7 water and are

often double the values for the next most concentrated site (No.ll). The
-1 -1
extreme values are 239 Vg ml (9.1.79,Site 7)and 1.0 Vg ml (5.6.79,

Site 101).

Concentrations of Na+ and K+ ions are given in Appendix Bvi and Bvii.

Sodium levels show a dramatic increase during the winter months in sites

located close to the road. These fluctuations correlate closely with periods

of road gritting conducted by the County Council. Potassium levels are

generally low at all sites, but are possibly slightly higher-than-average

in the ochre-precipitating seepages.

Total Fe concentrations (Appendix Bviii) vary considerably from site

to site. The lowest levels are recorded in surface runoff and karst waters
 TABLE 4: The Changing Proportions of Ferrous:Ferric
Iron on Storage

-,
A B e D

Total Fe Fe2+ Fe3+ F e2+ :Fe 3+

-
Sample (ug ml-1 )
(ug m1-1 ) (A-B) (ell)
-,
3P 88 13 75 5.8
7P 114 85 29 0.3
UP 121 16 105 6.6
16P 97 17 80 4.7

After twenty-four hours

3P 88 10 78 7.8
7P 114 61 53 0.9
UP 121 7 114 16.3
16P 97 2 95 47.5

After forty-eight hours

3P 88 7 81 11.6
7P 114 56 58 1.0
UP 121 0 121 - "

16P 97 0.2 96.8 484

-,.
 44

-1
« 1.0 pg ml ) and are once again highest at Site 7 and in the ochre
-1
springs (often> 100 pg ml ). Ferrous iron is only present in a handful
3 2
of sites (all ochre-precipitating) and the proportion of Fe + to Fe +

varies considerably between sites (see Table 4). The highest concentration

o f 85 ~g m 1-1 Fe 2+ was recor d ed at S~te

. 7•

Ferrous iron was not determined every week and the overall variation

in concentration is not known.

· ·1ar 1y Al 3+ was not analysed for on a regular b as~s,

S~mL . but Appen d·
LX

Bix lists the analyses for five consecutive weeks samples. Again aluminium

is most abundant at Site 7.

4.4.3 Anions

Levels of chloride ions as given in Appendix Bx are generally quite

high and it is thought that this may be due to an intrinsic fault in the

method of analysis (see Appendix A). However, variations in concentration

can be seen between sites and throughout the year. Ochre springs are

generally highest in chloride, with the exception of Site 7; this apparent

deviation from the 'norm' may be due to interference,from other ion~with

the analysis.

An identical response to that shown by Na+ is noticed for streams close

to the road (Figures 15 and 18) and a large contribution of sodium and

chloride ions from road salt in winter seems inevitable.

Sulphate is by far the most concentrated ion in solution at Mam Tor.

'Background' levels of 20-30 ~g m1-l are recorded from surface and karst
-1
waters but values of around 3000 pg ml are found in Site 7 water (Appendix

Bxi) •. This fact coupled with the low pH here justifies the use of the term

'acid sulphate' to describe these waters.

As can be seen from Table 5, the measured concentrations of the various

ions in solution, together with sample temperature and conductivity, are not
 U)
'.-4
U)
M
M
0
"-
ro4
"-
"-
N
.... M \0 ~ ex) 04'
0
0
CO \0
....
M ex) 04'

· · er· er· er· · ·

M \0 \0 \0 ex) 0
....I·
~ M

· er·
N ro4

0
~ ':l"' M ro4
I
M 0

U)
Cl)
QI '"
~ "-
CO ....
0\ N
0
\0
N
M M
'" .... ,...
~ 0
;
QI
CO
· · · ~· · · · · · ·
0\ \0 0\
0\
N
N
CO
"-
~ 0 \0
0 ro4 0 0 0 0 ro4 0 0 0
~
~

~ c:: .... .... 0 ~

0
'"
M \0
'" 0\ \0 0 ....0 ex)

· "-""'· · ....· 0· ·
t1l '004 N N ro4 CO M
-o~ ex) \0
N
....
N

·
0\

··
\0 CO N 0
c::t1l • t1l
.-4 • ~
0\ N N CO N
~ > ....0 ~ N ~ M M
ro4 \0
0
~ QI \0
~ \0

~
'" '"' 0\ ro4 N \0 ex)
~
0\ ro4
""' ........ ~
t1l 0
~ '"
"- CO M
....• ....0 M N \0

·· · · · ....· · · '"....'"·
\0 M 0 0\ "- 0 N
c:: '" N 0
~~
til
'" ro4 N ro4
M
0 0
M

\0 .... ....
c:: N N '"
\0
0\
0 ~
~
0
'"
.... '"
0\ '" CO N
0
0\

· ~~· ~· · · · · ~· ·
CO N 0\ "-
t1l 0 M N M

~
QI ex) M M 0 N ex) N
•
~
.... M M "- N
\0
N
0\

a:l
.... ........· .... "-
0
· ·8
0\ N
a ....
N
0\
M
0 0 M 0 "- ~
'.-4
><
t1l '" N ro4 ....
N N
M
M
N
\0
~ N

a
·
\0
~
'004
c::
• .-4
.... .... N 0 0 ~
.... CO
ro4
"-
N · 0
\0
0
~

~ c::
0
Cl)

0
.
",'004
QI ~
t1l
,...
~ 0\
~e 0\ 0\ 0\ ex) CO 04' C'\ 0\
0\ 0\ 0\ 0\ 0\ 0\

Z ~
M M M M M M M
'" '"
M M
\0
M

.0
0

ro4
-
la

....
~
Q) -
U
0
'-'
til

'-'
u
;l.

l>-
~
.~
Q)
• .-4
c:: ~
a:l
:> '"
t1l
§
§
'004 '004
0
....'"' ~
Q) Q)
4.J
4.J
:l
I1S
'004
4.J
'004
Cl)
Q) § Cl)
.... '004 I1S
..c:: '"' U

......c:'"
Cl)
QI :l
....u c:: • .-4 I1S I1S 0 Q, -0
00 ~ 4.J 4.J ro4 ~ c::
I1S t1l 0 0 0 :l ::c Q) 0
u ~ til Po. E-4 U til Q, E-4 U
 45

normally distributed about the mean. Large standard deviations attest

to the considerable spread of values while 'skewness' and 'kurtosis' are

measures of the shape and peakedness of the frequency distribution relative

to a normal curve.

Skewness is calculated as the 'third moment about the mean' (i.e.

the average of the sum of the cubes of the deviations) divided by the

standard deviation cubed:

(Gregory, 1963) (4)

For a normal curve the skewness index is zero. Here the data are

clearly positively skewed with a 'tail' on the high side caused by the

exceptionally high values recorded in the Site 7 samples.

Kurtosis is defined as the 'fourth moment about the mean' divided by

the standard deviation raised to the fourth power:

E(x - x)4
n
Kurtosis =------~------~
r-----~~- 4 (Gregory, 1963) (5)

J'E(X ~ x)2

A perfectly normal distribution yields a kurtosis value of 3. Here

most variables show a lower degree of peakedness (a greater spread).

Table 6 shows the calculated Pearson's product-moment correlation co-

efficients for every possible combination of variables. Particularly high

(>0.8) positive correlations are obtained for the following ion-pairs:

Ca-Mg; Ca-Fe; Ca-SO •

4'
 Q)

""=' M
N
""'
qj
N

""
0
Q)
0
~
Q)
E-4

0 ,....
0 ~
,....
::r: ~
P- o 0

0 0
I

Q)

""' N 0 ~
qj
0 ~
'" "
.c ~ ~
P- o 0
~
0

0
::s
til
? 0

Q)
"0
o~
N .... ,.... ,....
M

""0 "
0
0
0\
M
0
0
M
M
~
0 0

0 0 0 0
.c I
U

~
0
H"" ~
00
~
00
~
~
N ..:t
~
0\
~ 0 ~
qj
+J 0
0
"
0 '"
0 0

0
0
00
0

0
E-4
? 0

El
::s
o~ II"l ~ ~
en
en
~
II"l
..:t
II"l
M
11"'1
0\
00 "
~
0
'" 0\
qj
+J 0
0

0
0

0
0
"
0
0
'"0
0 0

0
Po.
I ?
,.... ~
9
.~
..:t
00
M
~
0
0
~
11"'1
N
0
"
M
00
00
M
N
"0 0 0 0 0
N 0 N
0 0 0
0 0 0 0 0 0 0 0
til I I I I

g
.~ M M ....
en
Q)
~
~
0
0
00
..:t
0
'"'",....
0
0\
M
0
0
0\

'"
0\
0
11"'1
11"'1
11"'1
0
00
~ 0
00
~
00
0
CO 0 0 0 0 0 0 0 0
qj I I
~

9
.~ II"l N ~ ~ ~ II"l
,....
0 0
'" ~ 0
" " ..:t

'" '"
II"l
CJ ~ N ~ ~ N 0\ ~ ~
~ 0 0 0 0 0 0 0 0 0
qj 0 0 0 0 0 0 0
u ? 0

>.
~
Q)

""::s .~
""'~
.~
~
.~
~
""
0
QJ Q) +J .~
9 en ~ .~ .c
H "0 +J qj
en
Q)
.~
en qj
""'CJ='
"" 'tl
Q)

-e .u .u ~"" ~
~ qj qj 0 P-
co ~ ~
qj
~
0
til
0
Po.
0
E-! (3 ='
til
::r: Q) 0
P- E-! U
 46

The close relationship between calcium and magnesium, and to a lesser

degree iron, suggests that these ions are derived from the same source

(diagenetic dolomite and ankerite). This also applies to the Na-Cl corre-

lation, confirming the belief that these ions are derived from road salt.

The strong Ca-S0 and Mg-S0 correlations perhaps suggests that the
4 4
precipitates forming from solution in the seepages are not only gypsum

(CaS0 .2H 0) but also epsomite (MgS0 .7H20), although similar examples from
4 2 4
the literature could not be found. Not surprisingly the dependent variable,

conductivity, shows a high degree of correlation with four of the most

2+ 2+ t 2-
concentrated dissolved species - Ca ,Mg ,Fe and S04 •

4.4.4 Trace Metals

Trace metal analysis was conducted on an additional set of water samples

collected from eight sites on January 23rd, 1980. Appendix Bxii gives the

full table of results.

Of the eight metals for which analysis was performed, only manganese was

present in sufficiently large quantities to have been detected by normal 'flame'

techniques (see Appendix A). Lead, zinc and vanadium were easily picked up

but only very minor amounts of copper, nickel, chromium and cadmium were

found. As for the greater proportion of major constituents the trace metals

were in the greatest concentrations in the Site 7 waters.

High Pb and Zn concentrations were also observed at Mam Tor by Ford

(1976), in a number of geochemical traverses across the area. These anomalies

were noted, however, in soil and stream sediment samples above the position of

the underground extension of Odin Vein. They were detected despite the vein

being covered by about 500 feet of shale. Unfortunately Ford does not give

quantitative details of measurements made and a comparison of values with

those obtained from stream water samples cannot be made.

 TABLE 7: Median monthly bulk precipitation concentrations, 1959-64
(after Stevenson, 1968) compared with values for two samples
from Mam Tor collected on 21st August, 1979(a) and 2nd
October, 1979(b)

Cl-
Ca2+ Mg2+ + K+ S02- Number of
Station Na
4 Observations

Aberdeen 1.1 0.6 4.6 0.4 5.7 2.1 70

A1dergrove 0.8 0.6 5.2 0.5 6.0 2.4 47

Belmullet 1.8 - 16.0 0.7 24.5 2.1 70

Birr 1.3 - 3.3 0.2 4.4 0.9 71

Camborne 0.8 0.6 7.7 0.4 11.3 1.9 57

Clones 0.6 - 2.7 0.2 3.6 1.0 72

Dublin 2.5 - 4.0 0.4 6.3 2.6 70

Edinburgh 1.8 0.3 3.8 0.3 3.0 1.7 68

Eskdalemuir 0.3 0.2 1.8 0.2 2.8 1.2 59

Leeds 2.7 1.0 2.3 0.5 5.3 3.9 63

Lerwick 2.0 3.2 22.5 1.1 32.3 3.3 61

Malin 2.1 - 21.0 0.9 33.0 2.6 71

Newton Abbot 1.5 0.6 5.5 0.7 6.3 2.9 59

Rosslare 1.7 - 14.8 0.7 22.0 2.3 69

Rothamsted 1.4 0.3 1.8 0.5 3.1 2.1 59

Stornoway 1.1 1.9 17.0 0.9 28.0 2.4 56

Valentia 0.9 - 8.0 0.3 12.7 1.4 72

Mam Tor - a 2.0 1.0 7.0 0.3 9.0 3.0* 1

Mam Tor - b 2.0 1.2 6.7 0.3 11.0 3.0* 1

*At lower detection limit of method used

 47

5. INTERPRETATION OF THE WATER CHEMISTRY DATA

5.1 Environmental Responses

5.1.1 Rainfall

After several weeks of analysis it was observed that high. concentrations

of ions in the stream waters occur at times of low flow. Conversely, the

lowest concentrations are measured after periods of heavy rainfall. It

appears therefore, that the waters are simply diluted by a large input of

atmospheric precipitation which is much less concentrated, in respect of

all ions, than the stream water.

Levels of dissolved solids in rain and snow fall are generally very

low over the British Isles (Stevenson, 1968), especially when compared

with the high concentrations present in the Mam Tor drainage waters.

Analyses of the two rainfall samples taken at Mam Tor during storm events
<,
are given in Table 7. These compare favourably with average values for

this part of the country, as given by Crisp (1968) and by Gore (1968). A

study of rainfall quality at Ladybower Reservoir, only 10 km northeast of

Mam Tor, is described by Rippon (1980). A full chemical analysis was not
2- -1
made but Rippon's values for S04 of between 3 and 5 ~g ml agree well

with others in Table 7. Rippon attributes the excess sulphate in drainage

water to additional input from dry fallout. Since stratigraphically similar

beds to those present at Mam Tor underlie the area, however, it seems

that a geologic source similar to that discussed in Section 6 is more

feasible.

The concentrations of ions in rainfall is usually somewhat higher

during heavy storms than during persistent light rain, and individual

ions can vary dramatically from one storm to the next; Ca2+ is supplied

from agricultural areas being limed, Na+, Mg2+ and Cl from storms at sea,
 48

and SO~- from industrial pollution (see Cryer, 1976a). Overall it is

considered that the supply of solutffito most of the Mam Tor waters from

atmospheric precipitation is minimal.

In other systematic· studies of solute inputs to catchment areas

(notably the detailed modelling at the Hubbard Brook experimental catch-

ment in New Hampshire, USA) similar inverse relationships to those found

at Mam Tor between stream discharge and ion concentrations have been

observed. These relationships are generally attributed to the mixing of

rain or surface water with deeper soil water (Johnson et al., 1969).

During the course of the Hubbard Brook study it was noted that some ions,
• I + 3+ •.• .
part1cu arly H and Al actually showed an ~ncrea8e 1n concentrat10n w1th

increased throughput. This is certainly not the case at Mam Tor, where

without exception, all ions show a decrease after heavy rainfall in streams

supplied partly by surface and rain water.

It must be remembered, however, that in the Mam Tor study there was

often a considerable time-lag between a rainfall event and the stream-

water sample being taken. It may be that immediately following heavy rain-

fall the concentration of some, if not all, constitutents does increase as

weathering products are 'flushed' out of pore spaces. This initial increase,

which may last for only a matter of hours or minutes, is then followed by

a progressive decrease in concentration over the next few days, producing

the observed dilution effect. Such a rise and fall in concentration with

the flood hydro graph has been noted by Cryer (1976b) and by Trexler et al.

(1975) •

Having stated that the Mam Tor stream waters are diluted by inputs

of precipitation, it should not be thought that all the streams show a

similar degree of dilution. As Johnson (op.cit.) notes, if a stream is

supplied with half of its water from surface runoff and half from deep

flow the effects of increased rainfall input will be felt much more strongly

than at another site which receives nearly all of its water from the surface.
 49

At Mam Tor most sites show a moderate dilution response. One site

(No.ll),however, shows very dramatic reductions in solute levels at high

flow and peaks during base-flow conditions. Still others (notably Sites

7 and 101) show very little or no response to changes in precipitation

input. Implications of these facts will be discussed shortly in examining

the various origins of the drainage waters in Section 5.4.

5.1.2 Temperature

Pitty (1976) defines five factors which influence the temperature of

stream water. These may be summarised as follows:

1. Air temperature at sink (limestone areas).

2. Air temperature at emergence.

3. Time of travel underground.

4. Mixing of underground waters.

5. Season (controls the effect of increased runoff).

Pitty concludes that air temperatures possibly play the major role in

determining the final temperature of most stream waters. Thus, it might

be expected that air temperatures during the Mam Tor study, although not

measured, will closely parallel the changes in water temperature given

in Appendix Bi.

At times of low flow, when the travel time of deep waters is increased,

stream water temperatures will be affected more by conditions underground

(Factors 3 and 4 of Pitty's scheme). Smith and Lavis (1975) note that

reductions of 4-50C can occur during such times, and temperatures tend to

equilibrate at about 70C.

At Mam Tor the various streams respond differently to changes in

ambient temperatures. Shallow, surface streams show the most dramatic

fluctuations, ranging from winter freezing to summer drying. Clearly such

streams are influenced entirely by air temperatures at the'time of emergence.

 50

Others show very little response to such seasonal changes, remaining

relatively warm in winter and cool in summer. Streams draining the lime-

stone cave system and emerging at Peak Cavern in Castleton (the control

sites) are such an example. These waters travel deep underground from

the swallet system at Perryfoot(GR. SK.lOl.812) to the resurgence at Peak

Cavern (GR. SK.148.827),a distance of 5 km. Pitty (op.cit.),using temp-

erature-lag data, estimates that the waters take some twenty-eight days to

travel this distance. Simpson (1954), however, records tests made by North

Derbyshire Water Board "which indicate a flow-through time of six days

from Perryfoot". It is likely that considerable variations from a few days

to several weeks occur depending on precipitation conditions.

The least variation of any stream at Mam Tor is shown by the waters

emerging below the retrogressive units at Site 7 (Fig. 5). In addition,

the waters here are much warmer than the 70C quoted by Smith and Lavis (see

Table 3). Again the importance of these observations will be discussed

further in Section 5.4.

5.2 Cluster Analysis

The degree of response to environmental variables discussed in the

last section was neither unique at each site nor the same for all sites.

Rather the sites seem to fall into a number of groups with strong intra-

group similarities and distinct inter-group differences. Similar groups

could also be distinguished on the basis of actual quantities of ions

present in solution (Section 4.4). Although it is easy 'by eye' to

divide the sites into groups on the basis of one or two attributes, it is

difficult to do so objectively and virtually impossible when dealing with

more than a handful of variables. There now exists a considerable volume

of literature concerned with statistical methods of multivariate analysis

(see, for example, Johnston, 1978) and after considering the possibilities
•

.
,

1S

10
-
LI'l
0
~

-
Vl
x

Vl
UJ

1 I
.1 I I
1 4 3 2 5 6 7

Groups
FIGURE 8: Cluster analysis hierarchy based on
annual mean data.
ESS = Ward's error-su~of-squares
distance measure.
Numbers 1 to 7 refer to the cluster
groups of Table 8.
•

15

10
-
Lno
~

-X

V)
V)
UJ

. 1
.~
4
~
3
,I
2
I
1
5
I
6 7

Groups
 .
Q)
'0
."
.
til
...-I .j,J
til
'0 ."§
~
til
...-I
Q) Q)
. >
."
Q)

..c::
.j,J
0
.j,J
Q)
. . til
til
Q)
~ Q) (J Q) 1-1

~
0 '0
."
.....
til
~
Q)
CO
1-1
0
.j,J

Q)
.
1-1
CO
0
1-1
.j,J
til '0 ::I '0 Q) Q)
...-I ~
'"
til .j,J 1-1
~ til
..... Q) ..... til
~ 1-1 til ~ ~
0 ..c:: '0 0
."
.j,J
.j,J
::I
~
0
~
1-1
~
til
'0
Q) Q)
Po 0 Q) ..... > til
."1-1 til til
Q)
>til
~ '"1-1 ,0
til
(J '0 CO C,) 0 Q)
til ~ til '0 .j,J
Q) til Po til til
A Q) ~ Q) Q)
..c:: Q) Q) CO (J CO
(J til Po. til til ~
.j,J
Po ~
'"1-1CO
'"
Cl)

~'"til
CO
~
.j,J
.
til
Q)
..c::
.j,J
Q)
Q)
til
1-1
::I
til Q)
13
~
'"
'0 .j,J
Q)
.j,J 0
CO
~
'0
~
Q)

0
~ '"Po '" til til '"til
.j,J
til 1-1
Q)

0 '" ~~ (J
Q)
1-1
Q)
.j,J
.j,J
Po
Q)
.j,J
til
~
til
Q)
1-1
Po
0
~
til
~
'"Po Q)
'0
~
.j,J I ::I
1-1
'"
(J ~ '.-I
'"til
Q)
1-1
..c:: Q)
Q)
~
0
Q)
1-1
Po
0
Q) '"
CO
1-1
(J (J .j,J I 1-1 0
~ 0 til til Q) ::I
Q) ~ Q) 1-1 .j,J Po
1-1 ~ 1-1 13 ..c:: >< Q)
0 ::I
'"..:I
.j,J (J Q)
Cl)
Z Cl)
0 '"~ A

0
0 til
M ..... ('f')
..... " ..
.. ..:r.. .......
til
Q)
.....
.j,J CO \0
." \0
.. .. .....
.....
.......
Cl)

N CO
0
.....
..
('f')
.....
..... ......

Po
::s
0 ..... N M ...r 111 \0 ......
1-1
o
 51

available it was decided that 'cluster analysis', a technique of hier-

archical classification, would be used in this case to produce an objective

grouping of the sampling sites. In this way it was hoped to reduce the amount

of data so that trends and responses to independent variables might be more

easily distinguished.

Various methods of cluster analysis are available on the ICL 19065

computer at Sheffield University, under the package name of CLU5TAN. For

the number and type of variables to be used here (seven numeric variables

with the same units - Ca, Mg, Na, K, Fe, Cl and 50 concentrations - and
4
three with different units - pH, temperature and conductivity) it was

decided that a most suitable method of analysis was that first described

by Ward (1963).Ward's method utilises an error-su~of-squares distance

measure to calculate individuals dissimilarity and gives the values of this

measure at which each group of individuals fuses to the next. The results

are best understood when represented in graphical form. Figure 8 is such

a cluster analysis hierarchy produced for yearly average data from each of

the fifteen sites. Essentially similar classifications were produced when

data for individual weeks analyses were used.

The most obvious split in the cluster analysis hierarchy is at a

very high level. This means that the groups on the left hand side of the

classification are quite dissimilar to those on the right. Table 8 shows

which sites make up the various groups.

Groups 1, 3 and 4, on the left of the hierarchy, contain surface run-

off sites on the landslide and the 'control'sites on Rushup Edge and in

Castleton (see Figure 5). The water at all seven of these sites has a pH

of 5.0 or above, and relatively low conductivity (low solute content).

Groups 2,5,6 and 7, however, comprise eight landslide sites whose waters

have a low pH and high ionic strength, most of which precipitate ochre.
 52

The most distinctive 'group' of all - that which branches from

the hierarchy at the highest level - is Group 7 which in fact contains

only one site. This is Site No.7, which was noted previously as being

different from the rest in terms of quantities of ions in solution and

response to temperature and rainfall fluctuations.

Table 8 shows that of the seven cluster analysis groups four contain

three sites and three contain only one site. For the three-site groups

the arithmetic mean value of every variable was calculated for each week.

In this way the data matrix was reduced in size from 10-by-15 to 10-by-7

(ten variables at seven 'sites'). The following series of graphs (Figures

10 to 19) show these group-averaged data plotted against time. Letters

along the abcissa correspond to the sampling dates given throughout

Appendix B.

Rainfall data obtained from the on-site rain gauge are given in Appendix

Bii and shownas a histogram in Figure 9. Differences between totals

recorded by the Mam Tor gauge and that of the Meteorological Office at

Hope, 3 km east of the landslide, were slight « 5%) and showed no

systematic variation. It might be expected that the Mam Tor gauge would

consistently record less precipitation than the one at Hope due to the

positioning of the rain gauge in the lee of the main scarp face in a

slightly sheltered site surrounded by trees (close to Site No.1, Figure 5).

This, however, was not the case and the two sets of data agree very closely.

Readings taken at the Hope station are used in Figure 9 between

09.01.79 and 27.02.79 as the Mam Tor gauge was buried beneath several

feet of snow during this period.

The responses of the group-averaged data to changes in precipitation

can be seen by comparing Figures 10 to 19 with the rainfall histogram.

Weeks with considerably less-than-average rainfall (e.g. J, S, CC) corres-

pond with sharp peaks on most of the other graphs. Conversely, sharp
FIGURE 9: Rainfall histogram for the entire study period.
Letters refer to sampling dates used in Appendix B.
•

z
o
...
c
...
0::
o
w
Cl:
e,

.,
e
(ww)

r
KEY FOR FIGURES 10-19

Group 7
---.-- Group 6
-------- Group 5
_a.._ ..._ .._ .._ ...._ Group 4
Group 3
----- Group 2
---------- Group 1
FIGURE 10: Variations in stream water temperature.
Group averaged data used.
t-
o
0

u
u
i
i ••
~
00
,..i ~
f N
\

'.
i, ~
I
I lC

! ~
j
:>
\
\ ::»
"'
.....
... .......... ...
, ...
'"
C

e
~
0

, Z

;
,i
:

~I
.
III

.•. ...
.. po
~
:>0
.. • '" . ... o
o

•
FIGURE 11: Variations in stream water pH.
 0
Q

CJ
\
U
\
10
'\, ID

I <C
i <C

,
I
I

I
:
N

>
)
, ,, I

,i I
I
I

)(

,, ,,
/

:
( ~
\
\
\
>
\,
\
\
,,
•
\ ,, :;)

\ , ...
\,...
", -,
'"
i
I IE:

\
\ 0
\,
~
,..'"
../ 0
I
; '\
I z
\
I \
:
\
I \
t

,,,
I
I

"
\
)

.. ,.. ""
,.. "
,
\.
./
" "
~......

.
",'
"..
=at ,..;\
i 0

0 0 0 0 0 0
0
...
0
CO III
• ,;, N

•
FIGURE 12: Variations in electrical conductivity.
-1 0
( 1-1 S cm ,25 c)
 \ ,.
-"J / ,,
\ I \
\

,,J
,,-' <, I \
,, :
\
" 1" \
\
'> .. , )
)
,....
~,J
i
r----" ( ( \
i
\. \ \ \ I,

-, : \
\
\
,
\
\
\ ~ , } \ I
\ \, I \\ ,
\..\ l
<, \
I. \
\
\
'
] ,i
\
, >

L
/ ~
'..
\
\
./ i
l
--'"
~~ .... -, \
" \~ .... ,\
,
/_,J....-
~:,/ \
i

,
-«
<' _,-
, __
~...L
(
I, ,/
I
I •
,

<. ~" ,
-, 1\ :
'", '
\
,
I'~ \I : I
,I
I, ) :
/ I" I \,
~
/
t ( " ,
..-
> ,~ ) \
-..
' ....~ <, <, \
> i
• ~,~~ '.> ;
CJ ~~~ ",'"
::J ;
cr r
--, / 1"-
C
Z (, i
o
CJ
\
/', \, 1
;
o o o o o o o
o
o
GIl
o
or
.....
o
o
o
cD
o
.... .,
o o
or
.... N
po po

•
FIGURE 13: Variations in calcium content of stream waters.
 c
c
N

•
Cl

•
>
• I· I""elf
..
o
o

•
FIGURE 14: Variations in magnesium content of stream waters.
 is
~/
-- -- -- -,~:" "
~

'
\
<,
\
\"
)
I

~8
I

,...__ -- -- ..., "I I '> "

,( ~
N
'
,
I
l I
"I
'\
~

'\ \ l
~
,, \

,I
~
I \
\',
)' IJ

"
>

--'
--'
~-_ -_, _
'-
-----,~/
c..( < ",
~
\ '
\J \ ..'
,)
I

... '-
...., "-
......
~I ....
,I
__ .-.rJ
<-- , / "
.........
1>" I )
("
-> \ ,.
\ I
"
z
\ ; \ (

/
> !; 'I
L.._ r» \{... 1...... ..
;> ; ~
-
.", -: (>

...'

~I
\
-~ - ~ ~ ~~
)---
-- __,
( -;1"'-
~ ,'_ ._", ,
C...... , \'. "~
" .' .\ /
) I
\ ji

o o o
8 •
.
t"
L~~"A1',;.
4i:': r. '.':
"
po ~

•
FIGURE 15: Variations in sodium content of stream waters.
 0
.........
'I 0
...._J,
»:
"
U
t U

\ '\\.<,
,1
III
III
\,' <, C
t' ')
~ .
,f (
C

"
",'.. <,
,,
,), ",
>

,'\ \ ",
./
( ~ f'"
"
)(

"" ...~\ \) ~

,;'-/..;, >

",4..'
, I "

/
)
~

,I ...
)_.., ~
~-;"
c..._
_- e,

--- ...~:~--- /
Co "

,
(It

er.:

,,
),
," 0
,
,,',' , I
--
__ ' ...~ J CL.
,
------ <...' , ... ' /'

<"-----
, .......".......... r"
",-
,'"
....
0

" , "-
... ',,\ z
,: \ I
,, \ /
\1
l
,,'/
....... A,..... ,'1
- ..,<...... "-
--~ _-:'._.>
_:.:..------ -r ....~...--
--_ 'l·--
- -,
(~~--t~-----
\
- - --.
~I "'-
1 -~
t,' »>
....... ~

T-------~------~------~~~--j
...
o

o o o o
• '" ....
,.IW Olf
•
FIGURE 16: Variations in potassium content of
stream waters.
 ell

, ! Cl
\ / !
J! .i ••
r " .I
r:
,
\
,•
(" c
c

,,, \

, \
\ '\ i
'\
..... !

>

I
I

•
\
I

-- -
\
__, / \
\

.-- ,\
--.....:::--_ - -::.
(
, ~
,
\
~ (
,-
,
~I \

)
,/
)

..
Q

•
o
• ..
o o
•
o o 0
f'I
0
N
0

\ .
•
FIGURE 17: Variation in total iron content of
stream waters.
 ,, 0
,, 0

,
\
,
\
I
I
J
, I

,,
(

,
•
,
I
I

\
\

~
\
\
)

- -~--
---
,
- - ....
-» ,- _..->
l
Lfl \. ....
.... "I
.-. -
o·
.. 0

•
FIGURE 18: Variations in chloride content of stream
waters.
 " ~
')'~\
I '

r ,'"
, ,
I
'./
<>
,')
.,
, ,«"
,, (,,~
""
,
,> '1
,, /',
<<, \ I I
<,
'\ ,
I

, \,
}~
c\.
I
'l "

,l,'
<.,,-,,.#

,-
....
;'l"
J JJ
..".,t~,,/
\
_- _-..- - -,
..... ..,. ... ,.,.~,
.
/
\ " /
\ '\ (
" I

\ \
:.'- ...- __ - -_ ... --,.,\,...... L)
< / --, ..

,,-- --
fIIIIII"~ ... --~'
- ?....
~
....
,..--~--- ;~) '--;;;..
...
~=-----:~;:

U !-I I
"
\
I

..s .,
o 000
,., '" .. 0

•
FIGURE 19: Variations in sulphate content of stream waters.
 }' I '
(\
\
<, \
\ \\ \\ "
...;._ :
.J- "> !
)
,~......
"'--7
" '
I ,-'
e" I I ;
'-<, {\ I
<, \ I
<, " ... ...
,>' I I

<,/ \ \ I

>
<, '. \
"........ ' '\ \ ,,
'41
,, •
~----pt ,,
........
..._ \
I I
\
\

~ \ ' , I
\~ I I

,I
->: I ,,
«:......... '' I I
-;.... I \
\
,
1:"'>1 I
I
I

,A I
r" ...... I I (
I

\ (,' \ I

\ '
\ \ ~
"" <,
\

,>
\'
\
'>
"I

I' / ,
t \
/ ...
I

"
~_, ...:» ,
(" <' / I
I
). '\ I
,_' :, \I

0
0
:II
0
0
N
f'I
0
0
•
N
..
0
0
N
0
0
0
N
o
..
o
o

,
•
 53

drops in solute content occur when precipitation is high (e.g. H, K,

AA).
Temperature variations throughout the year are considerable (Figure

10). The only exceptions, as noted previously, being 'Groups' 4 and 7

(Sites 101, and 7, respectively). The remarkable consistency of the

latter is clearly evident. So too is the clustering of pH values around

3.0-3.5 in Figure 11, which suggest that a natural buffering system is

operating on certain waters (see Section 8.3).

5.3 Charge Balance

In any ionic solution the overall net charge must be zero. This is

to say that the positive charge contributed by the cations must be exactly

balanced by a negative anion charge. The charge attributable to any

single ion can be calculated by dividing the amount of that ion present

by its atomic or molecular weight and then multiplying by the valency.

2
For example, the charge contributed by 100 ~g calcium (Ca +) is given by:

100
40.08 •
(+2) = + 4.99~E (~E = microequivalent)

2-
and that by 100 ~g sUlphate (S04 ) by

100
96.06 • (-2) = - 2.08~E

In th'lS stu dY th e quantltltes

" 0f f'lve catlons
, (2+
Ca ,Mg 2+ , N a 2+ , K+
- 2-
and total Fe) and two anions (Cl and S04 ) were determined each week.

A charge balance can be calculated for every water sample based onthese

seven variables. H+ is present in such small amounts, even in pH 3.0

waters, that its contribution to the balance will be very slight and is

omitted here. It is assumed for the purpose of these calculations that

2
any iron present is in the ferrous (Fe +) form.
 TABLE 9: Charge Balances Calculated for Group Averaged Data
(all values in ndcroequivalents)

Group
Week
1 2 3 4 5 6 7

D -0.310 0.032 -0.413 2.113 -0.323 -8.308 -20.554

E 0.012 -0.779 -1.093 1.714 0.340 -5.044 -8.906
F 0.743 0.243 -0.798 2.440 2.848 -1.987 -19.867
G 0.781 0.444 -0.378 2.054 -0.446 -0.293 -9.901
H -0.072 0.956 -1.077 2.349 1.520 -9.486 -7.076
I 1.295 0.618 -0.714 1.908 -1.490 2.425 -12.585
J 0.776 -1.544 -0.365 2.646 -2.426 -1.985 -12.363
K -0.473 -0.604 -0.832 2.330 -4.251 -4.287 -5.692
L 1.945 1.076 0.228 3.773 -0.235 1.593 0.001
M -0.758 -1.884 -0.684 2.425 -3.006 -8.243 -20.617
N -0.588 -2.376 -0.893 1.796 -5.990 -6.708 -14.728
0 0.740 0.783 0.412 2.959 2.352 0.207 -7.839
p 0.894 -0.417 0.279 3.233 -0.696 -4.456 -12.232
Q 0.070 -0.984 0.149 3.178 -1.650 -6.243 -10.291
R -0.422 -1.054 -0.297 4.535 -3.018 -7.371 -15.973
S 0.603 4.548 0.319 4.947 1.787 -0.718 -15.837
T 0.548 -0.441 0.106 3.551 -0.357 -3.431 -11.637
U 0.043 -0.869 0.012 2.748 0.372 -4.569 -12.415
V -1.127 -1.193 -0.048 2.600 -0.972 -8.441 -16.350
W -0.117 -2.128 0.101 2.918 -1.454 2.746 -17.792
X -1.041 -0.237 0.187 3.153 -2.373 -4.351 -18.077
y -2.047 -1.328 0.412 3.465 -4.280 -8.744 --9.108
z - - - - - - -
AA - - - - - - --
=
BB -2.779 7.700* 0.401 3.378 6.751 -7.421 -12.810
CC -1.283 6.745* 0.689 4.150 1.211= -7.511 -16.424
DD 0.318 10.348* 0.836 3.224 -2.164 = -7.216 -16.901

Mean -0.090 -0.325 -0.127 2.943 -1.079 -4.395 -13.040

Mean Sum
8.786 15.476 2.351 4.920 19.841 21.236 .38.561
Cations
Mean Sum
8.444 15.555 3.059 2.725 20.369 24.856 50.679
Anions

- Calcium values missing, balance not calculated

* Site No.4 dried up
= Site No.3a dried up
1 Values excluded from mean calculations
 54

Table 9 shows the values of Ecation charge minus Eanion charge as

calculated for the group data. The calculations were made using a

'Commodore PET' mini-computerusing a program written for the purpose in

BASIC language.

Group 1 waters show only a very slight deviation from the actual

charge balance of 0.00. The mean value of -0.09 vE is only 1% of the

total cation or anion charge. This value is considerably better than

that quoted by Rainwater and Thatcher (1960). In their experiments, using

only very dilute solutions (T.D.S. = 30 Vg ml-l) the best charge balance

they could obtain was ± 5%.

Groups 2, 3 and 5 also show a good balance, 2%, 5% and 5% respectively.

The charges do not balance, however, for the remaining three groups. The

waters of Group 4 show an apparent excess positive charge which is probably

balanced by additional anions not included in the analyses. Indeed it was

noted during the course of pilot study (p. 40 ) that HCO; ions were present

in these karst waters (this also accounts for deviations in the T.D.S.-

CONDUCTIVITY curve shown in Figure 7).

In contrast, a large excess negative charge is the characteristic

feature of Groups 6 and 7. This is partly explained by the fact that

iron was assumed to be in the ferrous form. It is known that some also

exists as trivalent ferric iron (p. 44 ) and this would increase the posi-

tive charge if taken into account in the calculations. Another trivalent

•
~on known to b e present 1n
• th e waters of Groups 6 and 7· ~s Al 3+ and aga1n
•

calculations would be improved by its inclusion.

Thus it appears that the calculation of charge balance serves as a

useful 'check' on the analyses. If the inbalance is constantly to one

side (positive or negative) for any given stream then the analysis of that

water is probably incomplete or is in error. It would seem that the use of

only seven ions in this study is sufficient for all but three of the sites,

and in these cases the 'missing' ions are known.

 -
§I~
>:
~Io
...... '-'
......
~ ,... ...;t ..;t 0 \0 0 N
'-' N .-4 M N ~
0
,...
,...
~
t-e

..;t

......
~
~
U
CO ..;t C""I 0 · N
M
\0 CO
M
'-' ~

N
·1· ..;t

...... ~
Cl
...... N
~
\0
0'1
..;t
~
M · '"11'\
......
M
0'1
......
'-' + N N
~
-e
'-'

\0

·
00

-
U
'-'
p..
::I
,...
0
t!)
..;t
0
N
N
0'1
~
~
M 00
CO
~
M
~
..;t
~
.-4
,
~
r-I
S
, >: 0()
;:1

r::
• .-1
......
·
..;t Ul
p.. Cl)
...... ::I ..;t 11'\ ...;t 11'\ 0'1 N \0 ::I
~ 0
,... 0 00 r-I 0 ..;t CO ~
'-' ..;t ~ ~ ..;t lIS

o
c.!> N >
r-I I>: ~
r-I
<
M

-
<
'-'
o,
::I
,...
0
0 ...... -e- · .
..;t

.-4
\0
N N M
......
c.!> N ~ M

,>:

+
N
lIS
+
N
0()
+ lIS
+ Cl)
,~ , N...;t
0
U ~ Z ~ ~ U Cl)
 55

Many other ions will be present in the waters sampled and all will

contribute to the charge balance (unless species with zero charge occur).

The additional quantitites of ions and charges will be small, however,

in comparison to the ten or so major ions considered here.

5.4 Origin of Drainage Waters

5.4.1 Deep and Surface Components

As noted in Section 5.2, the various groups of waters show different

responses to variations in precipitation. Following the model of Johnson

et al. ,(1969), it would appear that if the water emerging at any particular

site shows a strong dilution after rainfall then the water at that site

is derived partly from deep flow and partly from surface runoff supplied

by the rainfall event. In this study the waters of Groups, 1,2,5 and 6

can be thought of as originating in this way. The deep flow component.

comprises waters which have reacted with the rocks within the landslide

to give their characteristic chemical composition. Differences in responses

and overall water quality between these four groups reflect different

proportions of deep to surface flow.

A quantitative mixing model can be derived by trial and error if it

is assumed that simple proportions of deep and surface water mix at any

given site and provided analyses of the two components are available. At

Site 11 (Group 6), on Figure 5, water from Site 7 (Group 7) and Site 8

(Group 3) meet and mix. The chemical compositions at all three are known,

and Table 10 shows the results of a simple theoretical mixing of equal

volumes of Group 3 (surface)and Group 7(deep) waters.

Yearly average data are used at all times. The percentage differences

between the theoretically 'mixed' waters and the true composition of Group

6 are quite small considering the simplistic approach adopted. Sodium and

chloride values are in error presumably due to unknown supplies from road
 56

salt. Also iron may be involved in precipitation and re-solution between

Sites 7 and 11. 'Mixing' these waters in different proportions (60:40,

20:40 etc.) produces worse results than the simple 50:50 mix discussed

here.

There appears to be no dilution of the deep flow component by surface

runoff in the case of Group 7 waters. This results in the lack of response

to precipitation and temperature changes and consistently high solute con-

centrations noted in Section 4.4

One interesting feature of Site 7 is that the flow rate does not appear

to fluctuate from week to week. At all the other sampling sites variations

in discharge could be observed, albeit subjectively, throughout the period

of study. For the reasons outlined in Section 1.3, it was not possible to

regularly monitor discharge from the Mam Tor streams. Since Site 7 appears

I to be so distinctive, however, it was decided that discharge measurements

by the rather laborious technique of 'diluting gauging' should be made,

if not at all the sites, then at least here.

The very low solute content of the Group 3 waters results from them

being almost totally derived from surface runoff, with no deep flow com-

ponent. Any reactions taking place here involve relatively unmodified

precipitation and soil minerals/organics.

The control waters of Group 4 are chemically very distinct from any

other group. Their chemical composition is entirely due to reactions

involving carbonates as they pass through the limestone. These waters

and reactions in which they are involved are not central to the arguments

to be put forward in this study and need not be considered further.

Figure 20 summarises the likely mode of derivation of each of the

seven groups.
FIGURE 20: Flow diagram showing the derivation of the
seven groups of waters.
 57

5.4.2 Dilution Gauging

Dilution gauging is most easily carried out using a constant rate

injection method (see B.S.I., 1964). The principle of this technique

is that if a chemical tracer solution is added to a stream at a constant

rate then an increase in concentration of the solution's constituents in

the streamwater will eventually become apparent at a downstream sampling

point. For a given tracer solution the increase in concentration will

be inversely proportional to streamdischarge. It has to be assumed that

the tracer solution mixes completely with the streamwater, and hence a

sampling point located a considerable distance downstream from the in-

jection point is desirable. If complete mixing is achieved a constant,

r~ised level of concentration is observed downstream and this'plateau'

is maintained as long as the upstream injection continues. The dis-

charge of the stream is calculated in the following way:

Ct - Cds
Qs = Qt • Cds - Cus (6)

where: Qs = stream diSCharge}

-1
(1 sec )
Qt = tracer discharge
-1
Ct = tracer concentration (mg 1 )
Cds = streamwater concentration at
sampling point (mg 1-1)
Cus = streamwater concentration above
injection point (mg 1-1)

The tracer solution used in this case was NaCl and a constant rate of

injection was achieved using a Mariotte Bottle. This simply is a constant

head device of about 20 litre capacity shown in diagrammatic form by

Figure 2lc. Sodium chloride was chosen as the tracer because the con-

centration of Na+ and Cl- ions in solution at Site 7 are low and certainly

very much less than in the saturated NaCl solution in the Mariotte Bottle.
FIGURE 21: The principles of dilution gauging.
See text for explanation.
 a)

Cds plateau
----------------- GCn'Jing Curve
2.200

-
Vl
>-
~ 1.000
t=
u
:::>
0
~u teco

\600
10 20 30 40 50
TIME (min.'
b) c)
Calibration Curve

200
~Cds
c •
LLJ
o
~100 .:::_•
a: a. •'.-:::...
LLJ
':;J.
a:
-==-
I- t --------------

1600 1600 2000 2200 2400

CONDUCTIVITY (S)

Ct = 7609 NaCl in 20l Cds = 2250 S =185ml tracer

= 38000 mg le' = 7030 mg in 20t
= 350 mg r'
at = 2l in Smin
= 0004 l f'

•
 58

Thus Cus ~ zero and Ct is also much greater than Cds. Equation 6 can,

therefore, be conveniently reduced to:

Qs = Qt.Ct (7)
Cds

Qt can be preset by opening the tap to give a convenient flow rate and

calculated simply by timing the filling of a beaker of known volume. Ct

is known from the mixing of the original solution. Cds is most accurately

measured by titration of strearnwater samples taken during the 'plateau'

phase but a more suitable field method is to estimate Cds from simple

conductivity measurements made at the downstream sampling point using

the field apparatus described in Section 4.3.

This is done by firstly preparing a calibration curve of conductivity

against concentration by adding successive volumes of tracer solution (say

10 ml) to a large, known volume of streamwater. The conductivity of the

mixture is recorded after each addition. Once the calibration curve is

constructed the tracer concentration in streamwater can be found by

measuring only conductivity at the sampling point (see Figure 2lb).

A conductivity meter was, therefore, placed on the stream bank

approximately 30 m downstream from the injection point with the electrodes

immersed in the streamwater. Readings were taken every two minutes and

the resulting graph is shown in Figure 2la.

The discharge at Site 7 was calculated from Equation 7, as follows:

Qt.Ct
Qs - (7)
Cds

0.004 x 38000
= -.....;_;...,,3...;:,;5-0~;...;...;...

-1
= 0.43 1 sec
 59

The discharge at Site 7 was, in fact, measured on two separate

occasions. The first, for which the above data are given, was on 1st

November, 1979, when it was noted that the other streams and seepages

were flowing at well below their usual rates. The second occasion

followed a week of almost continual light rain, and on 20th December,

-1
1979, a discharge of 0.44 1 sec was recorded.

Although only two quantitative measurements were made they do attest

to the constancy of flow at this particular site under different climatic

conditions. When visiting the site regularly for a period of twelve

months differences in flow rate from week to week can be recognised. As

noted in Section 1.3.1 qualitative information regarding changes in dis-

charge at each site was entered into a field notebook. The only notes

recorded at Site 7 read 'no change' and it is believed that the flow
-1
rate of approximately 0.4 1 sec is maintained virtually constantly

throughout the year.

5.4.3 Additional Inputs

The constant flow rate discussed in the preceding Section poses an

interesting problem regarding the origin of the Site 7 waters. A dis-

-1
charge of 0.4 1 sec is not very great but would certainly require a

catchment area greater than that present to maintain it throughout the

year. The maximum reasonable area which might contribute water to this

site is the ground above and including the retrogressive units. This
2
immediate catchment has a maximum area of 6,400 m as measured from the

air photo (Plate 12). Total rainfall received during the period of

study was 1,137 mm, giving a volume of water supplied to this area of
6
approximately 7.28 x 10 1.
6
A stream flowing at 0.43 1 sec-l,however, removes some 13.3] x 10

litres in a year. Therefore, an additiona16..n9 x 106 litres (46% of the

total discharge) must be supplied from elsewhere. It must be remembered

 60

that this figure is a maximum since it is assumed that all the water

falling within this catchment (which is also probably an over-estimate

of the actual contributing area) arrives at Site 7. From considerations

of the geology and topography of the retrogressive units this seems highly

unlikely. The dip of the beds into the main scarp face would tend to

channel water away from the site, and evaporation losses from such an

area of bare, often fragmented, shale must be large. Also it is unlikely

that all the water held in the area of the retrogressive units should

emerge at only this one site. Indeed after heavy rainfall water has been

observed to seep out of the shales and move in the direction of an ephemeral

stream which flows between Sites 8 and 11 (Figure 5).

A larger catchment area than that of 6,400 m2 quoted here for Site 7

seems unlikely. Above the retrogressive units surface water would naturally

be channelled northwards, due to thegradient, on to the main unit. Any

deep water present here would be carried westwards and northwards, down-

dip, away from Site 7. Similarly the dip of the beds off the landslide
o
area to the southwest is in the 'wrong' direction (c. 5 northwest), and

surface water in this region can be seen on the air photo to be flowing

in gullies and shallow seepages parallel with the south flank of the land-

slide towards the Odin Sitch.

The most feasible alternative to a larger catchment area is that over

6 million litre of water annually are supplied by groundwater recharge.

One feature, clearly visible as a distinct break of slope on the air photo,

and also marked on Figure 5, which might channel such groundwater towards

the retrogressive units is the fault running along Rushup Edge in a west-

southwest-east-northeast direction.

Such a fault, which here cuts through the uppermost horizons of the

Edale Shales, would normally be associated with an area of highly fragmented

or brecciated rock - the fault crush zone - which acts like a 'sponge',
 61

attracting groundwater largely by capillary forces. The groundwater

could thus be drawn from a cons~derable area, and might possibly upwell

from great depths {the vertical extent and throw of the fault are not

known} into the fault crush prior to emergence at Site 7.

Since the eastern end of the fault is so closely associated with the

retrogressive units it might even be related to the initiation of movement

of this area after the main unit had itself slipped. This could have

occurred by re-activation of the fault during recent geologic times or,

more likely, the structural weakening of the rocks by its presence.

If the groundwater is uprising from depth towards Site 7 in this way it

may account for the elevated and constant temperatures which have been

observed {Section 5.l.2}. A long residence time deep underground, where

the geothermal gradient is higher, is easily envisaged. Alternatively,

the observed temperature and chemical anomalies at Site 7 might be due

to exothermic reactions taking place between groundwater and minerals of

the fault crush. Such reactions are considered in detail in the following

Section.
 62

6. DEVELOPMENT OF A WEATHERING MODEL

In the previous two Sections, the results of chemical analyses of

the water samples were discussed and the extreme composition of spring

7 was noted. It is intended in this Section to account for the pro-

portions of ions in the Site 7 water by reference to reactions taking

place at depth, possibly within a fault crush zone, between percolating

ground waters and solid mineral phases.

By considering each major ion in turn, starting with the most

abundant, sulphate, the most likely reactions responsible for their

liberation can be put together in sequence and quantitative values

ascribed at each stage.

In order to do this it is assumed that all the SUlphate generated

in the first reaction remains in solution and does not participate in

further mineral transformations prior to emergence at Site 7. This

assumption will be qualified towards the end of the Section before con-

sidering some of the wider implications of the model.

6.1 Weathering Reactions

6.1.1 Pyrite Oxidation

Since many of the streams draining the Mam Tor landslide precipitate

ochre and gypsum, the waters must be highly concentrated in iron, sulphate

and calcium. A possible source of the former two components is the oxi-

dation of pyrite by percolating waters charged with dissolved oxygen.

Since the Edale Shales, which comprise the bulk of the landslip material,

are known to contain abundant pyrite and the mass as a whole is highly

fractured by landslide movement (Section 2.2.1) the contact between such

waters and minerals is inevitable.

There is considerable disagreement, judging from published work, as

to the actual mechanism of pyrite oxidation, which is most frequently

 63

written as:

In an early paper by Winmill (1915) this equation is given as:

(9)

or
(10)

depending on whether ferrous or ferric sulphate is formed.

The true mechanism of pyrite oxidation is probably far more complex

than these reactions suggest. Banerjee (1971) describes thermoanalytical

techniques of investigating the various ~eps involved and intermediate

products formed. Much of this work, however, is conducted at high tempera-

ture and such experiments are clearly of little consequence in weathering

studies. An earlier article by Mapstone (1954) describes the reaction as

taking place by several steps even at low temperatures. Although some of

these steps seem a little unlikely the sequence of five reactions does

account for virtually all the possible direct'products of oxidation.

The sequence is written by Mapstone as follows:

1. Oxygen is absorbed on pyrite surfaces, liberating sulphur

dioxide in the presence of moisture:

(11)

2. The reactive ferrous sulphide is directly oxidised to sulphate:

(12)

3; The sulphur dioxide of step 1 is oxidised to sulphuric acid:

(13)

4. The interaction of ferrous sulphide with the acid liberates hydrogen

sulphide:

(14)
 64

5. Interaction of this with the sulphur dioxide gives

free sulphur:

Saksela (1952) and Nambu (1957) suggest alternative mechanisms for

the reaction, neither of which entirely agree with other literature.

Steger and Desjardins (1978) also note that: "Previous studies on weathered

pyrite do not show a consensus on either the products or nature of the

oxidation".

In an attempt to settle the dispute these authors oxidise three sul-

phide minerals - pyrite, chalcopyrite and pyrrhotite - in a constant tempera-

ture/humidity chamber at 520e and 68% RH (conditions which do not entirely

represent those of most weathering situations). They note that: "The

major difficulty encountered in the study of the oxidation of sulphide

minerals at temperatures less than 1000e is the identification of the

small amounts of products formed during reasonable periods of time".

Steger and Desjardins found that (in their experiments) the oxidation

of pyrite proceeds to give the metal sulphate as the predominant product.

Some metal thiosulphate which also forms is later oxidised to sulphate:

(16)

or
o
2FeS2 + 302 + 2FeS03 + 2S (or S02 t) (17)

In most natural situations where pyrite is being oxidised, however,

water is present and the chi~f product then is an acid iron sulphate

solution. Garrels and Thompson (1960), therefore, quote the reaction as:

(18)
 65

Although in their weathering experiments Garrels and Thompson use

solutions of extreme acidity (pH 0-2) it does seem likely, where ferric

iron is present in solution, that this mechanism will predominate over

direct oxidation by atmospheric oxygen.

Despite this lack of agreement as to the detailed mechanism of pyrite

oxidation the final products are reasonably well known. Much of the work

on sulphide mineral weathering has been conducted in the mining districts

of Colorado, in North America. Here pyrite·, in close association with

copper sulphide, is ubiquitous throughout most geologic formations,

occurring as disseminated grains, in veins, and as large discrete crystals

in rocks which otherwise show no effects of hydrothermal alteration (Harrer

and Tesch, 1959). Much of this pyrite is presumably of diagenetic origin.

Here in a temperate region with moderate annual rainfall totals, the chief

minerals formed as a result of rock weathering are 'limonite' and geothite.

Hanshaw (1974) describes the geochemical evolution of a geothite

deposit in Colorado formed where waters percolating through the pyritic

rocks emerge in springs and seepages very similar to those at Mam Tor.

Hanshaw states that the de-oxygenated waters, on emergence, first form

'limonite' according to the reaction:

(19)

This 'limonite' then dehydrates with time to form the oxyhydroxide geothite:

(20)

Most of the oxidation reactions cited above have one productin common

acidity. This is represented by either large quantities of hydrogen ions

(Garrels and Thompson, Equation 18) or a combination of these and sulphate

ions, sulphuric acid (Winmill, Equation 9 and 10).

As stated previously, much of the work on this subject has been con-

ducted in mining areas. Whether these be sulphide mineral mining or coal

 66

mining areas (pyritic rocks are commonly found in association with coal

seams) a common descriptive term is now applied to the water produced.

It is simply known as 'acid mine drainage'. A vast body of literature

has emerged on this topic since winmill's (1915) time, particularly from

such mining areas as the Appalachian region of the USA where the problem

is particularly acute and considerable environmental pollution has resulted.

A discussion of the consequences and possible controls of acid mine

drainage is contained in Section 6.2. It is sufficient to note here that

such waters have a very similar chemical composition to those draining

the Mam Tor landslide and that pyritic source rocks appear to be common

to both situations.

6.1.2 Carbonate Dissolution

ca2+ ions may be supplied to the Mam Tor waters by the weathering of

diagenetic carbonates (described in Section 2.2.1) which occur in close

association with pyrite in the Eda1e Shales.

The simplest and most frequently quoted carbonate dissolution reaction

is given by:

(21)

where carbonic acid, formed by the solution of atmospheric CO2 in rainwater,

2+
attacks limestone (calcite) to produce Ca and more stable bicarbonate ions.

The Mam Tor carbonates are commonly composed of dolomite or ankerite,

minerals which are less soluble in carbonic acid (Krauskopf, 1967, page

109). In a more concentrated acid solution - such as acid mine drainage -

the dissolution of dolomite can be represented by:

+ 2+ 2+
Cal.lM~.9(C03)2 + 4H = 1.ICa + O.9Mg + 2H20

+ 2C0 t (22)
2
 67

• reactlon
ThlS •• Yle Id s Ca 2+ an d Mg 2+. lons ln
• roughly the correct

proportions, as found in the Mam Tor waters. Streams draining the lime-

stone area to the south have ratios at least fifty times greater than

this (see Appendices Biv and Bv) although the actual quantitites of ions

are much smaller.

It should be pointed out that the above formula for dolomite rep-

resents an ideal composition. The actual proportions of calcium and

magnesium can vary considerably; many sedimentary dolomites contain far

more calcium. Iron substitution for magnesium is also common and where

Fe > Mg the mineral is known as ankerite (Read, 1970, page 277).

Since the exact proportions of Ca, Mg and Fe are not known for the

Edale Shale dolomites an ideal formula is used.

6.1.3 Clay Mineral Transformation

The only additional major mineral constitutents of the Edale Shales,

assuming the gypsum and iron oxides are weathering products and not primary

components, are clay minerals.

Cation exchange is the main mechanism of metal-ion release in clay

mineral weathering and the exchange of hydrogen ions in, for example, an

illite to kaolinite transformation, can be represented by:

K2A14 [Si2AI2020] (OH)4 + 2H+ + 3H20

= 3A12Si20S(OH)4 + 2K+ (23)

where an ideal muscovite formula is used. This is not true'exchange' since

the process is irreversible, but nevertheless potassium ions are released

into solution; these constitute yet another component of the Mam Tor

waters (Appendix Bvii).

Kaolinite is invariably the stable end-product of silicate weathering

(as examplified by the formation of china clay deposits after the weathering
 68

of Cornish granites). In extremely acid solutions, however, kaolinite

itself may be dissolved according to the reaction:

(24)

3
th1s 1n turn consumes more H + 10ns an d Li.b
o 0
1 erates A1 + • Minor amounts of
0

2+ 2+ 2+ +
other cations, including Ca , Mg , Fe and Na , may also be released

during such clay mineral transformations.

+
Na and Cl ions are not considered in this model since the greatest

proportion is supplied by road salt (Section 4.4.2).

6.1.4 A Quantitative Weathering Model

The net result of pyrite oxidation, irrespective of the various possible

mechanistic steps, can be summarised by the two end-member reactions:

(25)

(26)

depending on the degree of oxidation. Since most of the iron in the Mam

Tor waters is in the ferric form, the second equation is probably more

useful. If it is assumed that one product of this reaction remains in

solution and does not participate in further chemical reactions prior to

emergence in the springs and seepages quantitative values can be assigned

to the various reactants and products based on the molar quantities present.
2-
In reaction 26, S04 is most likely to remain in solution as a simple

ion. Therefore if the quantity of sulphate is determined the amount of

pyrite destroyed per litre of water can be calculated.

Waters emerging at Site 7 are least affected by mixing and dilution

(Section 5.4). Therefore, the sulphate content of these waters is probably

most indicative of the reactions taking place at depth. The mean annual
 69

S02- concentration at Site 7 is 2486 ~g ml-l or 25.880 mM 1-1 (the annual

4
mean value, calculated from Appendix Bxi).
-1 2- 26
Thus if some 26 mM 1 S04 are produced in Equation 26, then:l mM
-1
1 pyrite must be consumed (since 2 moles sulphate are produced for

every mole of pyrite).

2; mM pyrite = ;6 x (molecular weight of pyrite) ~g ml-l

= 13 x 119.97

= 1559.6 ~g ml-l
-1
= 1. 5596 g 1

Thus some 1.5 g pyrite are destroyed by every litre of water in liberating

2.486 g sulphate. Similar calculations show that, in addition, 0.7260 g

3+ +
Fe and 0.0131 g H are produced.

If all the iron in solution in the Site 7 water is assumed to be in

the ferric form (for the purpose of these calculations), then the average
-1
concentration, from Appendix Bviii, is 0.109 g 1 • This is considerably

less than the amount produced by pyrite oxidation. Therefore, it is postu-

lated that some of this 'iron (0.617 g 1-1) must be precipitated prior to

emergence as either 'limonite' or goethite according to the reactions:

(27)
or
(28)

If 'limonite' is precipitated according to the first equation 1.1804 g

hydroxide are precipitated and 3.34 x 10-2 g H+ are liberated. According

to the second equation some 0.9816 g goethite would precipitate and

-2 +
2.23 x 10 g H would be liberated.
 70

Such calculations balance the ferric ions but precipitation of ochre

in this way liberates further hydrogen ions, making the waters even more

acid. If the amounts of H+ ions present at this stage are summed the

following observations are noted:

-1
Quantity of H+ ions from pyrite oxidation = 1.31 x 10-2 g 1
-1
Quantity of H+ ions from ochre precipitation = 3.34 x 10-2 g 1
2 -1
Total = 4.65 x 10- ~ 1

This concentration represents a pH value of 1.33. The mean pH of the

Site 7 waters however, is only 3.49 which corresponds to a H+ concen-

-4 g 1-1•
tration of 3.2 x 10

Thepefope~ ovep 99% of the acidity genepated by PVpite oxidation

and ochpe ppecipitation must be consumed in furthep minepa~ peactions

befope these waters emerge at the surface.

Clay mineral transformations of the type discussed in Section 6.1.3

(Equations 23 and 24) would consume some 2 x 10-4 and 9.7 x 10-3 g H+,

respectively, and produce a solution 0.2 mM with respect to K+ and 3.2 mM

3+
with respect to A1 • These values, corresponding to solution concentrations
-1 + -1 3+
of 7.8 Vg m1 K and 86 Vg m1 A1 ,agree closely with the observed mean

concentrations of 5.4 and 77 Vg ml-l (Appendices Bvii and Bix).

An excess of hydrogen ions (0.0366 g) still remains. The acid de-

composition of dolomite, as discussed in Section 6.1.2 (Equation 22) could

account for this discrepancy. 3.66 x 10-2 g H+ would destroy 1.688 g

2+ 2+
dolomite, yielding 0.4002 g Ca and 0.1986 g Mg (9.98 and 8.17 mM,

respectively). Once again these values agree well with measured concen-
-1 2+ -1
trations at Site 7 of 0.404 g 1 Ca and 0.185 m 1 Mg (Appendices

Biv and Bv).

It would appear, therefore, that a very simple weathering model based

on only five or six reactions and the water analyses described in Section
FIGURE 22: Sequence of chemical reactions affecting the
composition of water at Site 7. Figures at
the bottom indicate calculated and, in paren-
thesi
eS1S, 0b serve d concentratlons
• 0f· lons ln
. g 1-1 •

Values with an asterisk are those used in calcu-

lations.
 OXIDATION

15%

85%

HYDROLYSIS

118049
"limonite"

a: Mg1• Al·· K· H·
e 41)')21 O.~9~6g 0.OB65g 000789 00003g
r ~ 4 ,)4';~) [0 ~850g) [00773g1 100054g1 (00003g)

..
•
 71

4 can account for the observed water quality at this site. Figure 22

summarises the various steps involved.

-1 .
It is interesting to note that if a flow rate of 0.43 1 sec 1S

maintained throughout the year at Site 7 (Section 5.4) and each litre of

water has oxidised 1.56 g pyrite, then some 21 tonnes of pyrite are con-

sumed per annum. Similarly, 23 tonnes of carbonate are destroyed and 13

tonnes of goethite are precipitated.

6.1.5 The Sulphate Problem

The results of mineralogical analyses are not considered until Section

7.3 but it must be mentioned at this stage that the sulphate minerals

gypsum (CaS04.2H20) and jarosite (KFe3(S04)2(OH)6) are found in samples

of shale from Mam Tor. The former is only found in surface outcrops of

Edale Shale but jarosite often occurs at considerable depth within the

landslide. The significance of this deep jarosite formation is that the

process consumes sulphate ions which would otherwise remain in solution.

Since SO~- concentration of the Site 7 waters was the basis for all cal-

culations in the preceding Section, the quantitites of minerals given in

Figure 22 must be under-estimates and more than 21 tonnes of pyrite must

be consumed annually.

Since some K+ is also consumed in this reaction more must be leached

from the clays in the first place. This may, in fact, account for the dis-

crepancy b etween teo

h b serve d an d ea 1eu 1ate d va 1ues 0f K+.1n F1gure
. 22 •

If some iron and sulphate is used up in jarosite formation the corn-

ponents of Equation 26 can be divided, in molar terms, as follows. The

equation must first be multiplied by a factor of 6 in order to produce the

24 millimoles (approximately) per litre of SO~- in the Site 7 water:

(29)
 72

This is the scale of the reaction necessary for the original weathering

model. More pyrite must be oxidised, however, in order to form jarosite.

This can be represented, schematically by:

14FeS + 105(0) + 7H20 - l4H+ + l4Fe3+ + 28S0!- (30)

2

8
1-----'---1 1--'----1
6 4 I 24

8FeOOH 2KFe3(S04)2(OH)6 24S0Z-

(gee thite) (jarosite) (stream)

Thus, 15% more pyrite is oxidised and 15% more iron precipitated.

Approximately 60% of this iron is in the form of goethite and 40% is

precipitated as jarosite.

6.2 Wider Implications of the Weathering Model

6.2.1 Catalysts

Curtis (1976) notes that the 'activation energy' of a reaction is

closely related to reaction rate. If a reaction can proceed via more than

one mechanism (which, from the discussion in Section 6.1.1, certainly

seems possible for pyrite oxidation) there will be several possible

'activation energies'.

According to Curtis "catalysts induce acceleration of reactions by

providing an alternative reaction mechanism with lower 'activation energy"'.

There is considerable debate in the literature of both microbiology

and water resources concerning the role of chemosynthetic autotrophic

bacteria as catalysts in sulphide oxidation (see tau, et al. 1970).

Such organisms, by definition, are able to build up their body substance

from carbon dioxide and certain salts using chemical energy obtained by

oxidising inorganic materials.

 73

Quispel et al. (1952), were among the first to suggest that the auto-

troph responsible for pyrite oxidation is ThiobacciZZus thiooxidans.

Leathen et al. (1956), however, point out that this organism does not

completely oxidise pure pyrite, and they suggest an alternative species,

FerrobacilZus ferrooridans, which apparently does, via the intermediate

production of ferric iron. It is nowadays considered that Leathen's

bacterium is really a variant of the species ThiobaciZZus ferrooxidans

which was first 'discovered' by Temple and Colmer (1951).

Numerous studies suggest that bacteria may be more important in

oxidising pyrite than oxygen. One proviso, however, would seem to be

that the pyrite is of small particle size (Temple and Delchamps, 1953) as

is the case when the naneral is disseminated throughout argillaceous rocks.

Leathen et al. (op.cit.) for example, found that the large concentrations
.
o f f errous iron (2
c. 00 mg 1-1) present in
. some mlne
. d ralnage
. waters was

completely oxidised by a culture of T. ferrooridans in three days whereas

the chemical oxidation of a sterile solution by atmospheric oxygen alone

took more than two years.

Similarly Colmer and Hinkle (1947), before ThiobaciZZi had been posi-

tively identified as responsible, treated acid mine drainage with anticeptics

and found a discrepancy in oxidation rates between the treated samples and

controls. Bryner and Jamerson (1958) have even demonstrated a technique

for the commercial extraction of metals from low grade pyrite ores using

these bacteria.

Bloomfield (1972a) in considering the role of iron-oxidising bacteria

in the formation of acid sulphate soils, describes an experiment whereby

Alum Shale is oxidised in the laboratory. He finds that significantly

more pyrite is oxidised in the shale when ThiobaciZZi are present and notes

that the maximum pH at which these organisms can survive is 3.5 to 4.0
 74

According to Kuznetsov et al. (1963) the growth of T.fer~ooxidans

cultures ceases above pH 4.5 and the optimum conditions lie between pH

2.0 and 4.0. They state that "Because of its ability to live under condi-

tions of high acidity, ThiobaaiZZus fer~~oxidQnscan withstand the large

concentrations of sulphuric acid that it forms as a result of metabolism"

(see also Lazaroff, 1963).

Kuznetsov et al. also note that the fracturing of pyritic rocks

produces favourable conditions for the development of micro-organisms by

promoting the access of water and by creating larger surface areas for

them to act upon. Nowhere do pyritic rocks become so fractured - except

perhaps, within a landslide - as in mine workings, where bacterial oxi-

dation can still proceed deep underground even though the supply of oxygen

may be severely limited (Singer and Stumm, 1970).

6.2.2 Acid Mine Drainage

6.2.2.1 Formation

The number of publications concerned with acid mine drainage has

increased dramatically over the last thirty years apparently in proportion

with those on microbiological oxidation.

Colmer and Hinkle (op.cit.) were among the first to suggest that acid

mine drainage is produced by the action of micro-organisms on sulphide

minerals. During the 1950's, as further work on ThiobaailZi was progressing,

many more researchers saw the connection (see for example Temple and

Delchamps, 1953; Ashmead, 1955). In the 1960's most of this research was

conducted in the coal mining areas of North America (Barnes, et al. 1964)

where in the Appalachian region alone some 6,000 miles ofstreams are polluted

by mine drainage (Appalachian Regional Commission, 1969). More recently

attention has focused on the more localised but more intense problems

in areas of sulphide mineral extraction (Hanshaw, 1974; Potter, 1976;

Wentz, 1974).
 75

In a study of extremely toxic discharge from a lead/zinc mine in

Idaho (see Table 11, sample 1) Trexler et al. (1975) note that in

addition to high acidity and sulphate concentrations, aluminium, cadmium,

copper, iron, lead and zinc are also present in undesirable quantitites.

The problem of pollution on such a scale has become so acute in recent

years that several attempts have been made at producing computer packages

which can predict the volume and quality of acid mine drainage likely to

be produced in any given situation (Ricca and Chow, 1974; Rae, 1977).

Caruccio et al. (1976) even use depositional palaeoenvironments as a

predictor of acid mine draina~characteristics.

Deep mines frequently discharge acid drainage into deep ground-

waters and so cause little surface pollution. After centuries of mining,

however, the vast accumulations of spoil above ground are now themselves

a threat. Indeed some of the worst cases of acid mine drainage pollution

are the result of leaching strip-mine overburden materials (Whaley et al.

1977; Anderson and Youngstrom, 1976).

6.2.2.2 Abatement

In a work concerned with the biochemical ecology of water pollution,

Dugan (1972) devotes a chapter (Chapter 11) to the role of bacteria in

controlling the rate of acid mine drainage production. This chapter ends

with a discussion of various means of treatment and abatement of acid

mine drainage. (The interested reader is also referred to a publication

by the Ohio State University, 1971).

Dugan suggests that the use of heterotrophic anaerobic bacteria of

the DesuZfovibrio type might reduce the vast amounts of toxic sulphate

generated in acid mine drainage. Their technique appears to work well in

laboratory experiments but the development of anaerobic conditions and the

supply of organic nutrients, necessary to provide the micro-organisms with

 76

the energy and carbon they require, on a scale great enough to deal with

most natural situations would be difficult to achieve.

Bloomfield (1972a) studies the possibilities of an alternative method

of treatment. In experiments which lasted five months he examines the

effects of adding lime to acid sulphate soils in order to raise the pH

and, therefore, prevent the growth of iron-oxidising bacteria. He finds

that the quantities of iron and sulphate in waters draining the soils are

reduced slightly with the addition of a quantity of CaC0 equivalent to

3
14 tonnes per hectare, although the pH only rises by 0.4 units.

Hill (1974) examines the use of both lime and limestone to treat

acid mine drainage in order to reduce acidity, remove heavy metals (which

are less soluble at high pH) and lower the concentrations of ferrous

iron and sulphate ions. He notes that in 1973 alone over 41,000 tonnes

of lime were used in this way (Gutschick, 1974). Hill quotes the reactions

involved in lime treatment as:

Ca(OH)2 + H2S04 ~ CaS0 + 2H20 (31)

4
Ca(OH)2 + FeS04 ~ Fe(OH)2 + CaS04 (32)

or 4Ca(OH)2 + 4FeS04 + 2HZO + 02 ~ 4Fe(OH)3 + 4CaS04 (33)

depending on whether sulphuric acid or ferrous sulphate are the reactants

and ferrous or ferric hydroxide the products.

Limestone treatment takes place according to the following reactions,

again depending on the reactants involved:

CaC0 + H S0 ~ CaS04 + H20 + CO2 (34)

3 2 4
or 3CaC0 + Fe2(S04)3 + 3H20 ~ 3CaS04 + 2Fe(OH)3 + 3C02 (35)
3

Most industrial treatment plants prefer the use of lime to limestone

for the following reasons:

 oil:
>.
~
.~ ~
>
.~ 8 ,.. N
M
~ 0

~='
(J

e
11"1
~
I I N
~
M
N
~
IS
-
0 U
U

-.~
Cl)
;:1

~>.
I 0 0
.~
\0 \0
N ..;t \0
'" CO >
~ '"
0 I g '>D 0'1 ~

-,..........
Cl) N 4.1
U
='
"0
!=:

-'" - -
0'1 0
I
~ I '"
CO
N
~
U

U N
N
I I
en ,..
\0

-
,..
-
C 0- >.
~
.~
--
4J
0-

+ ,..
~
,.. ...
CO
~
'-'J Qj
en
.~
>
4J
M
.... CO N
~
0'1
~ ,.. ~ ~
0 U

-
N I 0'1 Qj III
~
< N III
....~III "0
4.1 C +
Qj Qj
.. III Qj ::r:
,..
...
Qj
~ -
III
p::

... ,......
0
Qj
en
~='
III
00
0
,
-
0 0- M 0'1 =' > ~
0 ><
... -
0 M N 0
4.1
Qj 0
~
..;t
~ ~
... .~
Qj Qj
0-
... C
III
"" ~ f-t
.c: S
U
~
::I:
c,

.~~...
(I) Qj

0
~
... ::r: III e-,
4J
CO
.~ ~
~
0
~ .~
... >
C
III III
Po
Qj

+
III s M
~
0
~
~ .c =' .... Qj
~
U
><
4J 0 >. Qj
Z I ..;t
N Qj 0
='
.~
(I)

,..
en
til
C ....,
.~
~
C til
~
~
C
Qj ....s
........ !=:... !=: ~.. .~
Qj
Qj ~
Qj CO
+
N '>D 0- 0'1 \0 .~ C1I .~ .. Qj
Cl) ;:1

'" ~ .~
... ... ...0 U... f-t...0
00 11"1 '>D ..;t N CO ::I: Qj Qj d
~ ~ CO ~ ~ Q
~
Qj en
~ Qj ~ .0 Qj
d
;:l
> .~
.~ U .0
C1I ~ ....
;:l

+ !XI p:: :::> !XI ~ C1I

N CO 0 \0 ~ >

-- -- -- -- --,.. ........
III I CO '>D M 0
U ..;t N ~
~ N M ~
1
«
::I:
Po
0
N
. I I
\0
N
. '".
M

--.... -- -- -- --,..
Qj
~
~ N M ~
III
Cl)
 77

(1) The carbon dioxide produced in equations 34 and 35 buffers

the reaction and it is difficult to raise the pH sufficiently.

(2) Limestone is ineffective if the water has a high ferrous iron

content.

(3) The size. characteristics and method of application of the

limestone are critical.

(4) The treatment plant is more complex than for lime treatment.

Several different commercial schemes for limestone treatment are described

in Hill and Wilmoth (1971).

Whichever method of treatment is used the 'neutralisation' of acid mine

drainage has several shortcomings. For example the hardness of the water

may be increased, the sulphate and iron content can still remain unacceptably

high and vast quantities of waste sludge (mainly gypsum) are produced and

must be safely disposed of. Hill (op.cit.) concludes that a split treat-

ment of mine drainage by both lime and limestone may be the most cost-

effective method, and suggests the use of limestone for reducing toxicity

in surface spoil heaps (see also Deult 1974).

6.2.3 COmparison with Mam Tor Drainage Waters

Table 11 shows that waters grouped together under the all-embracing

term 'acid mine drainage' can in fact show considerable variation in

quality. The water of analysis (1) is clearly exceptionally rich in total

iron, a fact reflected by the conductivity value, while that of (2) is

extremely saline. Features common to all such drainage waters are a low

pH and high metal and sulphate concentrations.

It is interesting to note that the mean values given for Site 7 fall

somewhere in between these two extreme samples and the other two waters

included in the table.

 78

It seems perfectly justified, therefore, to consider the Mam Tor

waters as being a 'natural' example of acid mine drainage. This is the

result of fracturing of rocks by landslide movement, rather than by

mining activity, and the subsequent involvement of micro-organisms in the

process of pyrite oxidation. It is likely that the production of acid

sulphate waters in this way has been continuous since landslide initiation

and as such the time-scale involved is perhaps an order of magnitude

greater than for true mine drainage of similar composition.

Since the waters draining Mam Tor are not used for public supply and

pose no threat to wildlife or fish stocks, there is no need for remedial

measures, such as those discussed above, to be taken. It is an interesting

academic point, however, to consider how such abatement could be achieved.

The use of sulphur-reducing bacteria as described by Dugan (op.cit.)

would be precluded for the reasons given on page 75. Limestone treatment

would, presumably, also prove ineffective since the waters encounter

carbonate material in any case on their passage through the Eda1e Shales

and through road-fill; many hundreds of tons of limestone ballast have

been used and subsequently scattered over a wide area during 180 years

of road maintenance.

The most effective way of ame1eorating this acid drainage would seem

to be by natural processes which do not require man's intervention. Natural

beneficiation of acid mine drainage has been discussed at length by Biesecker

and George (1966) and Lovell and Lachman (1970). It has been noted that

simple dilution as the acid streamsmix with larger, less concentrated,

bodies of water cannot explain the observed fall in concentration of

certain ions. Interactions between the acid waters and stream sediments

(Crouse and Rose, 1976) and the effects of non-acid streams containing

significant bicarbonate (Barnes and Romberger, 1968) are probably responsible.

Chemically active stream sediments and karst waters are both present at and
 TABLE 12: Species List for Selected Water Sampling Sites

Higher Plants Bryophytes and Mosses Algae

Cares:nigra DrepanocladUs ftuitans EugZena mutabiZis

Deschampsia fZexuosa Sphagnum recurrum Zygogonium Spa
Eriophorum angustifolium Sphagnum eubeecundum
Bolcue mortis
JUJWU8 aeutifZorus
Molinia caerulea
Phragmites communis
Deschampsia j1exuosa DrepanocZadus fZuitans Euglena mutabZis
Eriophorum angustifolium Sphagnum reeurrum Zygogonium Spa
JUJWU8 actifZorus
Phragrrri tee comnunie
PotentilZa erecta
Agrostis tenuis EugZena mutabiZis
Dactylis glomerata Zygogonium Spa
Desckampsia cespitosa
Equisetum palustre
Holcue mol/l-ie
Juncus aeutifZorus
Agrostis tenuis DrepanocLadus fl,uitans ChLamydomonas applanta,
Carex nigra Polytrichum commune var. Acidophi La
Deschampsia j1exuosa Euglena mutablis
JUJWU8 effusus Hormidium Spa
Mougeotia sp
Agrostis camina Acroc~ium euspidatum Hormidium s p •
Agrostis etolonif'era A meum undu Latum Spirogyra Spa
Agrostic tenuis DicranelZa squarraso
Anthozanthum odoratum DrepanocZadus fl,uitans
Cardamine praterurie Eurhynchium praelonqum
Cerastium hotosteoides Phitonotis fontana
Cirsium palustre Polytrichum corrorrune
Desckampsia fZexuosa Rhytidiadelphus squarrosus
EpiLobium paLustre
Equisetum paZustre
Festuca ovina
Gal.ium paluetire
Gal ium saxatiZe
Soleus molUs
JUrlCUS effusus
LuzuZa campestris
Molinia caeruZea
Poa pratensis
Poa triviaUs
Ranuncu lue repene
Rumex acitosa
Sagine procumbens
Trifolium repens
TussiZago farfara
Vaccinium oxycoceus
Veronica becoabunqe
 79

in the vicinity of Mam Tor and the toxic constitutents of the waters have

all been removed by the time the streams leave the village of Castleton

(Sampling Site No. 102 was located below the confluence of waters draining

Mam Tor and a karst stream). Some consequences of mixing waters of con-

trasting compositions are discussed in Section 7.6.1.

6.3 Environmental Considerations

6.3.1 Phytosociological Responses

Lackey (1938) provides a detailed list of organisms which he found to

be living in acid ~ne drainage in Indiana and West Virginia. Unfortunately

his account does not state whether or not the species are optimally adapted

to the high acidity levels or even if they are actually growing. Species

diversity in such waters is known to be low and algae are particularly

restricted (Harrison, 1958; Steinback, 1966; Warner, 1968). Consequently,

most workers have concentrated on the lower forms of biota.

Bennett (1969) for example, considers the algal floras in seventeen

acid creeks where he finds no evidence that any chemical factors other than

acidity are limiting or controlling. The total number of algal species in

highly acidic streams in England is reported as being twenty-four (plus

two moss species) by Hargreaves et al. (1975). An interesting conclusion

of these authors is that the acid streams which have been in existence

the longest appear to have a richer flora.

A list of the species of plants found at five sites on Mam Tor is given

in Table 12.

Sites 2,3,6 and 7 are all acidic (pH 3.0-3.5) seepages and can be seen

to contain, in total, only fourteen species of higher plants, four mosses

and five algae. By contrast Site 8 (mean pH - 5.2) has twenty-five higher

plants, eight moss or bryophyte and two algal species. The greater diver-

sitv at the latter site, which can readily be accounted for by lower acidity,

is obvious. Phytosociological differences between acid sites are slight

and all contain virtually the same communities of moss and algae.
 80

Of the higher plants restricted to acid seepages one,Phragmites

communis, is particularly interesting. Although this reed is commonly

believed to grow only in neutral-pH environments this is the second site

known to the author where it thrives in water of pH <3.5. Indeed at Parys

Mountain on the Isle of Anglesey, North Wales, a large stand of Phragmites

is the dominant community in water draining abandoned copper mines where

the pH is often as low as 2.0 or less.

Of the plant species found at Mam Tor, those that are known to prefer

acid conditions are Juncus effusus, Deschampsia jtexuosa, Agrostis tenuis

and Carex nigra. Of the mosses DrepanocZadus jtuitans and PoZytrichum

commune are tolerant of highly acidic waters. The former species has

also been recorded in large numbers in an acid mine stream in Durham,

England, by Whitton (1972). The algae of EugZena, ChZamydomonas, Mougeotia

and Zygogonium species are all acidophilic and are commonly found in acid

mine drainage (Hargreaves, 1975).

High concentrations of H+ ions generally impare a plant's absorption

of calcium, magnesium and phosphate and can increase the solubility and

toxicity of aluminium, manganese and iron (Rorison, 1973). Mycorrhizal

activity and cell division in young roots are also restricted.

Sulphate is not usually considered to have much effect on plant growth

and nutrition, but at very high concentrations can prevent the uptake of

nitrate and phosphate, adversely affect osmosis and generally stunt growth.

Brock (1969) studies in detail the effects of extreme conditions on the

lower forms of aquatic life.

It would be interesting to observe whether the acidophilic species, or

indeed those not generally considered to survive in low-pH waters, at Mam

Tor have adjusted over the millenia to the toxic conditions and if the

species diversity is greater than that reported for true acid mine drainage
 81

situations. Unfortunately there was insufficient time in the present

study to conduct such an investigation and the author is not qualified

to perform the necessary experiments. From casual observation, however,

it would seem that some plants are growing in waters generally considered

unsuitable and that some species, particularly mosses, are more diverse

than reported by Hargreaves et al. (op.cit.).

Any interested reader would have the full co-operation of the author

in undertaking such work and the streamwater chemical data presented here

might prove invaluable in this event. It is thought that by using Mam

Tor as an example of what happens after prolonged exposure to acid mine

drainage possible ways of re-colonising areas devastated by pollution of

this kind might be found. The examination of plant remains incorporated

in the ochre deposits might yield evidence of past changes in plant

communities in the seepages.

Examples of studies concerned with re-populating the aquatic environ-

ment of acid mine drainage areas could not be found. In a study of the

recolonisation of black wastes from coal mining areas (very similar material

to the fragmented black shale at Mam Tor) Schramm (1966) found that low pH

was not the limiting factor. Nitrogen-fixing legumes and potentially

ectotrophic mycorrhizal species such as pines, spruces, birches, aspens,

oaks, and willows can tolerate a highly acid growing medium. High sur-

face temperatures resulting from the low albedo of black shales are far

more important.

Black shales exposed to the direct rays of the sun can reach a tempera-
0
ture in excess of 75 C at the surface; 25 mm below the surface this tempera-
o
ture can drop by as much as 18 C. Heat injury and even death both of seeds

and seedlings can result purely from insolation and it is thought that the

lack of vegetation cover on certain parts of the Mam Tor landslide (notably

the black scree .lopes of Little Mam Tor and the retrogressive units) may
 82

be partly due to elevated surface temperatures which are observed on fine

days, particularly on south and southwest facing slopes.

Several animal species can be found in the acid waters at Mam Tor

including a large population of frogs and several insects (beetles, water-

boatmen, etc.). Again it is not known if these animals are in some way

adapted to the toxic conditions or if they are barely managing to survive.

6.3.2 Biological-Chemical Interactions

The concentration of some elements in streamwater can be predicted

purely from abiotic chemical reactions of precipitation, complexing, and

so on. Levels of others may be significantly influenced by biochemical

reactions. According to Ruttner (1953), for example, "It is quite im-

possible to understand the chemistry of an aquatic biotope without taking

into consideration the causal relationships in the metabolism of its community

of organisms".

Higher plants and animals only become significant in influencing stream-

water chemistry when their numbers are large. Because of the high surface

area to volume ratio and the biochemical versatility of bacteria and micro-

scopic plants and animals these generally play the dominant role.

In a study of the relation of biological activity to water composition

Lee and Hoadley (1967) observe that the effects of such organisms are largely

indirect. The number of free-living bacteria in natural waters is shown by

these authors to be extremely low. Most are attached to particulate matter

where they "create special localised conditions, permitting solution of

mineral constituents or the metabolism of certain organic materials by

altering the pH, redox potential, or by preventing dispersal of extra-

cellular enzymes".

Changes in pU result from normal photosynthesis and can lead to precipi-

tation of chemical species which are at or near saturation. The classic

 83

example is the photosynthetic precipitation of CaC0 which is accelerated

3
by the presence of certain bacteria (Lee and Hoadley, op.cit.).

Since concentrations of iron in water in excess of a few tenths of a

milligram per litre can impart an unpleasant taste, and even stain materials

with which it comes into contact, the solution characteristics of this ele-

ment have probably been more closely studied than any other. The of ten-

controlling influence of organisms,which all require iron to build up

their body substance, on iron concentrations is well known. Perhaps the

most comprehensive review of literature on this subject is that of Oborn

(1960). While this work pays particular attention to microscopic forms

that of Oborn and Hem (1962) concentrates on the effects of the larger

aquatic organisms.

It is thought that at Site 7 at Mam Tor the filamentous green alga

Horrnidium is sufficiently abundant to influence oxygen levels during photo-

synthesis, thereby bringing about diurnal cycles in stream chemistry (Whitton,

B.A., 1979; written communication). The filaments can also be seen to act

as nuclei for the precipitation of iron oxide. Spring and summer growths

of Hormidium at this site become completely encrusted with iron by the

Autumn, and it is thought that this process may be in part responsible for

the accumulation of ochre at this and other localities (Section 7.6.3).

Sykora et al. (1972) have also noted that algae can act as nucleation sites

for precipitating ferric hydroxide, which ultimately results in the death

of the organisms.

The close association of algae and gypsum is also observed in the Mam

Tor seepages, and can be seen in Plate 20.

6.4 Conclusions

The conclusions which can be drawn from this Section, concerned

essentially with the chemical reactions taking place at Mam Tor, are summarised

as follows:
 84

Observations made in Sections 4 and 5 can now be explained in terms of

quantitative chemical equations. Pyrite oxidation is the source of

sulphate andiron, calcium and magnesium are derived from dolomite

dissolution and silicate transformations give rise to the aluminium

and potassium in the waters.

At least 99% of the acidity produced in pyrite oxidation and ochre precipi-

tation is consumed before the solutions rise at Site 7. The residual

1% of H+ ions leave the water considerably acid, at a pH of approxi-

mately 3.5. Over 20 tonnes of pyrite are consumed annually in the

generation of this acidity.

The waters produced by these reactions closely resemble so-called 'acid

mine drainage'and the involvement of chemoautotrophic iron oxidising

bacteria as catalysts in their generation is suspected.

Pronounced vegetational differences occur between acid springs and surface

drainage on the landslide. It is not known whether the organisms in

the acid seepages have become adapted to the extreme conditions.

Algae almost certainly act as nucleation sites in the precipitation

of ferric hydroxide as well as raising the oxygen status of the waters.

The next logical step in this research would be to examine the above

equations more carefully to see if they make sense, for example, in terms

of chemical thermodynamics. Firstly, however, we will consider the other

half of the descriptive story - the analysis of solid materials.

 85

7. MINERAL ANALYSIS

The chemical weathering of individual mineral grains takes place

by surface attack. Therefore. it seems logical to examine the finest

fractions within a rock sample as the effective surface area is much

greater than for coarse grains and the results of chemical alteration

will be more readily detected.

Structural changes may also occur in the clay minerals as a conse-

quence of weathering which might be particularly relevent here since an

increase in the proportion of. say. expandable clays can affect the

stability of the material as a whole. Consequently the separation and

methods of analysis of very fine clays « 0.5 ~m) will be described in

detail.

We are also interested,however. in overall compositional changes

involving the removal of certain minerals and the precipitation of others.

The identification of mineral phases and the detection of inter-sample

variations or trends are facilitated by the use of powdered whole rock

samples. as opposed to any particular size fraction. Both types of analysis -

clay mineral and whole rock - will be discussed after brief consideration of

the nature and location of the samples themselves. This Section ends with

a broader discussion of the occurrence of gypsum. jarosite and ochre.

7.1 Sampling of Solid Materials

Appendix Ci gives a description of each sample taken for X-ray analysis.

Essentially these samples can be split into three distinct sets; shales

taken from surface exposures on the landslide. shales from borehole cores,

and samples of various other surface materials. These will be discussed in

more detail below. with a further division of the surface rock samples between

the two principal sampling localities.

 86

7.1.1 'Little Mam Tor'

Since the Edale Shales contain the various mineral phases thought to

be responsible for the unusual drainage water quality, numerous samples

of these mudstones were collected for analysis. The largest exposure of

Edale Shales on the landslide is on the minor scarp face known as 'Little

Mam Tor'.

Samples were not systematically located throughout the exposure here.

Rather, materials which appeared visually distinctive were taken from

various locations marked on Plate 8, in a general 'down slope' succession.

The reason for this is that material being weathered on the face, falling

off to accumulate as scree, and eventually becoming incorporated into the

soil profile is undergoing progressive alteration. Any mineralogical

variation produced by weathering, therefore, would be most easily detected

between samples of relatively fresh shale on the scarp face and highly

broken-down shale beneath the grass cover at its base.

As noted in Section 2.2.1, the Edale Shales at outcrop here are

fractured considerably due to jointing and expansion along bedding planes.

Consequently samples are easy to collect and can be picked from the face

by hand. Occasionally a small geological hammer is necessary particularly

for breaking off pieces of the carbonate bands which are much harder than

the shales.

Samples collected in this way were placed in self-sealing polythene

bags and marked with a number representing the sample location and date.

Any relevant comments were recorded in a field notebook.

7.1.2 Retrogressive Units

Some of the highest beds in the Edale Shales succession outcrop in

the area of retrogressive units (Plate 9). Here the rocks are even more

highly fragmented and the enlarged joints and bedding planes are lined
PLATE 21: Joint filling material in the retrogressive
units.
FIGL~ 23: Soil profile in a pit dug above Site 7
on the retrogressive unit.
Letters refer to soil horizons as defined
by Cruikshank, 1972.
Fig 23 Soil Profile Above Retrogressive Units

10

20

30

40

SOcm

•
 87

with gypsum and iron oxides (Plate 21). Discrete carbonate concretions

are present in addition to continuous beds.

Samples of joint-fill material and highly weathered shale were

collected from the face of the main retrogressive unit in addition to

pieces of the concretions and carbonate bands.

A soil pit was dug on top of the main retrogressive unit. Plate 9

shows the location of this and other sampling points. Material collected

from the soil pit ranges from very fine shale fragments in the turf layer

to weathered bedrock at a depth of approximately 50 cm. Figure 23 shows

the section in this pit.

7.1.3 Other Surface Materials

A piece of the main failure surface some 5 em square, was removed

from the base of the main scarp face in order to determine the mineralogy

and to compare this with the other shale samples. In addition a large

carbonate concretion (Plate 2a) was taken from the top of the main retro-

gressive face and returned to the laboratory where different types of

surface coatings and oxide layers were carefully removed and separately

analysed (Samples A to F, Appendix Ci).

Samples of precipitates from the seepage areas, previously described

simply as 'ochre' were also taken. The ochre at Site 3 was sampled by

inserting a barrel auger with a sharp cutting tip down into the stream

bed with a corkscrew motion. The auger penetrated to a depth of about

40 em and yielded a good intact core.

At the top of this core the ochre is a purplish-red colour and is

very brittle despite having a high water content. Further down the ochre

is reddish-orange with a high organic content (the remains of Sphagnum

and PoZytriahum moss), and at 40 em depth it is yellow-brown, very hard

and has a lower organic content. Sub-samples of these three distinctive

ochre types were taken.

 88

Gypsum crystals were removed both from the carbonate concretion and

from plant stems in the seepages (Plate 20) and returned to the laboratory

in stoppered glass bottles. A milky-white precipitate from the Odin Sitch

(Plate lIb) was also collected for X-ray analysis and some of this material

was precipitated in the laboratory for comparison by mixing water from

Site 11 with karst water (see Section 7.6.1).

A sample of unweathered Edale Shale was collected from the quarry at

Hope Cement Works. Unfortunately the management of the Works would not

grant permission for more comprehensive sampling as they considered the

existing quarry workings "too dangerous". The single sample was taken

from a stock pile of recently quarried material. It is not known, there-

fore, from whereabouts in the succession it came, but it must be lower

than any beds outcropping at Mam Tor.

7.1.4 Borehole Cores

In the Spring of 1978, Derbyshire County Council sank eight boreholes

down through the landslide (see Section 3.3). The cores from these bore-

holes proved a useful source of material from depth within the landslide

and small samples were obtained, with permission from the Council, from

various depths in two of the cores. The location of these two boreholes

is shown on Plate 12.

Borehole 7 penetrated to a depth of about 20 metres and samples of

shaly material, some 'fresh' and some 'weathered' were collected at regular

intervals down to 18 metres. The material was fragmented to varying degrees

throughout this distance. It is not known for certain if the fracturing

is due to coring procedure or to inherent cracks in the rock. It is

thought likely, however, that since certain portions of the core were

relatively undisturbed that much of the fracturing was the result of

past landslide movements.

FIGURE 24: Location of samples 1n Boreholes 7 and 8.
Correlation is made on distinctive units.
 FIG 24 LOCATION OF SAMPlES
SH7 IN BOREHOLES 7 & 8

BH 8

-- -

~""'_--Top of Edale Shales;

pyrite and carbonates
below this horizon

lEft shale

~ Siltstone

D coarse micaceous sandstone Possible failure zones

~ shale With carbonate bands

rocks slightly fractured EJ scale in

rocks highly frac tured ~ °5~ metres

•
 89

Borehole 8 was sunk to a depth of 30 m and samples were collected

to 28 m. Not all the samples taken from these two boreholes were

subjected to X-ray analysis - once again the emphasis was on comparing

'fresh' and the 'weathered' material.

A sketch log of the boreholes compiled from notes made at the time of

sampling the cores is shown in Figure 24. Correlation between the bore-

holes is based on distinctive formations such as the coarse sandstone

unit which contains large flakes of white mica.

Due to the lack of pyritic, carbonate-containing shales in Borehole

7, it is thought that the Edale Shales were not penetrated. The top of

the Edale Shales is believed to be at a depth of 14 m in Borehole 8.

7.2 Particle Size Separatio~

7.2.1 Disaggregation

The shale samples described in 7.1 were to be subjected to two types

of X-ray analysis; whole-rock and clay-mineral. For this reason the

samples were divided into two parts. One part was treated as described

in Section 7.3.2 and the other was disaggregated, in order to separate

out those clay particles less than 2 ~m in diameter.

Approximately 10 g of shale was lightly broken up using a geological

hammer; for many of the weathered samples this procedure could be effected

by crumbling the rock between the fingers. This material was then placed

in clean, wide-necked polythene bottles of about 1 litre capacity and

300 ml of 'calgon' solution added. This dispersing agent, added to prevent

the clay particles from flocculating, consists of 1.5 g sodium hexameta-

phosphate per litre of distilled water.

Tightly fitting lids were then placed on the bottles which were

secured to a metal frame rotating about a central horizontal axis. The

'end-over-end' tumbling was continued until the samples had completely

broken down to sand-, silt- and clay-sized particles. Anything from

 90

between 1 to 24 hours of tumbling may be necessary, depending on the

degree of consolidation.

When the shales were completely broken down they were transferred

to sedimentation cylinders for further size separation,

7.2.2 Sedimentation

7.2.2.1 Theory

The following discussion concerning the separation of clay-, silt-

and sand-sized materials is based on that presented by Blatt, Middleton

and Murray (1972, Chapter 3).

The settling velocity of particles in solution is a function of particle

size andfluid.resistance. This resistance comprises both inertial and

viscous forces which are proportional to the kinetic energy of the fluid
2
(pu /2) and to the cross-sectional area of the particle, as follows:

2
= ~ • A (36)
2

where p is the mass density of the fluid

u the veloci ty of the fluid relative to the particle

A the cross-sectional area of the particle

Cn its 'drag coefficient'

This relationship is known as Newton's Law of Resistance.

The viscosity of the liq.uid determines the value of Cn' The cgs+
-1 i h
system unit of viscosity is the 'poise' (g cm s-1) W h1C .
can vary 1n

a given fluid depending on the temperature and salt content. A relative

measure of the total viscous forces of a fluid is the Reynolds number

which is defined as:

R
e
= (37)
 91

where d is the diameter of the particle

~ is the 'dynamic viscosity'.

It is found experimentally that the drag coefficient CD is a unique

function of the Reynolds number, for a constant particle shape.

Graf and Acaroglu (1966) have shown that for very fine-grained

spherical particles settling in water at low concentrations:

(38)

If this equation is combined with (36) above, we obtain:

which is known as Stokes' Law of resistance. Following from this, for

a particle settling in water with a constant velocity w, Stokes' Law

of settling can be written as:

(ps - p)S
w = IS].!
(40)

where (ps - p) is the density difference between the particle and

the fluid

g is the acceleration due to gravity.

This law is used as the basis of determining the size of clay and

silt particles since the value inside the square bracket has the value

of 0.892 x 104 at 200C (w measured in cm 8-

1 and d in cm). Thus:

d - J( w ) (41)
- 0.892 x 104

In practice the application of Stokes' Law is limited since clay

particles are not spherical, their density is not precisely known, and

high concentration of particles may retard settling. However, if the

clay particles are adequately dispersed, and it is remembered that the

FIGURE 25: A method for separating < 2 11mclays from
a sediment suspension using a graduated
syphon tube.
 FIG 2S A METHOD FOR SEPARATING < 2~ CLAYS

graduated
sedimentation /' syphon
cylinder; tube.

constant
temperature
vo ter tank

< 2~ motenct
in suspension
above gradua non

> 2\-1material

•
PLATE 22a: The Sedimentation Tank

PLATE 22b: Thin-section of Edale Shale.

x 5 Negative View.
 92

'equivalent diameter' (the diameter of a quartz sphere having the same

settling velocity) is being measured then sedimentation is a perfectly

acceptable technique of particle size determination.

7.2.2.2 Practice

From Stokes' Law can be calculated the time taken for particles of

a known equivalent diameter to settle a given distance. Pure samples of

clay <2 ~m in diameter are most frequently used in X-ray diffraction

studies of clay minerals. It was calculated that particles of this size

4
will settle in a 'calgon' solution at a rate of 4.0277 x 10- cm s-l.

A convenient settling time in experimental studies is approximately 16

hours; samples can be mixed up at 5 p.m. and the fine fraction drawn off

at 9 a.m. the following day. In sixteen hours 2 ~m particles will settle

a distance of 23.2 cm.

Consequently, sedimentation cylinders approximately 30 cm in length

were obtained and syphon tubes constructed so as to draw off the top 23.2

cm of liquid from inside them. The whole apparatus was enclosed in a water
o
bath kept at a constant temperature of 20 C by an electric thermostat.

The set-up is shown diagrammatically in Figure 25 and also in Plate 22a.

At the commencement of sedimentation the samples from the 'end-over-end'

tumbler were placed in the glass cylinders and made up to volume, according

to the graduation on the syphon tube, as necessary. The cylinders were

then inverted several times before leaving them to stand in the water bath.

After sixteen hours of undisturbed settling no material coarser

than 2 ~m remains in suspension above the lower graduation(Figure 25).

The supernatant liquid, containing particles of 2 ~m and smaller, is

then drawn off by opening the clamp and activating the syphon (the syphon

tubes are filled with distilled water before lowering them into the

sedimentation cylinder).
 93

Changes in clay mineralogy due to weathering are most noticeable

in the finest size fractions (greatest surface area). Therefore, it

was decided to separate out the <0.5 ~m particles from the <2 ~m cut.

The sedimentation time required for this by the above procedure is far

too long to be practicable. If the value of g in Equation 40 can be

increased, by using a centrifuge, the time is reduced considerably.

Therefore, it was calculated that a spin-time of seven minutes and

two seconds at 3000 rpm (for a 100 ml-capacity centrifuge head) would

effectively bring down from suspension all particles coarser than 0.5 ~m.

The supernatant fluid, containing clay particles of <0.5 ~m, was then

further centrifuged at 4000 rpm for several hours until the liquid in

the centrifuge tubes was clear and the clay particles had settled to

the bottom and could be removed.

The subsequent treatment and analysis of these <2 ~m and <0.5 ~m

sub-samples is discussed in Sections 7.3.2 and 7.4.

7.3 X-Ray Diffraction

7.3.1 Principles

X-rays consist of electromagnetic radiation of very short wavelength.

-8 -11
This typically ranges from 2.S x 10 m ('soft' X-rays) to about Ix 10 m

('hard' X-rays). Visible light has a longer wavelength, and shorter over-

all range, in the order of 10-7 m. The very short wavelength of X-rays

means that they can penetrate solid matter.

If monochromatic X-rays are incident upon a crystal surface at some

angle e, they will penetrate the solid to lower layers in the crystal

lattice. They will penetrate further for higher angles of incidence

and vice versa. If monochromatic, or 'coherent' X-rays emerge once again

from the crystal, having been reflected from the lattice, then the path

differences between the incident and reflected beams must be an integral

 94

number of wavelengths. Otherwise destructive interference will lead

to extinction in much the same way as happens to polarised light in

transmitted light microscopy.

Bragg (1913) showed that the wavelength A, and the angle of

refraction 8 are related to the crystal lattice spacing, d, in the

following way:

(42)
nA = 2d sin 8

The X-ray source in modern X-ray diffractometers is held stationary

and a device known as a 'goniometer' rotates both the sample and the X-ray

analyser, a'proportional counter' (similar in principle to a Geiger-

"
Muller counter), through various angles of incidence with the beam. Since

the angle of reflection (8) is equal to the angle of incidence (8) the

beam of X-rays is diverted through an angle 2s in total. The counter,

therefore, moves with twice the angular velocity of the sample.

The proportional counter records the intensity of the radiation it

receives as a certain number of 'counts-per-second'. This information

is sent to a chart recorder which gives a continuous visual reading of

intensity. Peaks of X-ray intensity are observed, above a general low

background level, when crystals of the same mineral in the sample are

reflecting X-rays simultaneously and through the same angle. The diffe-
o
rent positions of these peaks on the chart, which has 2e along the

horizontal axis and intensity along the vertical, indicate the presence

of different mineral phases. Thus, if the goniometer is driven over

a sufficient range of 028 to cover the possible peak positions of the

minerals likely to be present, the overall mineralogy of the sample

(assuming that its components are crystalline and not amorphous) can

be determined.
 TABLE 13: Operating Conditions for the X-ray
Diffractometer

Tube Type Cu
Voltage and Current 35 kV, 40 rnA

Filter (secondary) Ni

Pulse Height Analyser:

Lower Level 180 divisions
Window 220 divisions x 10-100%
Gain 64

Range 4 X 103 c.p.s.

Time Constant 1 second
Chart Speed 2 cm per minute

Wavelength of C~a Radiation o -10

1.54 A (1.54xlO m)
 95

By altering the type or intensity of X-rays, the speed of rotation

of the goniometer, and by using various methods of signal enhancement

(such as filtration, 'pulse height analysis' and integration) satis-

factory operating conditions can be obtained for most experimental

situations. The standard operating conditions for the Philips

diffractometer used in this study are shown in Table 13.

7.3.2 Mineral Identification in Whole-Rock Powder Samples

To analyse a whole-rock sample by XRD it is necessary to crush the

rock to a very fine powder. This is most easily done in a mechanical

'Tema' grinder, after the rock has initially been broken up using a

hammer or 'jaw' crusher.

After 'Tema' grinding the resulting powder is homogenised to produce

a representative sample. Very little powder is needed to form the back-

filled mount required for this type of analysis, although a considerable

amount can be crushed at one time.

About 109 of the powder is ground still finer by hand in an agate

mortar. This material is then pressed into a hole in a 1 mm-thick alu-

minium plate, which is covered on one side by a microscope slide cover

glass. When the hole is full, a second solid plate is clipped on top

and the mount turned over. On removing the cover slip the surface of

the powder is found to be smooth and flat, although the mineral grains

have random orientation. This powder mount is then placed in the

sample chamber of the goniometer and analysed as described in the pre-

vious section.

Such a method of sample preparation gives good peak intensities on

the chart because the powder is thick and many thousands of crystals will

be reflecting X-rays at the same time. With such randomly-oriented grains,

however, the resolution of separate peaks may be poor.

 96

0
It is usual to scan such a sample from about 40 to 45 29. This

range covers d-spacings from 22X to 2X; most common rock-forming and

secondary minerals have lattice spacings in this range and will give

one or more diffraction peaks. Each mineral has a number of possible

peak positions corresponding to different crystal orientations.

A scanning rate of 1026 per minute is usually sufficient to produce

an interpretable diffraction trace. If carbonate rocks are being analysed,

however, it may be desirable to scan the specimen over a more limited 2e

range, and at a slower rate to obtain better resolution of the peaks. In

this way subtle variations in composition between, say, dolomite, ferroan

dolomite and ankerite (which have peaks very close together on a standard

trace) may be discerned.

In this study most of the samples from 'Little Mam Tor', the retro-

gressive units and Borehole 7 were X-rayed using powder mounts. The samples
o 0 0 •
were scanned from 4 to 45 26 at 1 29 per m1nute. In addition, if carbonate
0
was found to be present, the samples were scanned from 26° to 33 26 at

0.25°29 per minute. Appendix Ci gives further details of these and any

other analyses.

7.3.3 Identification and Estimation of Clay Minerals

There is considerable debate in the literature of sedimentary petro-

logy as to the best way of preparing clay mineral samples for analysis by

X-ray diffraction. The aim is to produce a highly oriented specimen so

that only basal reflections are produced. Most authors agree that the use

of<2 um (or smaller )size fractions give the best results.

Perhaps the most commonly used preparation technique is that of smear

mounting on a glass plate as described by Gibbs (1965). Kinter and Diamond

(1956) however, believe that air drying on glass gives little better than

random orientation and they propose a method involving suction of a clay-

 97

water suspension through porous ceramic plates. This technique has

the advantage that any dispersing agent or other chemical additive likely

to affect the results can be washed off simply by sucking through dis-

tilled water. Such a technique is preferable, therefore, in cation

replacement studies.

In a discussion of the various available mounting techniques, Gibbs

(1968) recommends the use of only three; Kinter and Diamond's suction

method, smear mounting on glass, or powder press mounting (as for whole

rock analysis). The actual quantity of clay used in these techniques

can also affect the results (Stokke and Carson, 1973).

Schultz (1975) believes that an overall accuracy in quantitative

estimation of the minerals present better than! 10% can be achieved with

several techniques provided that sampling and preparation are standardised,

the grinding procedure is controlled and uniform and that the results are

consistently interpreted.
As a result of Cubitt's (1975) work it was decided in this study

to use the glass smear mount of Gibbs (1965), which is also the most rapidly

prepared type of sample. Cubitt concludes that: "Detailed statistical

tests of competing sample preparation methods show that the commonly

used smear technique is superior for distinguishing between shales on

the basis of quantitative assessment of mineral content".

Since maximum structural information is required to determine subtle

changes on weathering two size fractions - the < 211 and < 0.5 II - were

separated as described in Section 7.2. Each was allowed to almost completely

dry before being lightly ground in an agate mortar to break up any aggre-

gate which had formed. A very small amount of distilled water was then

added to the clay to form a paste. If too much water is added at this

stage the particles may actually go into suspension and, on drying, the

very fine clay flakes will take longer to settle onto the plate. As a
 98

result of adding too much water Gibbs (1965) observed enrichment of

fine-grained montmorillonite in the surface layer of the mount.

After the paste has been thinly spread onto the glass slide using

a small spatula the sample is quickly dried by passing a slight air

stream over its surface. If this stream is too rapid, material may be

built up more thickly at the edges of the mount. If the mount is success-

fully prepared a highly reflective and uniform surface, indicating good

orientation, is achieved.

Basal reflections from the common clay minerals produce the most
o 0
intense diffraction peaks in the range 4 to about 30 ze. This is not

to say, however, that each peak is at a distinct position on the trace.

Rather, some peaks occur in the same place as, or very close to, those

of other minerals. For example the [OOZ] chlorite peak and the [OO~

kaolinite peak both occur at about lZ.4° Z8 (7.lR d-spacing). Further-

0
more a fairly broad peak is often obtained at about 8.9 ze (lOR d-

spacing) which comprises both a [001] illite peak and a reflection from

expandable clay material, often referred to as smectite, with a basal

d-spacing anywhere between loR and lzR.

To distinguish between these peaks and to determine whether in fact

any expandable clays are present, a number of tests have been devised. The

first of these involves solvating the smear-mounted sample with a molecule

larger than the inter-layer cations, typically ethylene glycol. Any expand-

able clay minerals will swell in a saturated atmosphere of glycol as the

molecule is pushed into inter-layer sites, expanding the crystal lattice

from about loR to anything between lzR and l4R. Keeping the sample in
0
glycol for about twenty minutes at 60 C is usually sufficient to produce

maximum swelling.

If the sample is subsequently heated the glycol ~ driven off and

expandable clays dehydrate, collapsing back to the fundamental structure

 99

with a loR d-spacing. The difference in area between the glycolated

and heated loR peaks is commonly used as a quantitative measure of

the relative proportions of expandable and illitic clays.

According to Austin and Leininger (1976): "The optimum temperature

of collapse of illite-smectite mixed-layer clay minerals to loR ••••••

o
is about 375 C, and the maximum collapse of layers is complete after

thirty minutes". They also note that maximum peak intensity is obtained

if the sample is scanned when still at a temperature of over 100 o C, so that

even partial rehydration is prevented. Samples at such high temperatures

are not to be recommended for use in standard X-ray diffractometers,

however, and somewhat of a compromise was achieved in this study. The

o
Mam Tor samples were heated to 550 C for one hour and then cooled rapidly

in a dessicator before being quickly transferred to the sample chamber

of the goniometer.

An intermediate heating to 2500C for one hour (thought to induce

only partial collapse) was also conducted on certain samples. The relative

merits of the two techniques together with a full description of results

obtained will be discussed in Section 7.4.2.

o 0
The <2 um and < 0.5 um smear mounts were scanned from 4 to 30

28 and glycolated and heated specimens from 40 to about 20028. Appendix

Ci gives details of those samples treated in this way.

7.3.4 Other Materials

When only very small quantities of material are available for analysis

it is not possible to produce a powder press mount. Although a highly

oriented specimen is not necessarily desirable in many cases the only

realistic alternative is to produce a smear mount (which in any case is

far less time consuming).

FIGURE 26: Examples of typical powder mount diffraction
traces for a carbonate (a) and mudstone (b).
Samples MT2c and MT3, respectively.
a - ankerite; c - calcite; f - feldspar;
g - gypsum; i-illite; j - jarosite; k -
kaolinite; p - pyrite; q - quartz; s -
siderite.
ex
w
Cl

~ '"
w
U
LL «
0 ex
~

'"
w
_J
o,
Z
0
J: ~
« u
x « a::
w LL
LL
.0
N
5
C) 0
LL N

- N
N

~
N

-0
N
0-
co
N

0
tv'!

N
tv'!

.....
..
Qj
a c
c 0
0
D \It N
l- "'0 ~
a ::l
u J: ~
a .0 ~

•
 100

For the analysis of materials described in Section 7.1.3, particu-

larly gypsum from the plant stems and surface deposits on the concretion,

only small amounts were obtainable. These were ground to a paste with

water and smeared onto a glass plate, as for oriented clay specimens,

but without the preceding particle size separation.

o
These smears were subsequently scanned from 4 to anything up to

50029 so that the presence of virtually any crystalline substance could

be detected. The low intensity of peaks noted in many samples may be

due to the fact that random orientation was not used. An alternative

explanation for small peaks may be that the normally low background count

rate was enhanced in many cases by the presence of iron-bearing minerals;

CU - radiatianis known' to produce this effect. The influence of

Ka
fluorescence was kept to a minimum by more precise signal discrimination,

achieved by using an optimum combination of integration time, full-scale

deflection and 'window' settings.

Results of these XRD investigations are considered in the following

Sections, and shown in tabulated form by Appendix Cii.

7.4 XRD Results

7.4.1 Whole Rock

A typical XRD trace of a powder mount specimen is reproduced in

Figure 26. Samples prepared in this way generally show peaks which are

easily interpreted. The trace in Figure 26b, of sample MT3, is given as

an example but the following discussion concerns the powder-mount samples

1n general.
The most noticeable features of the shale samples is that the dominant

mineral present is quartz. This may be surprising since the Edale Shales,

in particular, are very fine grained mudrocks which only occasionally

contain more silty horizons. It is assumed, therefore, that most of this

 101

quartz is of very small particle size. Alternatively, one or two

fairly coarse detrital quartz grains, when ground to a fine powder

would appear to be a major constituent of the rock.

The main quartz peak is at 3.35~ with much smaller peaks occurring

at 4.2IR, 2.46R, 2.29R, 2.24R and 2.13R.

Kaolinite and illite are perhaps the two most abundant minerals

after quartz. The main kaolinite peaks are at about 7.l5~ (001) and

3.57~ (002). Illite peaks are at 10~ (002), sR (004), 4.4SR and 3.32~.

The latter peak coincides with the main quartz peak, but is usually much

weaker, producing a bump on the high 2e angle-side of the quartz peak.

The only other clay mineral detected using powder mounts, in this

case, is chlorite at about l4R.

In addition to these sharp peaks formed by the common silicate

minerals certain of the Mam Tor samples (particularly borehole samples)

show less pronounced X-ray peaks characteristic of the mineral goethite.

Also the hydrous sulphate mineral jarosite (KFe2(S04)2(OH)6) was present

mainly in scree and other weathered surface material. Gypsum, which was

noted as being a major constituent in weathered surface exposures, was

also present in many of the samples containing jarosite.

Pyrite was not detected in samples from Borehole 7. This fact,

combined with the higher quartz content of these samples, suggests that

this borehole does not penetrate to sufficient depth to encounter the

Edale Shales. Pyrite was present in samples of Eda1e Shale from surface

exposures.

Feldspar was present in trace amounts in several instances.

In addition to those peaks accounted for by the minerals described here, one

which appeared on every powder mount trace could not be ascribed to any

mineral considered in standard X-ray mineralogy texts{see Brown, 1961;

Figure 27: The effects of glycolating and heating a <0.5 ~m
sample (Sample BR89)

x air dried
Y glycolated at 60°C for 20 minutes
Z heated to 550°C for one hour

C - Chlorite
I - Illite
J - Jarosite
K - Kaolinite
 U"\
·0
o
.-

N
,.."

,.."

co
.- r-- -0
r--: 0 .-
0 ~
.-- .--

y
~ 0
0
e--
.-- 0
.--
~ ~

-
0
rn
-0
Lf)

,.."
-
0
0
Lf)
W

1.1"\
C!
~
rri .--
Wi """'\
"""'\

(T"I 0-
CO
r- '0 -0
..j LIl LIl

30 26 22 18 14 10 6 ·29
.,
I

•
Figure 28: The effects of glycolating and heating a <0.5 ~m
sample (Sample SP2)

Lower Line air dried

Middle Line glycolated at 600C for 20 minutes
Upper Line heated to 5500C for one hour
 o
..-

co ..j'
eX)
..- ..-
0
~
Ln
r--
..j'
Ln 01 ..-
fVl 0 e-
tTl
rri

eX) eX)
..- N ~
0 0 ..- 0-

f""- Lf'l
,.,...~
trI trI
Lf'l
,.,; O· 0-.
...0
0
tritri Lf'l L.Il

30 26 22 18 14 10 6

•
 102

Brindley, 1955; Chao, 1969). This peak consistently appears at 44.64°28

corresponding to a d-spacing of 2.03R. The only mineral giving a peak at

this position is elemental iron (ASTM Powder Diffraction Data File, Card

No. 6-0696, according to Chao, 1969), but its presence in weathered sedi-

ments of this type is impossible.

The carbonates sampled are found to be of two distinct types. Those

from Little Mam Tor are composed almost entirely of ankerite (Ca(Mg,Fe)

(C0 )2)' whereas both bands and concretions from the retrogressive units
3
contain in addition a large proportion of dolomite (CaMg(C03)2)·

Calcite, siderite and pyrite are present in all the carbonates

analysed. Quartz, feldspar, and clay minerals are also present in trace

amounts as is gypsum, presumably from surface coatings. A typical carbonate

X-ray trace is shown in Figure 26a.

7.4.2 Clay Minerals

Figures 27 and 28 illustrate the effects of glycolating and heating

a <0.5 ~m smear mount sample.

After solvating the loR peak diminishes as expandable clays absorb

the glycol and swell to give a broad peak at between 12 and l4R. The

kaolinite and chlorite peaks are unaffected by this procedure.

Heating to 550°C restores the loR peak to its former proportions

and in some cases it is even intensified as the expandable clay portion

collapses back to loR (Figure 27). Kaolinite and jarosite are destroyed

by the heating process and their peaks almost completely disappear.

Chlorite is again unaffected.

o
An intermediate heating to 250 C for one hour produces only a partial

collapse in the expandable clays and the kaolinite peak is hardly affected.

No additional information or degree of discrimination is obtained.

 TABLE 14: Peak Areas and Proportions of Clay Minerals in the
< 0.5 um Samples

A. Areas (cm2)

Kc{ 01 ill i ~e Illite Chlorite Expandab le Clay

Sample Total lO~)
(air dried) (heated) (air dried) (air dried loR ~ glyco1ated

BR715 22.43 18.38 4.74 45.55 8.55

BR 86 3.82 3.93 0.00 7.75 2.20
BR 88 11.56 15.95 2.20 29.74 7.98
BR 89 19.54 44.86 5.20 69.60 14.80
G2 7.75 15.72 4.74 28.21 11.56
SP1 8.79 12.01 9.48 30.28 8.90
SP2 5.78 10.98 6.13 22.89 7.63
SP3 4.28 11.00 6.36 21.64 7.28

B. Percentages

Sample Kao1inite% Illite % Chlorite % *Expandab le Clay %

BR715 49 40 11 47
BR 86 49 51 0 56
BR 88 39 54 7 50
BR 89 28 64 8 30
G2 27 56 17 74
SP1 29 40 31 74
SP2 25 48 27 69
SP3 20 51 29 66

C. Ratios

Sample Kaolinite: Illite Ch lori te: Illi te Expandable: Illite Expandable:Kaolinite

BR715 1.22 0.26 0.47 0.38

BR 86 0.97 0.00 0.56 0.58
BR 88 0.72 0.14 0.50 0.69
BR 89 0.43 0.12 0.33 0.76
G2 0.49 0.30 0.74 1.49
SP1 0.73 0.79 0.74 1.01
SP2 0.53 0.56 0.70 1.32
SP3 0.39 0.58 0.67 1. 70

*Percent of loR peak, not of total clay

(see Appendix Ci for location of samples)

 103

Most workers involved with clay mineral analysis use the <2 ~m size

fraction, presumably because it is more readily obtained. It was found

during the course of this study, however, that the coarser material pro-

duced inferior diffraction traces to the < 0.5 ~m smear mounts. Although

the background level is reduced for the < 2 um samples so too are peak

intensities. Also, additional minerals are present in the coarser fraction

and quartz and gypsum peaks - which are absent from the < 0.5 ~m material -

interfere with interpretation of the trace.

Since the finest particles will be more at equilibrium with the waters

which contact them it was decided here to make semi-quantitative estimates

of the proportions of various clay minerals on the < 0.5 um samples only.

It must be remembered that this is no indication of the proportion of

minerals in the sample as a whole since the finest fraction is unrepresent-

ative.
Pierce and Siegel (1969) studied several methods of quantitative analysis

based on either peak-height or peak-area measurements (although they also

used only the < 2 um fraction). Their results are inconclusive and each

technique yielded substantially different data. It appears, in fact,

that anyone procedure, if consistently applied, can be used to obtain

an indication (but no more) of the varying proportions of clay minerals

between samples.
In this study it was decided that peak areas would be.used and that the

total clay content would be taken as the sum of the 7.16R kaolinite (air-

dried trace), the loR (heated trace) and the 14R chlorite peaks. The areas

of each clay mineral were then calculated as a percentage of this total

for each sample and the results obtained are shown in Table 14.

The peak areas were measured by cutting out the relevant portions of

the diffraction trace - which had first been transferred onto draughting

film - and comparing the weights of these pieces to the weight of a known
 104

area of film. This technique is considered more accurate and is more

rapidly performed than any other method of area measurement of complex

shapes.

In addition, by superimposing the air-dried and glycolated traces

the difference in the loR peaks, resulting from a shift in the position

of the expandable clay peak, could be measured directly. This area was

then calculated as a percentage of the loR peak (heated trace) in order

to give an estimate of the proportion of expandable clay present in each

sample. These percentages are given in Table l4B, Column 5.

In practice it was found to be extremely difficult to rapidly cool

o
a smear mounted sample from 550 C to room temperature; many samples frac-

tured due to contraction and could not be placed in the diffractometer.

Consequently full analyses (air-dried, glycolated and successfully heated

samples) were only performed on the eight samples for which results are

given in Table 14.

These results show that there is a much greater proportion of chlorite

and expandable clays in the surface weathered material and soil samples (G2,

SPl-3) than in the shale from the borehole.

Assuming the mineralogy of both surface and deep material was originally

similar the above observations can be explained in one of two ways. Firstly,

the chlorite and expandable clays of the near-surface samples could be

., formed, by weathering, from both kaolinite and illite. Alternatively,

kaolinite might be selectively leached out and the weathering products

form entirely from illite. Although in most situations kaolinite is not

preferentially removed this possibility must not be ruled out in solutions

such as these where the high cation content might favour the persistence

of dioctahedral clays.

Within the soil profile samples SP3+SP2+SPl represent a weathering

sequence, whereby illite appears to be transformed to expandable clays and

1
 105

the proportion of kaolinite also increases.

The most obvious conclusion that can be drawn from the information of

Table 14, is that the material at depth is very unlike the surface shales.

If weathering underground by percolating solutions containing inorganic

acids produces the same results as surface weathering by organic soil acids
('WhlCh '()'
Curtls 1975 belleves i
to be the case - H +, Lon content a 1one beelng

the important factor) then none of the borehole samples appear to be highly

weathered. It might be expected that failuIBsurfaces are, however,

weathered due to the action of percolating solutions, in which case

neither borehole contacted a failure surface. Alternatively, the irregular

sampling might have failed to include material from such a failure zone,

although the most fragmented shale was taken.

It is apparent that a major sampling programme and more detailed analysis

of the clay minerals are needed in order to determine whether any of the bore-

holes encountered failure surfaces. The problem concerning chlorite which,

although it is a less stable mineral than illite or kaolinite, increases

in proportion on weathering also remains for the time being unresolved.

One observation which does make sense, however, is that the proportion

of expandable clays increases on weathering. This is important in an engin-

eering sense since clay swelling is frequently responsible for failure in

rocks containing a high proportion of expandable clays (Nakano, 1967;

Mielentz and King, 1955).

Chemical analyses earlier showed that the contribution of solutes from

clay mineral transformations is small and it is thOUght that the overall

degree of alteration on weathering is slight, except in the soil layers.

This might be expected since the finest detrital clayshave already under-

gone one cycle of weatheringand.transport prior to sedimentation, and should

be reasonably stable in a weathering environment.

 106

Accurate quantitative assessments of the amounts of various clay minerals

present in a given sample are impossible to obtain. In the first place the

proportion of<0.5 ~m clay which the sample contains is not known. Also

peak intensities depend not only on the amount of each clay present but also

on the degree of orientation, which is impossible to reproduce exactly

in consecutive samples. In addition several subjective procedures are

involved, such as the construction of a 'best-fit' background line and the

terminating of peaks at abitrary 29 values for the purpose of area measure-

ment.

Despite these limitations the techniques outlined above do give an

indication of inter-sample variations in the relative proportions of clay

minerals.

7.4.3 Other Materials

7.4.3.1 Carbonate Coatings

Smear mounts of the various carbonate coatings described in Appendix

Ci generally produced excellent diffraction traces.

Jarosite and goethite appear to be responsible for the orange and

yellow-brown colouration in the weathered skin (Samples B, C, D), while

.
• the soft white efflorescences are composed entirely of gypsum (Samples A,F) •

The fibrous, white, crack-filling material from the concretion (Sample E)

is found to be calcite. The similar-looking encrustation on plant stems

in the seepages (Sample H) is confirmed as being gypsum.

7.4.3.2. 'Allophane'

, Water at Site lIon the main unit, having been supplied from Sites 7

and 8, flows through a drain beneath the upper section of the road before

emptying into the Odin Sitch (see Figure 5 and Plate 12). Here the water

meets a stream draining the limestone area to the south of the landslide

and a spontaneous chemical reaction takes place.

 107

From this point in the Odin Sitch downstream to Sites 13 and 1, and

even beyond the toe of the landslide, a milky white precipitate lines the

banks and bed of the stream (Plate llb). In an accessible side-branch of

the Odin Rake which lies directly beneath the Odin Sitch some water permeates

the overlying limestone and the white precipitate is formed on the ceiling

and walls of the worked-out vein.

Some of this material was collected and analysed by Wilkinson (1950).

Although the author notes that the precipitate is "x-ray amorphous" he

identifies it as being the mineral allophane (an aluminosilica gel) which

is defined by Grim (1968) as the non-crystalline part of a clay that is

soluble in hydrochloric acid.

Some of the precipitate was analysed in the present study and in

addition a quantity was synthesised by mixing various proportions of

water from Site 11 and from the limestone stream. The natural, stream

precipitated, material indeed appears to be amorphous and no peaks are

observed on the X-ray trace.

The laboratory precipitated material was allowed to dry completely

before being ground and X-rayed in the normal way. This sample (Sample G)

gave very strong gypsum peaks and no others. It would appear, therefore,

that if the precipitate is given time to dehydrate it does become crystalline

and is composed principally of gypsum rather than allophane.

7.4.3.3 Ochre

Samples of the ochre core described in Section 7.1.3 show a systematic

variation in composition, as well as in appearance, with depth below the

surface.

Brittle ochre at the top of the core is completely X-ray amorphous

(Sample I) but gives a very high background reading suggesting fluorescence

due to the presence of iron (see Section 7.3.4). Further down the core,
Figure 29: A comparison of four goethite samples
(letters refer to samples of Appendix C)
 - - c
0
••
-
OJ E
u u OJ
0 C> E L-
t.. .- U u
:J I C> C
VI LJ"I ~ 0
u

- ~ ~ U
CD
N
•
....t'

N
.-

o
N

Q)
N

'{ 69'Z ---

y S+J',Z _.--

)
3Jrnn03S Hld30
•
 108

although fluorescence still occurs, distinct peaks begin to emerge above

the background in the same position as those in Sample C (goethite).

Figure 29 shows this apparent sequence whereby material becomes pro-

gressively more crystalline with depth and, therefore, time.

The tables of Appendix Ciii show that the identification of selected

minerals is consistent with published work (after Brown, 1961). The rela-

tive intensities ~ peaks are often different due to the fact that Brown

uses powder press mounts and the samples from Mam Tor are mostly oriented

smear mounts.

7.5 Thin Sections

The petrographic microscope is of limited use in the study of very

fine grained rocks, and thin sections are often difficult to make since

mudrocks break up very easily when cut using a diamond-edged wheel. Two

thin sections of Edale Shales were prepared, however, in order to compare

the fabric and texture of 'fresh' and 'weathered' samples.

'Fresh' Edale Shale was collected from Hope Cement Works - the same

samp~e as was used for XRD analysis (HCWl). Plate 2b is a x 20 enlarge-

ment of part of this thin section viewed in plane polarised light. The

rock is extremely fine grained and the identification of individual crystals

is, therefore, virtually impossible. Some larger-than-average quartz grains

and some sericitic mica can be seen. Most of the dark colouration is thought

to be due to the presence of carbonaceous material, the relative abundance

of which distinguishes between the dark and light bands.

A few larger, sUbhedral grains of an opaque mineral can also be seen

in Plate 2b. Observations in reflected light suggest that these grains,

approximately 0.25 mm across, are pyrite.

The 'weathered' section of Edale Shale taken from the retrogressive

unit above Site 7, was prepared by first impregnating the rock with 'Araldite'
 109

resin so that it could withstand being cut and ground without disintegrating.

Similar dark, fine-grained material as in Sample HeWl makes up the bulk

of this shale,but instead of a sharp transition from dark to light laminae

highly crystalline material can be seen to clearly separate the two (Plate

3b). This mineral is identified as gypsum.

Inclusions of matrix material within these crystals give a sense of

the gypsum forming between laminae and forcing the shale apart. Where

larger cracks have opened well-formed euhedral crystals of gypsum, showing

good cleavage, have developed (Plate 4a).

Plate 22b is a x 5 enlargement of the resin-impregnated section of

shale,made by placing the thin section in a photographic enlarger. Although

this results in a negative image the whole section can be studied at lower

magnification than is possible with a microscope.

It is interesting to note that the gypsum (dark areas) not only forms

along the (horizontal) bedding and laminations but also in cracks which

intersect the bedding at between 400 and 600• The presence of cracks on

a much larger scale, but at a similar angle to bedding, were noted in

exposure in the field (Section 2.2.1).

7.6 The Mechanism of Formation of Three Precipitate Minerals

7.6.1 Gypsum

Saas et al. (1965) report on the weathering of pyritic concretions

in an arid area of southern Israel. Their description of the appearance

of the concretions could be applied, without modification, to the weathered

carbonate bands of the Edale Shales. According to these authors "the

outer part of the concretions is usually composed of red to deep violet

hematite and dull-brown goethite. Sometimes a coating of gypsum scales

is found on the outside". (see page19 for a description of the Mam

Tor carbonates).
 110

Saas et al. believe the gypsum is formed by acid (resulting from

pyrite oxidation) attack on calcite in the country rocks. Where calcite

is scarce the acid attacks clay minerals, leaching outfue alkalis and

alumina, to form jarosite and natrojarosite. Other alteration products

are listed as alunite, baryte, celestine, goethite, hematite, sideronatrite

and sulphur.

In more humid environments the presence of gypsum in surface material

is generally thought to be only temporary and possibly even seasonal as a

result of its high solubility (van Breemen, 1972). Where it is present

in large amounts it is considered evidence that leaching is not taking

place (Spears et al., 1971).

Smith and Drever (1976) and Vora and Raymahashay (1978) discuss the

formation of gypsum in joints, similar to the situation at Mam Tor. They

conclude that authigenic gypsum can form from undersaturated solutions due

to "non-equilibrium evaporation" during dry periods. During winter rain-

fall the gypsum dissolves once again.

Gypsum crystals are present in the Edale Shales at Mam Tor, however,

throughout the year; although a greater number of very fine rosettes do

appear during warm summer weather. Downward leaching by rainfall must

be considerable here due to the porous nature of this highly fractured

material yet the gypsum still persists. It is thought that this may be

in part due to a coating of less soluble material such as jarosite on the

crystal surfaces, which also accounts for their yellow-brown colouration.

The precipitation of gypsum on the plants in the seepages is definitely

related to dry weather evaporation and the crystals disappear even after

moderate rainfall.

X-ray analysis also showed the white precipitate in the Odin Sitch

to be composed of gypsum (page 107). This material is thought to be similar

both in character and origin to the sludge formed during the lime and lime-

stone treatment of acid mine drainage (see Section 6.2.2).

 .-..
u
0
."
N
+J
et!
..-l
'13
c.J

tn
-3
0 0 0 0 0 0
:>- ." N M \0 ~
~ as M 00
+J
.,.j

.,.j
:>
N
."
""
..-l ..-l ..-l ..-l

+J
c.J
::l
't:I
c:
0
U

::cc,
M
M
. . ~"". ". ~~. .
."
M
,0
M
"

c:
0
J..I
C1l
J..I
C1l
+J
et!
)
C1l
J..I
C1l
J..I
--
en
J..I
::l
0
en
J..I
::l
0
.,.j +J ::l ::l ..c:: ..c::
+J CIS ..c:: +J +J
~ ) c.J ~ ~ N N
.,.j
J..I ..-l
+J
.,.j
.,.j
13
.g " "
c.J ..-l tn J..I J..I
en 0 0 C1l C1l

0
C1l C1l
+J
.,.j
c:
.,.j
"C 0
.. ..
." N

0
~
+J

CIS
+J
~
et!
tn 0 ." 00 '-' '-'

C1l
..-l

et!
~ < ~ U 0 U 0
tn
 111

The components necessary to form gypsum from solution - Ca 2+ and S042-

ions - are both contained in the waters emerging at Site 7 and flowing

via Site 11 under the road to the Odin Sitch. No additional calcium is

required, although a small amount is contained, in the water of the

Odin Sitch.

In order to investigate the mechanism causing the rapid precipitation

of gypsum where these streams mix,two 20-litre samples of the waters

were collected and removed to the laboratory. Here they were mixed in

5-litre polythene beakers in 50:50 and 80:20 ratios of Site 11:Odin

Sitch water. Changes in pH and electrical conductivity were monitored

and the results are shown in Table 15.

Af ter three days there was 1ittle change in pH and conduc tivi ty over

the original change on mixing. Very little precipitate had formed in

the 80:20 mix, whereas a considerable amount of orange 'gel' had

accumulated in the bottom of the other beaker (50:50 mix). This precipi-

tate was allowed to dry, after pouring off the supernatant liquid, before

being subjected to XRD analysis as described in Section 7.4.3.2. It

would appear from these simple experiments that a rise in pH from about

3.3 to 4.4 is needed to cause the gypsum (probably an impure form containing

iron) to precipitate~.

7.6.2 Jarosite

Another mineral which is known to precipitate from certain solutions

after a change in pH is the hydrous sulphate of iron and potassium, jarosite.

It has been suggested that this mineral is one end-member of a solid

• . d . • 3+ •.
solutlon serles an lncreaslng Al substltutlon tends to the other end-

member, alunite (KA13(S04)2(OH}6). Intermediate minerals, however, are

rare (Brophy et al. 1962).

 112

Much of the work on the alteration of pyrite has been done in the

USSR (see: Srebrodol'skiy, 1977; Yakhontova et al. 1979) where it is

noted that jarosite is one of the most common oxidation products

(Khlybov, 1976).

According to Nordstrom and Dagenhart (1978) there are two distinct

types of pyrite weathering products. These are efflorescent crusts of

sulphate minerals on the unoxidised sulphide and those minerals which

precipitate directly from the drainage solutions. These authors suggest

the following alteration sequence:

PYRITE+MELANTERITE+ROZENITE(+ SZOLMONOKITE)+COPIAPITE+COQUIMBITE

The final stage of this sequence "Is the formation of jarosite and hydrated

iron oxides which have only been found as direct precipitates from water

solutions".

In order to characterise the formation of such minerals, Bladh (1978)

simulates the weathering of sulphide-bearing rocks in a rather unusual

way - using differential equations describing equilibrium and non-equilibrium

mass transfer between aqueous solutions and minerals. His calculations

predict the formation of jarosite or alunite with the solution of only

-8
1.8xlO g pyrite per litre of water being enough to precipitate geothite.
Brown (1971) considers the stability relationships between the three

minerals; geothite, jarosite and pyrite. It is found that the dominating

+ 2-
influence is pH although the activities of K and 804 must also be high

for jarosite, rather than geothite, to form (Kelly and Goddard, 1969).

Brown notes that if the pH of the weathering solution is below 3.0

jarosite is the dominant phase. Above pH 3.0 goethite is the stable

+ + 2-
mineral. An increase or decrease in the activities of H ,K or S04

(perhaps caused by concentration or dilution) at a given Eh and activity

 113

of iron will cause the precipitation of one or the other mineral. It

is thought that bacterial activity, necessary in the first place for

pyrite oxidation, can raise the Eh and lower the pH sufficiently to

form jarosite rather than geothite.

Brown also notes that the replacement of jarosite by goethite has

been observed and that jarosite precipitation is always limited to sites

above the water table.

The formation of jarosite from goethite takes place according to the

following reaction:

3FeOOH + K+ + 2802- + 3H+ KFe3(804) 2 (OH)6 (43)

4

This reaction must be spontaneous since the free energy is -19.7 kcal

mole-l (Brown op.cit.).

7.6.3 Iron Hydroxides

The characteristic orange colour of the ochre deposits at Mam Tor is

typical of the ferric hydroxide minerals 'limonite' and goethite and of

the oxide hematite. There seems to be no consensus of opinion in the

literature, however, as to which mineral will form in a given situation

and the terms often appear to be interchangeable and are used loosely

to describe a ferric hydroxide with indeterminate water content.

Here 'limonite' is taken to be the freshly-precipitated phase which

is amorphous and, therefore, not a recognisable mineral; hence the use

of quotation marks. Goethite is the crystalline ferric hydroxide with

formula FeOOH (or HFe0 ). Hematite is the ferric oxide given by Fe203,
2
although formulae of the type Fe20 ·x H20 may appear in later equations,
3
representing a hydrated form.

In an article concerned with acid sulphate soil formation, which has

been cited several times previously, Van Breemen (1972) states that the

precipitation of iron minerals is pH-dependent. Below about pH 3 or 4

 114

jarosite precipitates whereas above such values poorly-crystallised

('limonitic') goethite, lepidocrosite (y Fe203) or hematite will form

together with sulphate ions in solution. At Mam Tor, however, the pH of

seepage waters is often as low as pH 3.0 and ferric hydroxides are the

only minerals to precipitate. Jarosite is only found encrusting highly

weathered shales where it is thought that evaporation and supersaturation

play an important role (the availability of K+ may also be greater here).

Changes in appearance and composition of the ochre formed in the seep-

ages with depth as noted in Sections 7.13 and 7.43, are consistent with

the observations of other workers. For example, Hanshaw (1974) notes that

the first phase to precipitate is amorphous 'limonite' which, with time,

dehydrates to form goethite as shown by reactions (19) and (20), page 65.

This sequence is also recorded at Mam Tor where goethite X-ray peaks begin

to emerge after prolonged burial (see Figure 29).

The persistence of organic matter at depth appears to contradict the

common belief that red beds indicate oxidising conditions. It would appear

that where ochre forms by chemical precipitation, as in this case, organic

matter is preserved as long as anoxic conditions are maintained inside the

deposit.

Hanshaw (op.cit.) reports organic matter being incorporated into a

goethite deposit in much the same way as at Mam Tor. He was able to calculate

a mean rate of accumulation of hydroxide of 0.9 mm per year, based on

carbon-14 dates of the lowest organic materials. Carpenter and Hayes (1980)

placed unglazed ceramic plates, which acted as artificial substrates, at

various levels in a stream from the surface to below the sediment-water

interface. The rate of iron oxide accumulation was found to be 120 ~g

cm-2 yr -1 near the surface and fell off to zero on the stream bed. The

iron content of Carpenter and Hayes' streamwater is far lower than at Mam

Tor and it is believed that the accumulation rate of goethite here is much

greater.
 115

Although there was insufficient time in this study to measure the

accumulation rates it is thought that such work might be worthwhile in

providing data concerning the age of the deposits and, ultimately, the

age of the landslide (assuming the goethite began to form soon after

the first movement opened up the rocks to acid attack). Hanshaw's

goethite deposit is estimated to have first begun to accumulate some

9,500 years before present, corresponding to the retreat of the most

recent mountain glaciers in that part of Colorado.

Iron hydroxide deposites are known to 'scavenge' and act as 'traps'

for base metal ions (Jenne, 1968). Indeed Hanshaw (op.cit.) reports

Cu, Ph, Zn-enrichment of goethite relative to the seepage water. Un-

fortunately, once again, there was not time to examine the trace element

content of the Mam Tor ochres although several water samples were analysed

for a total of eight base metals.

From Appendix Bxii it would appear that samples taken downstream from

ochre deposits (Sites 2, 11 and 16) contain lower concentrations of trace

metals than springs above the seepages (Site 7). This might suggest

that these metals co-precipitate with the ochre, although the normal

effects of dilution (as for other ions) must also be taken into account.

Also no definitive statement can be made based on such a small number

of samples and anomalous values at certain sites (e.g. Zn at Site 2,

Pb at Site 11) remain unexplained.

Future work, including more detailed sampling, could obviously be

undertaken concerning these goethite deposits and it is hoped that some

of the questions raised may subsequently be answered.

A more detailed knowledge of the various aqueous, crystalline and

ionic species involved in the weathering reactions has now been obtained.

Various inter-relations will be examined in the following Section from

the thermochemical viewpoint and with special emphasis placed on iron

chemistry.
 116

8. CHEMICAL THERMODYNAMICS

One of the most important chemical approaches to hydrology is to

establish whether or not natural water bodies approach equilibrium. By

calculating equilibrium conditions and comparing them with observed

conditions the degree of departure from equilibrium can be assessed, and

from that the probable direction and extent of spontaneous reactions such

as precipitation.

In this study it is essential to establish the state of equilibrium

of emergent waters with respect to atmospheric oxygen and to precipitation

of iron oxides and sulphates. Oxidation-reduction potential (Eh)is a good

indicator of the P02 status of waters and the approach used here to examine

pertinent equilibria is essentially the same as that of Garrels and Christ

(1965), who describe the methods of construction of Eh-pH ('stability

field') diagrams.

The heavy reliance on the above-cited work is easily explained. No

other text describes the principles of construction

.~
of these d~grams so

fully and in so logical a manner, and it is thought that a better under- "
standing of the equilibria involved here will be obtained by following

Garrels and Christ's methods fairly closely, with only slight modification

to suit our particular needs.

In Section 8.2, equations marked with an asterisk are those given by

Garrels and Christ (op.cit.). Many have been re-calculated using the

recently revised data of Table 16.

Before considering these equilibria in detail the thermodynamic data

of Table 16 will be used to calculate temperature changes induced by the

reactions described in Section 6.

 TABLE 16: Values of Standard Free Energies of Formation
-1 0
(kca1 mole ,25 C, 1 atmos)

Formula State llGof Source

Ca2+ aq -132.2 Latimer, 1952

Mg2+ aq -108.8 Langmuir, 1965

CaMg(C03) 2 c -518.7 Robie, 1959

Fe2+ aq - 20.3 Latimer, 1952

Fe3+ aq - 2.52 Latimer, 1952

Fe20 c -177.1 Rossini et al. 1952

3
Fe 04 c -242.4 Rossini et al. 1952
3
Fe(OH)2+ aq - 55.91 Rossini et al. 1952

Fe(OH)2 c -115.57 Rossini et al. 1952

+
Fe(OH)2 aq -106.2 Rossini et al. 1952

Fe(OH)3 c -166.0 Latimer, 1952

FeOOH c -116.88 Berner, 1969

FeS c - 23.32 Rossini et al. 1952

FeS2 c - 38.29 Robie and Wa1dbaum, 1968

KFe /SO 4) 2 (OH)6 c -794.0 Brown, 1970

H2O 1 - 56.69 Kubaschewski and Evans, 1958

H2S aq - 6.54 Rossini et al. 1952

HS - aq 3.01 Rossini et al. 1952

HS0
- aq -179.94 Rossini et al. 1952
4
S02- aq -177.97 Robie and Wa1dbaum, 1968
4
 117

8.1 Temperature Considerations

Chemical reactions can be divided into two main categories - those

which proceed spontaneously liberating heat (exothermic) and those which

require heat energy to be supplied externally (endothermic). It has been

noted previously that the oxidation of pyrite by atmospheric oxygen

(Equation 26) is spontaneous and exothermic. The heat liberated is calcu-

lated fromfue free energy of formation of the reactants and products as

described in Section 1.2.2:

= (26)

AGo f• = -38.29 0 - 56.69 -2.52 - 177.97 o

These values of AGo are in kcal mole-I. The total energy liberated is
f
calculated as:

= *(44)

= (2x(-2.52»+(4x(-177.97» + 0 - (2x(-38.29» + 56.69 +0

-583.368 kca1.

Therefore, two moles of pyrite (2 x l19.97g) yield 583.368 kcal. Since

1.5596 g pyrite are consumed per litre of water at Site 7 (page 69) the

total amount of heat liberated is given by:

1.5596
x 583.368 =
2 x 119.97

Thus the temperature of each litre of water at this site is raised

by riearly 4°C as a result of pyrite oxidation. Similarly, during goethite

precipitation according to equation 28:

(28)

= 12.52 o - 56.69 -116.88 -56.69 o

 118

!J.Gof= (2x(-116.88)) + (2x(56.59)) +0- (2x(-2.52)) - (5x(-56.69))-0

= -58.646 kca1.

Since 0.9816g goethite are precipitated the temperature rise is given by:

0.9816
x 58.646
2x88.858

A substantial temperature rise can result from oxidation and precipi-

tation reactions such as these, provided none of the energy is subsequently

used up in endothermicreactions. Carbonate and silicate dissolution

reactions generally have positive free energies, however, and require

energy in order to proceed. For example, the dissolution of dolomite

has the following free energy values:

* (45)
o o
!J.Gf = -518.7 o -132.2 -108.8 -56.69

!J.Gor= -132.2+ (-108.8) + (2x(-56.69») +0- (-518.7)-0

+ 164.32 kca1.

This energy reduction corresponds to a temperature drop (for 1.688 g

dolomite destroyed) of:

1.688
184.41
x -164.32 =

It is interesting to note at this stage that a set temperature rise

of 2.6oC is produced by these three reactions. The mean annual temperature

of the water emerging from Site 7 is 9.SoC. If these reactions did

not occur, therefore, and no others were involved, the temperature of

the water underground would be approximately 7°C. This is the value said

by Pitty (1976) to be the normal for this area (see also Pitty et al., 1979).
 119

It is not intended to take these temperature considerations any

further since additional reactions playa minor role and the quantities

of mineral species involved are not so well known as those described

above. It is worth noting simply that these reactions can account for

the observed level and consistency of temperature of the waters emerging

at'Site 7.

In certain instances pyrite oxidation does notm1y raise the tempera-

ture of drainage waters. In highly porous surface material water may not

be in direct contact with pyrite for more than a very short time during

a rainfall event. In this case the heat of oxidation is released into

the surrounding material which, in the case of spoil heaps, frequently is

highly combustible. Both Winmil1 (1915) and Guney (1968) have examined

the role played by this reaction in the spontaneous combustion of coal

and coal slag.

As early as 1915, Winmi1l considers the "heats of formation" of

various phases involved in the reactions given by Equations(9) and (10).

He records a value of 623 kcal of heat produced in this way and describes

laboratory experiments which produced results of a similar order. The

discrepancy between this value and the one of 583 kca1 given on page 117

is due to the refinement in free energy values over the years, and the

slightly different reaction used in each case.

Winmil1 found that for crushed pyrite, assuming all the heat developed

is used to raise the temperature of the material, a temperature rise of

600C inthree hours was obtained. The equivalent rise using oxidisable

coal alone would take 48 hours. According to Winmill, "it is quite clear

that where finely divided pyrites occurs in considerable proportion among

coal, it may be the chief factor in originating combustion".

The amount of pyrite in the exposed Edale Shale at Mam Tor is far too

small to engender such catastrophic events and readily combustible material

 120

is not at hand. Nevertheless, pyrite oxidation could substantially

raise the temperature of this material.

8.2 Mineral Stability and Eh-pH Diagrams

8.2.1 "
The Nernst and Debye-Huckel .
Equat10ns

The relationships between pH and the oxidation potential (Eh) of

an environment were first studied in detail by Clark and Cohen (1923) in

the context of aquatic biochemistry. ZoBell (1946) reviews the early

work on the subject related specifically to geological situations.

Neither of these authors portray the stability fields of minerals in the

now familiar form of Eh-pH diagrams, however, and the first publication

in which these appear is the classic work of Pourbaix (1949).

As mentioned previously, principles of construction of Eh-pH diagrams

described here are based on the method of Garrels and Christ (1965) who

note that for every oxidation-reduction system the half-reaction can be

written as:

reduced state = oxidised state + ne *(46)

where n denotes the number of electrons involved.

The oxidation-reduction potential, Eh, for such a half reaction is

measured in volts relative to the standard hydrogen electrode and is

given as:

Eh - EO + RT In [Oxidised State] *(47)

nF [Reduced State]

where EO is the standard half-cell reaction

(The voltage of the half-cell when all the
activities are unity)

R is the gas constant

T is the temperature in degrees Kelvin

n is the number of electrons

F is the Faraday constant

 + 00
<
+ ~
..
~ ::x:
.~
o z
..
+J

~ , + til
C"
r.LI
+.~ N~
0 U

...... ..
..:I 0
~ ..
~
Q)
+
N , .. +
~
:;j

T ..
U
0 NM
0
til , ..
..
.~ ,
Q) N
>. + 0
.0
o
Q)
N

Z
NM
0 , ~ , .. Z

Q)
Cl
.~
0
+
.. .. U
0
p., 0
M

-5 N + N Z
::x: ..
, .. ,
Q) N
~
.~
~
..
.0
p.,
.~ , ..
M 1-1

III
+
N
::x: ,
0
U ..
f N
til
.. '"'
'"' .. + .. !Xl

,........
Q)
+J +
+
N N
III
Z
~
'"'
III
p.,
+::x::
.. ..
~
til
..
00
::x: , .. ..
N~
u

+ + + 0 +
II-l M N
~
N
'"' ~..
, .. ,::x:
"'Cj U
o
'"' N U
en
Q)
U
.. +
.. +
.. til
+ + N~
~+
,~..
;j
...... M N N N 0
III ~ Q) Q) ;j III Q)

:> til ~ !Xl U !Xl til

.. + .. +
.. + .. + .. , .. , ..
+
..;t M N N N N~
...... 00 III 0 ::x:
e:: < ~ U '"'
til til 0

.~
III
......
......
0\ 00 \0 Ln ~ M
 121

Square brackets indicate activities or the~odynamic concentrations of

participating solutes.
0
Converting to base-lO logarithms at 25 C and 1 atmosphere total

pressure for the reaction:

aA + bB - cC + dO (48)

the half-reaction becomes:

[CJc [OJd
0.0592 (49)
log
n
[AJ[BJb

This is known as the Nernst equation.

In addition, for EO in volts:

... (50)

The only problem remaining is that actual quantities of solutes as

determined by chemical analysis cannot be used in the Nernst equations.

Instead such values must be converted to activities by mUltiplying the

molar concentration by the corresponding activity coefficient. Activity

coefficients are computed for each ion using the Oebye-H~ckel equation

given as:
Az.2 If
- log y. I: ----
1 (51)
1 l+Ba.1f
1

where y. is the activity coefficient of the ion

1

A is a constant relating to the solvent

(for water at 2SoC it is 0.5085)

z. is the ionic charge

1

B is a constant relating to the solvent which for

water at 2SoC is 0.3281

a. is a constant related to the effective diameter

1
of the ion in solution (values of which are
given in Table 17).

I is the ionic strength of the solution

 -.::t
......
00
I 00 " '"
'"· 00
N-.::t
0
U)
-o
00
·
'"
N
· 0 N
N
-.::t
N

0\

I 00
...... '" '" ......
......
U
N
-.::t
......· · · 0\
0 ......

M
+ -.::t
M N -.::t

·
-.::t
0·
en
'"'
Q)
~
Q)

0\
'"0· " 0
+J N
I'd
~
" + N
"
M

'"......· "·
Q) N

·
+J Q) 00 M
.,-l ~ 0 ......
U)
'"
-
00
Q)
..c::
+J N
I 0\
~
. · '"'"· ·
0 -.::t
.,-l ...... +~ -.::t
-.::t 0 0\ 0
en ><
~ 0 0 0
0
..... -.::t
M
'"
14-1 0\ N
0
II + I'd ...... '"
'" 0\

· · '"·
en \l) 0\
Q) H Z
.,-l -.::t 0 0 0
+J 14-1 ......
.,-l 0
::-
.,-l Q)
+J ;::I 00

-<
() ......
I'd + ...... '"
"
"C
~
::- N
CO
l: · 0·
00
"· \l)
I'd
Q)
..c::
+J
'" "
00
......
en
Q) ~
.,-l 0
+J
..... "C M
+
·...... "· ·
Q) 00
'"'
I'd
......
en
I'd
N
I'd
0 00

0
00

00
.0 U -.::t 0
~ '-' 0
-.::t

-
-
00 M
...... I
0
>- ......
~ '-'
><
~ ,-.. +J
~ ......
E-I §
~
......
M
I
0
-
......
Q)
.,-l
()
.,-l
14-1
14-1
I

.,-l
Q)

'"'
+J

......
I'd
;::I X Q)
~
~
'-' 0
CJ
a
.._,
I'd ......
'-' I >. >.
Q) +J +J

a·
.,-l .,-l
'"'
+J ::- ::-
c,· .,-l
......
.,-l
+J
.,-l
+J

p.· l: -<
() ()
-<
 122

The parameter I is computed from the expression:

I .. (52)

where M is the concentration of a given ion, in moles per litre. The

summation assumes that every ionic species present in solution is known.

In practice only the ten or so major contributing ions are needed for a

reasonably accurate estimate of ionic strength. Using mean compositional

parameters for the Site 7 waters, I = 0.0934.

Table 18 shows the computed values of each ion in the Site 7 waters

based on this value of I. In subsequent calculations activities of pure

solids and pure liquids are taken to be unity and the activity of a gas

phase equal to its pressure in atmosphere.

8.2.2 The Stability of Water

since we are primarily interested in the stabilities of minerals in

the presence of an aqueous phase the first stage in construction of an Eh-

pH diagram is the plotting of the region in which liquid H20 is itself

stable. The reaction can be written:

2H2O ... +
+
4H aq + 4e *( 53)
l °2g

!J.Gof • -56.69
° °

!J.Gro ...
° + (4xO) - (2x(-56.69» ... + 113.4 k.cal ,

-
l!.Go
r
nx23.06 - 113.4
4x23.06
.. 1.23 volt

P02 [H+ ] 4
0.059
Eh • EO + log *(54)
n
[H 0 J2
2

Since P02 and H20 are unity,

Eh 1.23 + 0.059
OIl

4 *< 55)
Figure 30 Eh-pH diagram for water and iron oxides
 .0.8 water oxidised

.0.6

+0,4
hematite + water
.02

Eh (volts) 00

-02

-04

-0.6
water reduced
.. .. ...
-OB .. ... ...
..
-1.0 L-_~ ---,,....- ""

2 4 6 8 10 12 14
pH

•
 123

Substituting -pH for [H+]

Eh = 1.23 0.059 pH *(56)

A line representing this equation can be plotted on an Eh-pH diagram,

having an intercept of 1.23 volt Eh and a slope of -0.059 pH, (see Figure

30). This line representsthe equilibrium between water and oxygen at 1

atmosphere total pressure. A similar line representing the lower stability

limit of water (in equilibrium with hydrogen gas) can similarly be found

from the half-reaction:

H2g = 2H+
aq + 2e * (57)

t:.Gof= 0 0 GO
r
.. o - 0 0

EO = 0 =
and 0
2x23.06

Therefore:

[H+J
0.059
Eh o +
2
log *(58)
PH2

= 0-0.059 EH for PH2 = 1 * (59)

The region between these two lines (shown on Figure 30) represents

those conditions of Eh and pH under which water, at or near the surface

of the earth, is stable.

8.2.3 The Iron Oxides

By expressing reactions between minerals (in the presence of water)

in oxidation-reduction terms it is possible to calculate stability fields

in much the same way as for water itself. Since iron minerals are of

particular interest in this study the stability relationships between

the various possible phases containing iron will be studied in more

detail.
 124

The stable iron oxides in surface environments are hematite and

magnetite. By eliminating oxygen from the reaction describing the

oxidation of iron to magnetite and adding in its place the water

dissociation half-cell we can write:

+
+ 8H aq + 8e * (60)

-56.69 -242.4 o

o
llG r = -242.4 + (8xO) - (3xO) - (4x(-56.69» = -15.6 kcal.

-15.6
8x23.06
= -0.084 volt

Eh = -0.084
0.059 [Fe30 J [H+J8
+ 8 log
[Fe J3 [H20 J4

*(61)

Substituting -pH for log [H+]

Eh = -0.084 - 0.059 pH *(62)

Therefore, the reaction between iron and magnetite is a straight line

with the same slope as the H20-H boundary. Since the Fe-Fe 0 line is
2 3 4
below the lower stability limit of water, this reaction cannot take place

in the presence of water if equilibrium is maintained.

The oxidation of magnetite to hematite can be represented by:

+
= + 2H aq + 2e *(63)

-56.69 -177.1 o
o
llG r = (3x(-177.l» + (2xO) - (2x(-242.4» - (-56.69) = +10.2 kcal.
Figure 31 Eh-pH diagram for iron hydroxides
 to ,.-----oIt..---------__,.----.
08

0.6

0.4
ferric hydroxide + water

Eh(V)

-10~--~--~----~------~----~~
2 4 6 pH 8 10 12 14

•
 125

EO +10.2
= 2x23.06 = 0.221 volt

0.221 + 0.~59
[Fe20
3
J3 [H+J2
*( 64)
Eh = log
~e304]2 [H2O]

= 0.221-0.059 EH * (65)

This line once again has the same slope but this time lies within the

stability region for water (see Figure 30).

8.2.4 Iron Hydroxides

Hydroxides of iron are usually the first minerals to form as precipi-

tates from solution. Both ferrous and ferric hydroxides, however, are

unstable with respect to the more ordered forms considered above, but

nevertheless may persist for some time as colloidal suspensions or stream

bed (ochre) coatings. The fact that considerable periods of time are

needed for the conversion to hematite or goethite has been noted in

this study as well as in numerous others described in more detail in Section

7.6.3.
The stability fields for hydroxides of iron, although ephemeral, can

be plotted in the same way as for the oxides. The reactions involved are

given below, without the intermediate stages:

Ferrous hydroxide: =

Eh = -0.047 - 0.059 pH

Ferric hydroxide:

Eh = 0.271 - 0.059 pH

The resulting diagram (Figure 31) is remarkably similar to that for the

oxides.
 126

The fact that freshly precipitated ferric hydroxide will in time

change to hematite or goethite is predicted by the negative free energy

of reaction in the following equation:

* (70)

= -166.0 -177.1 -56.69

l::,.Go = -177.1 + (3x(-S6.69» - (2x(-166.0» = - 15.17 kcal. * (71)

r

8.2.5 Relationships Among Ions

Since free energy data are also available for ions in aqueous

solution it is possible to calculate fields of stability for ions in

equilibrium with the solids. Certain of these reactions do not involve

oxidation or reduction and are, therefore, independent of Eh. Such is

that between the ferric ion and ferric oxide:

= 2Fe3+ + *(72)
aq

l::,.Go = -177.1 o -2.53 -56.69

f

l::,.Go (2x(-2.S3» + (3x(-S6.69» - (-177.1) - (6xO) = +2.0 kca1 *(73)

r

Since it is not possible to calculate Eh the equilibrium constant for the

reaction must be used. For the reaction:

aA + bB = cC + dD (48)

the equilibrium constant is given by:

[Cl [DJd
(74)
=
K
r-r [BJb
and K is in turn related to the standard free energy as follows:

AGO
t: = - RT 1n K (75)
r
 127

where Rand T are the gas constant and absolute temperature, as before.

~Go
r
= -1.364 log K (76)

Therefore for the reaction of ferric iron with ferric oxide:

~~
r 2.0
log K = -1.364 = -1.364 = -1.45 (77)

Since ~~J and [H2O] are unity

log -1.45 (78)

Rearranging and substituting -pH for [H+]

2 log [Fe3+] = -1.45 - 6 pH *(79)

Thus, if the activity of ferric iron is stipulated then pH is fixed.

Figure 32 shows contours of [Fe3+] plotted in this way. [Fe3+] = 10-3

is shown as a heavy line since this is approximately the value for Site

7 water (see Table 18).

Oxidation is involved in the reaction between ferric ions and magnetite

which can be represented by:

Fe 0 + 8H+ = 3Fe3+ + 4H 0 + e *(80)

3 4c aq aq 2 l

where Eh = 0.337 + 0.177 log [Fe3+] + 0.472 pH *(81)

Here a value of either Eh or pH must also be stipulated in addition to

[Fe3+J. The contours on Figure 32 are easily extended through the magnetite

stability field by taking the pH values at the point where the Fe203-Fe 0
3 4
line is reached.

The reactions necessary to calculate the relations of other possible

ions to hematite and magnetite are given below without intermediate steps.
Figure 32 Composite Eh-pH Diagram Illustrating the 'Solubility'
of Ferric and Ferrous Ions. Total Activity of Iron =
10-3 molar
 F ,.
e"'1

0.6

Eh (V)

-02

-0.4

-06

-0 a .

-10~--~--~--~----------~__j
2 4 6 pH 8 10 12 14

•
 --
N
co
'-"
~
co
'-"
- - - - -'" -'"
co
'-"
'"
co
'-"
o
'"
'-"
r-I

0'\
'-"
N

'-"
~

'-"

::cc, ::cp.. ::cc, ::cp..

::cc,
::cc, ::cc, CO CO
::cp..
'"
11"1
CO
r-I
r-I \0
r-I
r-I
--r
r-I
'"0.
N

.....
Q) N
..... ~
.~
.j.I

.j.I
N
.....
..;t
0
+
0
r-I
N

0
0
I
0 0
+
Q) + ,.-, 0
rI"--' rI r--1
~
--
0
+ + I I

-
eo r--1 + N I I
N N
I'd + + r+t r+t ::c
::c 0 0 0
~
"0
r
~
N
::c
0
::c
0
+
N
+
N
0
'-" '-" ~
Q)
~
Q)

~ Q) Q) '-" Q) Q) Q) Q)
t.3~
I'd
~ ~ L.3
Q)
~~ ~t.3 en eo
Q)
00 0 0
.~
.j.I 00 00
00
00
0
00
0
00
0 0 r-I r-4

-'"
0 0
.j.I r-I r-I 0 r-I r-I r-I r-I
::cp.. ..... CO
I'd
S .....
::cp.. ..... ..... ::cp.. .....
r-I

'"0
CO
CO '"
11"1
CO
CO
'"
11"1
0
CO
0
Q)
::c
CO
r-I
~
r-I N r-I

0
r-I
.....
r-I
11"1

0
0
0
0
0
0
0
. 0 0
0 Q) 0
.j.I 00 --r 0 I
I'd + r-I + CO I I
+ 0'\
Cl)

.~ ..
Q)
p.. N
..... ..... 11"1
.
r-I
&3 ..... 0 0 ..... --r '"
~ r-I
~
.
I r-I N CO r-I r-I r-I CO
~ ~ ..... .....
.~
.j.I 11"1
\0
0
I II
\0
'"0
N N
r-I r-I
I
.~
:>
'" 0 I 0 r--1 r-I 0 0 0 I
+
-
r-I
.j.I
(J
..et:
..
.j.I
II
,.--, .c
II
~
~ .c ::c .c
N II II
.c .c
II II
.c .c
II II
.c .c
::c
.~ .~
!:il !:il !:il
~+
Cl) !:il !:il 0 !:il !:il !:il !:il
(J 0 '-"
~ Q) Q) Q)
~ .c ~
L-..J L.5 t.5
0 U
H
"0 00 00 00
bO 0 0 0
~ ~
.~ r-I r-I r-I

.j.I Cl)
I'd r-I
r-I Q)
Q) ~
~ ~
I'd
!90 C!)

.~ ~
.j.I Q)
(J .j.I Q)
I'd 4-1 Q) <II <II
Q)
..et: + N N
N
~ '-" + Q) Q)
+ +
<II + N N
+
+ + N

'"
r-I
0
N
::c
0
+
N
::c 0N
--r ::c
+ +
-
::c
N_
::c
0
0 '-"
'-"
~
Q)
+
\0
+
::c +CO
::c
+ + +
::c ::c
--r 11"1
+ +
N
Q)

+
0
::c
N
N

~ ..;t ~ Q)
~ --r
~ ON ~ ~ + + +
+ + N_ N 0 0 0
e;] <II ::c II ~ --r N ~ N ..;t
~ o 0 Q) Q) ::c 0
E-t
~ ~+
+ '-"
0
+ N ~ ~ ~ N
<II Q) II Q) + ::c ~
Q) <II
~
+ Q)
~
~ ~
N ~
~
~
::c
N
N II II
~
0
N + II II 0 0 0
II II ::c +
0 0 0
N
::c ::c
N N
<II +
+ + + + 0 N N N ~ ::c
::c CO
::c ::c N ::c ::c
Pi ..;t + +
+
+ ::c II

--
\0 11"1 N

+ +
~ ..;t
0 0
~
-
N

::c
0
'-"
+
0
~
N
+ +
0
~ ...;j"
0
~
+
N
+

<II
+
N
+

<II
+

'-"
::c
0
<II
0
+
::c
'-'
<II
I
0
<II
~
N
I
0
~
N
Q)
N <II N
<II <II ~ Q) <II <II ~ ~
~
~ ~
~ ::c N
::c
~ ~ N ~ ~ ~ N N N
 128

All this information is brought together into a composite diagram in

Figure 32. Boundaries are drawn in by joining points of equal activity

2+ 3+
at their intersection, as illustrated in the case of Fe and Fe

in Figure 35, and represent a set of points where the activity of the

ion dominant in the field on one side is equal to the activity of that

dominant on the other side.

The fields of many of the ions considered in Table 16 do not appear

in the final diagram (Figure 32). This is due to the fact that they occupy
2+ 3+
parts of the fields already assigned to Fe and Fe • The latter ions

dominate these fields and the effect of other ions is 'negligible. Thus,

very few of the total possible species are in fact present in appreciable

amounts within the stability field of water. Possibly the two most striking

features of Figure 32 are the small field of Fe3+ at highly acid, oxidising

conditions and the large field of hematite which is stable over a considerable

range of conditions from acid-oxidising to alkaline-reducing.

A line representing the stability of ferrous ions in equilibrium with

goethite rather than hematite is shown on Figure 34 as calculated from the

relationship:

Fe2+ 2H 0 = + + e (94)
aq + 2 1

2+
Eh = 0.73 - 0.118 log Fe - 0.177 pH (95)

2
This lies very close to the Fe2+-Fe20 boundary for [Fe +] = 10-3 and
3
shows that the minerals hematite and goethite have virtually identical

stability fields.

8.2.6 Dissolved Sulphur Species

By adding sulphur to the iron-oxygen-water system which has been

developed here, information can be obtained regarding the iron sulphide

Figure 33 Eh-pH diagram for dissolved sulphur
species. Total activity of sulphur
10-1 molar.
 10

0.8

HSO;
06

04
SO:-
0.2

Eh(V} 00

-02

-04

-06

-OB

-, 0
2 4 6 pH B 10 12 14

•
Figure 34 Composite Eh-pH diagram for the system
S-Fe-H20. Activity of sulphur = 10-1
molar; activity of iron = 10-3 molar.
 10
Fe
OB

0.6 boundary for HFeCh

04
Fe10) + HID
0.2

Eh{V) 00

-02

-04

-06

-08

o 2 4 6 pH 8 10 12 14

•
Figure 35 Method of construction of field boundaries
and the plotting of Eh-pH values of Mam Tor
waters. See Table 22 for description of
Samples 1-4.
 , 2 3
pH
4 5

08

0.7

06

Eh(V)

05

'$0
03
~.
./
\\
0" 'c?
0" S 10
-~
UJ
-n -=n ..."

02 ~-
II
.
ft\.. ~~
II
...
It)
It):

II

...
I
I
0-
t
OJ

01

/
 ,..... ,..... ,..... ,.....
~

- -
r-.

-
00

-
\0
~ ~ ~ ~
-IC -IC oj( -IC

M
I
0
M

,...,,...
~
..c::
p.
M
~
CIl
r-1 ,..., r+t ,...,ro
i'o11""""'
r-1

I ::c
M I
ca ..;t CIl
N..;tN
+J 0 N N..;t
I CIlN ~..;tll
0 CIl ::c o o CIl CIl

.___,H ::c L...l ~I-.J

CIl ::c CIl
L-.J
L-.J
.
I-.J L-J L...J
,... eo eo eo eo
0 0 0 0 0
4-1 M ,_oj ,_oj M

til r-. r-. r-. r-.

0
or-!
CJ
Q)

Q)
0
0
0
0
0
0
0
0
0
. 0
0
.
c,
CIl + + + +
,... ::cp. ::cc, ::cp. ::cc,
~
..c::
c, ..;t \0 ~
\0
M
~
CIl
\0
0
0
.
r-,
0
0
.
\0
0
0
.
11"1
0
0
.
"tj
Q)
:> M N 00
M 0
0
til
~
N
g 11"1
N
..;t
M
til
or-! 0 0 0 0
~ II II
I::l
Q) ..c:: ..c:: ..c:: ..c::
Q) r>:l r>:l r>:l r>:l
E Q)
~
til
I::l
0
or-!
+J
CJ
ca
~ Q)
00
Q) Q) Q)
00 00
+ 00
0
N + + + +
::c +
~ + + ::c
r>:l ::c ::c
....:I + 0 ~ 00
M
~ + +
H -::r + I
0 N..;t
CIl I I
::c N -::r N -::r 0
0 0 CIl
CIl CIl

II II

0 0 0 0
N N N N
::c ::c ::c ::c
..;t
..;t ..;t ..;t

+ + + +
CIl CIl I I
N N CIl N
::c ::c ::c CIl
 129

minerals pyrite and pyrrhotite. The reactions listed in Table 20 help

delimit the stability fields of the various dissolved sulphur species

plotted on Figure 33.

At the boundaries the activities of sulphur species on each side

are equal and lines can be drawn as simple functions of Eh and Ph. The

boundary between HS04 and SO~- is obtained from the relation:

HS0
4 aq
H+
aq
+ S02-
4 aq
* (lOO)

[So~-J -10-1.9
~S04
= = * (101)
lHS04] [ H+ ] [ H+ ]

2 -1
Since the activity of S04- in the Site 7 water is approximately 10

moles per litre the boundaries of Figure 33 are drawn for this value.

8.2.7 The Composite Diagram

In considering reactions between the sulphides and iron oxides it

is found that many of the theoretical boundaries lie outside the stability

field of water, and pyrite is the stable mineral throughout the H S field.
2
The reactions and Eh equations used to calculate the boundaries on the

composite diagram (Figure 34) are given in Table 21.

The Eh-pH diagrams presented here are the simplest that can be drawn

based on the water compositional parameters of Table 18. Other dissolved

species included in Table 16 could be incorporated into the diagram together

with dissolved CO2 and other gaseous phases which are undoubtedly present

before the Site 7 waters emerge at the surface.

It is, however, beyond the scope of this study to do so since the

final composite diagram (Figure 34) adequately describes the relationships

which have been observed in the field. Discussion of these observations

in terms of Eh and pH follows in the next Section.

Figure 36 Eh-pH diagram for iron hydroxides and
sulphides, including jarosite. After
van Breemen, 1972.
 +1.2

.1.0

jarosite

••
•
.06
•
Eh [V]
+04 "limonite"

o
pyrite
-0.2 ......

.....
-04 ......
.....

o 1 2 3 4 5 6 7 8 9
pH

•
"%j "%j N "%j "%j N N W N "%j
(l) (l) ;:c (l) (l) ;:c "%j "%j "%j (l)
Cl) Cl) Cl) Cl) Cl) N (l) (l) (l) Cl)

N I N N Cl) Cl) Cl) Cl)

+ N N N +
;:c
II + + + + ;:c
+ + +
+ I-rj I-rj cc I-rj
......
Cl)
(l) (l) ::c ~ (l) ......
\0
N
\0
I
II N NN N N 00
+ + 0 0 + ;:c ;:c ;:c
"%j N N N
0 I-rj
(l)
N
+ II II II II 0 0
(l)
+ N
Cl)
"%j
(l)
N
Cl)
N
;:c
"%j
(l)
II II II Cl)
N
+ Cl) 0 Cl) Cl) ~ "%j "%j
+ N ~N 0 N ;:c (l) (l) +
::c I ~ I en w N
;:c
Cl) N + + 0
~ ~
0 0
I (l) + + I w + t-3
N
;:c
~
;:c >
"%j "%j + + + + t:d
t"'
+ (l) (l)
N
+ I-rj 0- ~ N t:.::I
N
+ + + + (l) en en CD
N 0 0 N
N + + N 0 ~N~N ......
(l) (l) w I I
...... ......
0- ~ + + +
;:c ;:c
+ + W 1.11 N ~
~ 0' 00
;:c
CD
+ + ;:c ;:c III
...... ......
+ + + (')
no
~ ~ + + + 1-"
0
(l) CD
W ~ W ~
0 ~ 0
(l)
(Jl
(l) CD
0-
(l)
no
~
(l)
...... t:.::I t:.::I t:.::I t:.::I t:.::I t:rl t:.::I t:.::I t:.::I (l)
0 ::r' ::r' ::r' ::r' ::r' ::r' ::r' ::r' ::r' ~
OQ
II II II II II II II H
t;;j1 0
(l) I I ~
N 0 0 0 0 0 0 0 0 0 (Jl

U w
~
~ w
VI
w 0
VI
w
0-
w
ex:>
w
00
w
~ t:j
II "-J W
0 0 ~ \0 "-J 0- ~ 0 0 1-"
(Jl

+ I I I I I (Jl

~ 0

~ .
0
0
0
· · ·
0
0
0
0
0
0
......
. . . .
0
0
0
0
0
0
0
0
......
<:
(l)
N VI 0- 1.11 ...... 0' "-J "-J N 0-
\0 \0 "-J \0 00 "-J VI VI \0
Cl)
'0
::c ...... 'd '0 'd 'd '0 'd 'd
;:c
'd
;:c
s::
......
0 ;:c ;:c ;:c ;:c ;:c ;:c
OQ 'd
+ + + + + ::r'
...... r;;j' c::
· · · . . . . .
0 (l) 0 0 0 0 0 0 0 0 t;
OQ N
...... L....J
+ 0
VI
0
0
0
0
0
VI
0
0
0
0
0
0
0
N 'd
Cl)

0 \0 ex:> ex:> \0 ex:> ex:> 00 \0 (l)

(')
I
...... ...... ...... ...... ...... ...... ...... ...... ...... 1-"
0 0 0 0 0 0 0 0 (l)
OQ OQ OQ OQ OQ (Jl
OQ OQ OQ

...... ...... ...... ...... ...... ...... ...... ...... III

0 0 0 0 0 0 0 0 ~
I I I I I I I I 0-
...... ...... ...... ...... ...... ...... ...... ......
Cl)

I + + I 0
......

· · · .
0 0 0 0 1-"
0-
(Jl
0 0 0 0
N 0 0 N
\0 ~ ~ \0

...... ...... ...... ......

0 0 0 0
(IQ OQ OQ OQ

r;;j1 r;;;-' .r;;" r;;j1

CD (l) (l) CD
N N N N
.....:t ~ ~ ~

,....
* ,....
* *
,.... ,....
* ,....
* *
,.... *
,....
'"'* * *
...... ...... ...... ...... ...... ...... ...... ...... '"'
...... '"'
......
...... ...... 0 0 0 0 0 0 0 0
...... 0 \0 ex:> "-J 0- 1.11 ~ W N
"'" "'" "'" "'" "'" "'" "'" "'" "'" "'"
 130

The reader interested in Eh-pH diagrams in general, and their

extension to other systems, is referred to the following publications.

General and iron system diagrams: Hem, 1961; Hem, 1970; Hem and Cropper,

1959; Barnes and Back, 1964. Aluminium, fluoride and sulphate: Hem,

1968. Bicarbonate and iron: Hem, 1960a. Manganese: Hem, 1972. Sulphur

species: Garrels and Naeser, 1958; Hem, 1960b. Trace metals: Hem, 1977.

Van Breemen (1972) considers the mineral jarosite in an iron stability

field diagram. This is reproduced in Figure 36 purely for interest, as

the activities of certain species are far removed from those used in this

study. The jarosite field would be appreciably smaller had Van Breemen

constructed the diagram for the stable minerals goethite and hematite

rather than for the metastable 'limonitic Fe203'·

8.3 Further Consideration of the Eh-pH Diagrams

The measurement of oxidation-reduction (redox) potentials depends

on the use of an electrode made from a noble metal, commonly platinum.

When such an electrode is placed in a solution containing, say, ferric ions

3+
the rod becomes positively charged as the Fe ions react with electrons

from the platinum:

3+ - (112)
Fe + e =

Conversely, when the electrode is placed in a solution containing only

2+
Fe ions it becomes negatively charged.

The amount of negative or positive charge developed, measured relative

to a zero reference electrode (commonly of calomel), is dependent upon

the ratio of reduced to oxidised species present. A solution having a

negative redox potential is reducing, and one with a positive potential

is oxidising.
 .....
· · · ·
('oj ('oj ('oj ('oj

::cp., M M M M

· · ·
r-I Lt) M M M
::cp., M · M M M

('oj
0
"
+J Lt) \D 0 ..... 00
r-I M 0 ..... ..... 0"1 0
..:t
· +· +· +·
0 Lt) Lt) Lt) r-I 00
>
.._, 0
+
0 0 0
>.. '"..
r-I

..c 14
w cO >..
;:l 14
s:: cO
;:l
cO
I-) 14
,.c
"Ij Q)
14 ~
r-I M
('oj ..c
"
+J Lt) 00
..:t
0"1
Lt)
0"1
M I=l
+J
\D
r-I \D
0 M ..:t ..:t ..:t 0
>
.._,
~
· 0· 0· ~· 00
s::
..c
+ + .~ 00
c
w r-I ,~
r-I
p.,
~
til ~
til
~
Q) $.I
..... Q) +J Q)

Q) .~
+J 4-1
cO
+J
4-1

.~
+J

Ul
til

00
til
Q)
cO
til
I=l +J
;:l
~Q)
8 ,~
0 >c: I=l Q)

0
I=l
14
4-1
,~
8 'S )

0
,~ Q) 0
+J ~ > ..:t ~
('oj cO Q) 0
('oj cJ ~ ,.c " II II
0 +J cO..c
cJ r-I ('oj
H til
r-I+J
r-I ,~
~ ..... cOUl
..... "Ij ..... 4-1
~ s:: til
Q) 8 Q) Q),~ Q)
+J +J +J"Ij >c:
,~ 0 ,~ cOO ,~
en r-I en ~.._, 4-1
4-1
::I
Ul

.....Q) ~
Q)

r-I ('oj M ..;t

~~
enz
 131

In highly oxygenated:waters a layer of oxide may form on the sur-

face of the electrode and reactions such as thatgiven above (Equation

112) may become irreversible. It is essential, therefore, to periodically

clean the Pt in a strong acid solution, such as hot H2S04•

Many systems of interest to the geologist, such as dilute aqueous

solutions and soft sediments, present problems in measuring redox

potential and in the interpretation of results (see Golterman, 1975;

Schindler and Honick, 1971). Complex ions and ions with valencies

greater than about 3 are often responsible for inconsistent and ambiguous

readings.

The aqueous system presented by Spring 7, however, appears to be one

which can, at least in part, be adequately described in terms of Eh and

pH. This is due to the fact that the ionic strength of the solution is

relatively high (facilitating accurate redox measurements) and the system

is dominated by a few simple ions.

In addition to illustrating the method of construction of field boun-

daries, Figure 35 also contains plots of Eh-pH values recorded at Mam Tor.

Redox potentials were measured using platinum and calomel electrodes on

an Orion Research Microprocessor ionalyser (Model 901). This instrument

is fragile and requires an electrical power supply and, therefore, could

not be taken into the field. Instead samples were collected in airtight

polythene bottles, from which all the air was excluded, and returned to

the laboratory for analysis within forty minutes of collection.

Values plotted on Figure 35 are from Table 22. The four sites

are located at irregular intervals along the stream flowing between

Site 7 and the Odin Sitch. In addition to readings taken immediately

on return to the laboratory corresponding measurements were made, two

weeks after collection, on the same samples. During the interim period

the sample bottles were left covered but not sealed and placed in full

light (not direct sunlight) on a windowsill.

 132

These solutions originally plot on Figure 35 in the stability

field of ferric hydroxide. Since ochre can be seen to precipitate along

the length of the stream this is not entirely unexpected, and the water

must be supersaturated with respect to ferric hydroxide.

With continued oxidation, presumably caused by mixing with atmospheric

oxygen, the waters attain higher redox potentials while at the same time

pH falls slightly. The precipitation reaction has been given previously

as:

2+ +
ZFe + !OZ + 5HZO = ZFe(OH)3 + 4H (19)

and it was noted in Section 4.4.2, that with time ferrous ions are indeed

lost from solution. This cannot be accounted for, however, by simple

oxidation to ferric ions as seems to be suggested in Table 4. The gene-

ration of hydrogen ions by reaction (19) clearly illustrates why ferric

hydroxide precipitation makes the solution more acid.

Buffering of the waters to the narrow pH range of 3 to 3.5 (noted

in Section 4.4.1 and illustrated by Figure 11) can also be explained in

terms of the stability field diagram. It can be seen that as the solutions

move up towards higher Eh and lower pH values the solubility of ferrous

iron becomes vanishingly small. Once a pH of about 3.0 is reached the

8
log of the activity of ferrous iron is less than 10- , so small that

further ochre precipitation is precluded and the pH can fall no more.

Extrapolating back from the initial pH and Eh values of Figure 35

the 'starting point' for the oxidation of these waters can be found (or

more precisely, the region in which the starting point will lie). This

can be seen from Figures 33 and 34. to be the region where pyrite is in
...
equ111br1um . h S02-
W1t 4 an d F e 2+ 10ns
. - the precise environment produced

by pyrite oxidation according to the mechanism discussed in Section 6.

Baas Becking et al. (1960), have accumulated most of the Eh-pH data

available from the literature on natural waters up to 1960. This inform-

Figure 37 Eh-pH diagram showing the field of
natural waters. After Baas-Becking
et al., 1960.
 10

08

06

04 s;!'.
~Qf!)s

02
v
Eh (V) 0.0

-0.2

-04

-06

-08

-,0
2 4 6 pH 8 10 12 14

•
 133

ation is reproduced in diagrammatic form in Figure 37.

It can be seen that highly oxygenated surface waters (including

acid mine drainage) plot along a line parallel to the upper stability

of water itself. Sato (1960) believes that this represents the true

upper limit of water in as much as further oxidation can only proceed

via a rate-determining, intermediate hydrogen peroxide step.

From Figure 37 it is apparent that the 'starting point' of the

(underground) Mam Tor waters is in the region which overlaps the three

labels, 'ground water', 'bog water' and 'waterlogged soils'. After

emergence and continued oxidation the Eh and pH approach the region

labelled 'mine waters' at the extreme upper limit of aqueous environ-

ments. This once again reaffirms the similarity between these and acid

mine drainage solutions.

8.4 Gypsum Solubility

The only other mineral which can be seen to precipitate directly from

solution in the seepage areas is gypsum (see Section 7.6.1). Since only

a short dry spell of weather is required for this mineral to begin to

precipitate on the seepage vegetation by non-equilibrium evaporation (Smith

and Drever, 1976), it is thought that the waters must already be close to

saturation with CaS04•

Table 18 lists the activities of the major solutes in the Site 7

waters. From these figures can be calculated the soZubiZity product of

gypsum as follows:

K [ca2+] [ so~-J (1l3)

= 0.0088 x 0.0225
4
= 1.98 x 10-

= 10-3•7
 134

From published equilibrium constant data, however, the solubility

product of gypsum is given as 10-4•6 (Sillen and Martell, 1964). The

value calculated for Site 7 is nearly eight times greater than this,

which predicts that the seepage waters are supersaturated with respect

to gypsum.

This clearly is incorrect since gypsum only precipitates after

capillary rise up the plant stems and subsequent evaporation. Any crystals

falling back into the water are re-dissolved, proving that the waters are

undersaturated.

One possible explanation of this anomaly is that the ion pair CaS04°

(zero charge), or more complex ligands, are forming in solution. These

would have the effect of lowering the solubility of gypsum whilst at the
• .• 2+ 2- .
same t1me contr1but1ng to the Ca and S04 content of the waters dur1ng

chemical analysis. Thepresence of other species such as MgS04 or FeS04

are also known to lower the solubility of gypsum in high ionic strength

.
solutions (Tanji and Doneen,1966) •

Alternatively, gypsum solubility might be a function of Eh, pH and

ionic content of the waters, in much the same way as for ferrous iron,

and may not be explained in terms of simple solubility products.

The conclusions that can be drawn from this Section on thermodynamic

chemistry are four-fold.

In the first place, the elevated and constant temperature of the waters

emerging at Site 7 can be accounted for. Chemical reactions raise the

temperature some 2.50C above the normal groundwater value in this area

of 7°C. The importance of such reactions in the spontaneous combustion

of certain coal spoils is also noted.

Secondly, oxygenation of these waters causes supersaturation with

respect to ferric hydroxide and deposits of ochre form in the seepage

 135

areas. With time this freshly precipitated 'limonite' will convert

to the more stable forms goethite or hematite (see Section 7.6.3).

The precipitation of ferric hydroxide liberates H+ ions and greatly

acidifies the solutions. A natural lower limit on the acidity is imposed

by decreasing solubility of ferrous ions in high Eh environments and

the system is buffered at values around pH 3.0. This is in close agree-

ment with the observations made in Section 4.4.1.

Finally calculations of the solubility of gypsum appear to be

complicated by the presence of ionic CaS0

4
° or other species containing

calcium or sulphate. Gypsum does precipitate with increasing pH and

on evaporation which suggests the solution must be at least close to

saturation.
 til
s::
.~
0
01-1
CJ
til ,-..
...... ...... 00 C'1 -0
. .
Q)
~ S
......., C'1 11"'1 N -0

Q) 0 0 0 0
...... e- I I + I
.~
p.,
CJ
<l

.~
s::
~
~
Q) ii51
0
.
11"'1
0"1
N
N

-e-
......
CO
01-1
Q) 0
~ H
~
Q)
..c::
01-1
'CS
Q)
01-1

.~
s::
'CS
'CS
Q)
>..
til
.~
01-1

'CS
Q)
Q)
>..
0
~ .~
c,

>
......
0
~
01-1
til
CJ
Q)

0 01-1 Q) ~
til 'CS c,
~ Q)
H 'CS Q) Q)

til Q) .~ .~
01-1 01-1

Q)
00
s::
.~
01-1

~
S
0
..c::
......
01-1
Q)

CO >.. 0 0
c, 'CS 00
cS
~
;::I
...... 00 00 00
0
:> -0
0"1
g -0
......

. ....... .
11"'1 00 00
11"'1 -0 0"1

C'1 ...... 0
N

r:<l
~
~
H
 136

9. CHEMICAL WEATHERING AND SLOPE STABILITY

This Section considers some of the ways in which chemical weathering

could possibly influence the stability of the landslide. Mechanisms

involving the dissolution and precipitation of minerals, the alteration

of clays, and forces induced by repeated wetting and drying, and the

maintenance of failure surfaces will be discussed. Finally, the

importance of identifying the true cause of landslide movement will be

stressed with reference to past management policies at Mam Tor.

9.1 Dissolution and Precipitation

Quantities of various mineral phases involved in the weathering

sequence proposed in this study were calculated in Section 6.2. By

considering the specific gravities (relative densities) of these

minerals the volume changes involved in their formation or destruction

can be calculated by simple division.

Table 23 lists the amounts of pyrite, dolomite and goethite involved

together with average specific gravities (individual mineral specimens

can show a range of values) and approximate volume changes incurred.

These reactions result in a net porosity increase of 0.66 ml per

litre of percolating groundwater. With a calculated flow rate of 0.43

1
I sec- at Site 7 this means a net void creation of 0.89 m3 per year

accounted for by this stream alone. Volume changes induced by jarosite

precipitation and clay mineral transformations are not taken into account

as the effects are considered small in relation to the three dominant

processes of Table 23.

Dissolution of pyrite and dolomite is probably not occurring at the

same sites as secondary mineral precipitation. Once the various ions

have entered into solution migration through pore spaces and fissures can
 137

occur prior to secondary mineral growth. Other authors have also reported

the formation of alteration products of these and similar reactions some

distance from the source of the primary reactants (see Grattan-Bellew

and Eden, 1975; Nordstrom and Dagenhart, 1968).

This implies that internal stresses may be built up within the rocks

as voids are created in some areas and precipitating minerals create

pressures in others (particularly along joint planes). Individually

I
these internal stresses must be very slight, but the cumulative effect

might be great enough to influence the stability of the mass as a whole

or at least the most intensely weathered parts of it.

One of the most expansive weathering products of pyritic, calcareous

shales is the sulphate gypsum. Pressures induced by the precipitation and

growth of this mineral have been described previously (Section 1.2.3) in

an engineering context.

At Mam Tor gypsum is only found near the surface and is not thought

to form at depth. Nevertheless the nature of the surface accumulations,

in joints and on bedding planes (Plate 4b, 5 and 21) must contribute

to instability of the exposures particularly on steep slopes.

The appearance of this joint-filling gypsum is remarkably similar

to that which causes floor heave (Grattan-Bellew and Eden, 1975; Quigley

et al. 1973). Excellent scanning electron micrographs of gypsum crystals

prising apart shale laminations are included in the former article

(particularly that on p.375). The similarity between these pictures and

the photomicrographs of weathered Eda1e Shales shown in Plates 3 and 4a

is obvious. A further similarity is that Grattan-Bellew and Eden also

also observe crystal growth in cracks both normal and at steep angles

to bedding~similar to those shown in Plate 22b.

 138

9.2 The Mechanism of Shale Breakdown

It was noted in the Introduction (Section 1.2.3) that certain

authors believe purely chemical weathering processes to be of secondary

importance in the breakdown of mudrocks (see, for example, Spears and

Taylor, 1972). Essentially physical processes such as expansion in mixed-

layer clays are thought to be rate-determining and after an initial rapid

breakdown along joints and bedding planes further weathering (solution

and cation exchange) proceeds relatively slowly.

In the shale samples from Mam Tor the quantity of expansive clays

is generally small and only reaches significant proportions in the soil

layers (Section 7.4.2). It is thought, therefore, that a much more

effective process than clay mineral swelling might be the formation of

expansive minerals in the natural fissures and discontinuitiestt the rocks,

as described in Section 1.2.3 with reference to Goudie et al. (1970)

and Cooke and Smalley (1968). Initial swelling of the expandable clay

components,however, may create the necessary space for these minerals to

begin to precipitate.

Spears et al. (1971) observe the presence of gypsum in coal mine

spoil heaps and suggest that since it is a soluble mineral its occurrence

implies insignificant leaching. It was noted in Section 7.6.1 that

euhedral crystals of gypsum also persist in surface exposures of Eda1e

Shales at Mam Tor. Here,rapid leaching takes place during heavy rainfall,

since the material is so fragmented, but the exposures dry out rapidly

when the rain stops and it is thought that the overall rate of gypsum

precipitation (during evaporation and capillary rise of solutions through

the shale) exceeds removal by percolating waters. A surface coating of

iron oxide on the crystals, which appear red-brown, might also protect

them from dissolution.

Spears et al. (op.cit.) also note that pyrite is abundant~en 1n

material at the surface of the spoil heaps and they propose that this is
 139

further evidence of very slow weathering. P,yrite is also found in the

exposures of Edale Shales (see Appendix Ci), usually in close association

with carbonates which themselves have undergone intense weathering. Its

appearance at the surface is somewhat of a mystery since it might be

expected that over the years it would have been completely oxidised.

It is suggested that the persistence of pyrite could be due to the

absence of ThiobaciZZi from these surface environments, where the pH may

be too high to support them. Oxidation by atmospheric oxygen alone is

too slow to completely remove pyrite in the space of a few years.

The repeated wetting and drying thought to account for the persistence

of gypsum might in itself cause breakdown of shales, at least near to the

surface. Taylor and Spears (1970), for example, conduct slaking tests

on various Coal Measures rocks. They observe that breakdown is arrested

if the tests are carried out in vacuo.

The mechanism causing air breakage is explained by Taylor and Spears

as follows. During dry periods the shear resistance of rock fragments

is increased by high contact pressures caused by surface evaporation.

Any voids become filled with air which, when the water returns, as during

a heavy rainfall event, is pressurised due to capillary forces in the

outer pores. Failure along the weakest planes results, which further

opens up the material. Taylor and Spears note that this mechanism is

similar to that of soil slaking reported by Terzaghi and Peck (1967).

Since~fective breakdown in vacuo has only occasionally been reported

(see Badger et al. 1956; Berkovitch et al. 1959) air breakage of shales

is considered an important mechanism and may contribute to instability

at Mam Tor since oxygenated waters can readily gain access through the

major joint systems and failure surfaces.

Once again there is clearly scope for further work concerning the

breakdown mechanism of the Edale Shales. Engineering tests, which are

 140

not considered here, could also prove useful if representative samples

can be found and access gained to unweathered material.

9.3 Pore Water Pressure and Failure Surfaces

In surface exposures of weathered shales gypsum, jarosite and goethite

can be seen to form along bedding planes and in joints. By contrast

gypsum is not encountered in the borehole cores and jarosite is found

only infrequently, presumably due to the fact that it only precipitates

above the zone of groundwater saturation (Brown, 1971).

Goethite, however, is found at considerable depth within the land-

slide and although failure surfaces could not be positively identified

(Figure 24) it is thought that this mineral is occurring principally as

joint-fill or along surfaces of rupture. Hanshaw (1974) also observes

ferric hydroxide in fissures of old mine workings well below the zone of

aeration, illustrating that oxidation is possible so long as oxygenated

waters can gain access.

Failure surfaces are often developed during times of unusually high

pore water pressures resulting from a rise in the water table. Conditions

at the end of the Pleistocene glaciation were favourable for the initiation

of mass movements and many landslides in the Southern Pennines were

activated at this time, later becoming covered by end-Pleistocene soli-

fluction deposits (Said, 1969). Landslides resting on top of such soli-

fluction mantles, as in the case of Mam Tor, must be younger although a

more precise dating is often not possible (see Section 1.1.1).

At exactly what time and however the surface (or surfaces) of rupture

was established does not concern us here. Rather we have been considering

processes which can maintain such a surface on the verge of failure for

many hundreds of years, to be activated at regular intervals where conditions

are favourable, i.e. when the water table rises after the spring thaw or

during any 'freak' rainfall events.

 141

In order to predict the stability of slopes and the likelihood of

failure, Hodge and Freeze (1977) believe that information concerning

the nature and position of the water table is essential. From obser-

vations made in this study it would also appear that information concerning

groundwater chemistry might be equally important in certain geologic sit-

uations.

Subsequent to the drilling of the boreholes at Mam Tor piezometers

were placedm the ground by the County Council in order to monitor

fluctuations in ground water level. Results of these piezometric

investigations would be invaluable in assessing the effectiveness of

any drainage scheme but, in common with the pending report, they have

not been released.

It must be stressed once again that water in the failure zones does

not act as a lubricant but rather reduces the internal strength and shear

resistance of the rock mass. Changes in groundwater level are only a

trigger mechanism, as suggested by Nakano (1967) and graphically demon-

strated by Curtis and Williams (1978).

9.4 Future Management

Cooke and Doornkamp (1974) perhaps make the most pertinent comment

concerning landslide management, in view of the information obtained on

the Mam Tor slide. They state that "The physical control of landslides

is an engineering problem whose solution lies in the correct identification

of the causes of movement".

It is clear from the small amount of literature, mainly in the form

of newspaper articles, concerning Mam Tor that the true cause of movement

is not generally understood. For example, a typical article concerning

one period of rapid movement states that "Flood water has loosened the

sides of the 1700 ft. mountain" and goes on to describe the remedial

measures taken by County Council workmen who "dumped 300 tons of gravel

at the edge of the road to prevent further subsidence" (Anon, 1965).

 142

Equally futile and short-sighted maintenance work is described

in another article concerned with the January, 1966, movement which

states that "Teams of Council workmen have been working round the

clock to seal cracks (in the road) as they appear, to stop water getting

into them" (Anon, 1966 - my italics). Clearly the highway engineers did

not understand that the cracks are a manifestation of the movement and

not the real cause; although it is accepted that a certain amount of

oxygenated water will gain access to the failure surfaces in this way.

In 1977 it was commonly believed that heavy lorries using the Mam

Tor road (in particular lorries from the Hope Cement Works) were the cause

of repeated failure. After two years of closure to all but light traffic,

however, the latest and most dramatic damage occurred (Plates 16 to 19).

Once again the nature and scale of the true situation was not realised.

The drainage scheme proposed for Mam Tor (discussed in Section 3.3)

although on a very large scale would, in the opinion of the author, have

also proved unsuccessful in stabilising the slide. Apart from the supply

of deep groundwater irrespective of infiltration it has been shown in

several instances that the opening up of pyritic materials of this type

can actually increase the rate of weathering and, ultimately, the scale of

the problem. For example, Bloomfield (1972b) notes that considerably

more ochre than naturally precipitates in pyritic soils is deposited in

field drains installed specifically to alleviate the problem. Such drains

can become completely blocked by iron hydroxides and are rendered useless.

Although it is accepted that excessive pore water pressures can be

relieved by adequate drainage, in a chemical system so active as that at

Mam Tor other effects could be intensified. Not only would pyrite oxidation

speed up but also the products of reaction would be more efficiently trans-

ported away from the reaction sites and might not precipitate again within

the landslide. In this event the amount of void space created would be

greatly increased.
 143

Several million pounds have been spent over the years on repairs

to the Mam Tor road. These maintenance operations have been strictly short-

term, aimed at getting the road open again to traffic as soon as possible,

and none has had the slightest effect on long-term stability. Since the

landslide poses no direct threat to any major structures or community

(other than the road,and the barn in Plate 15) it is unlikely that any

effective long-term remedial measures will be taken. This is even more

evident in view of the recent decision to build a permanent alternative

route at a cost of at least 2~ to 3 million pounds.

Figure 38 Pyrite oxidation model.
 a'

_C S2 _ °2 1
1 PYRITE 1- - - - - ~I~I-.r: - - ~1~Fe2+1 I~I
I I
I I'
I I
! !
I~I
I
ThiObac~zzus~1
Ferroo~dans I~

__F_e
3
,-+_1
\1
1 so~-

d ~171/
1 Feoo-;-I-+-- __ ~ r-KF-e-3-(-SO-4-) 2-(-OH-)-6-

----+ Slow reaction

--_ Fast reaction

 144

10. FINAL CONCLUSIONS

The following is a brief account of the major conclusions that can

be drawn from the preceding chapters. Special emphasis is placed on

certain points but generally a more detailed account can be found at

the end of the relevant Section in the text.

Precipitates of ochre and gypsum in the seepages at Mam Tor suggest

that the composition of drainage waters is somewhat unusual. Detailed

analyses reveal that these streamwaters closely resemble so-called acid

mine drainage, the root cause of which is pyrite oxidation catalysed by

the presence of iron-oxidising bacteria.

Although the precise mechanistic steps are not considered in detail

the model illustrated in Figure 38 (similar to that proposed by Temple

and Delchamps, 1953) represents, schematically, the oxidation of pyrite

as thought to be occurring within the fractured Edale Shales at Mam Tor.

The pathways involving elemental sulphur and jarosite are added from van

Breemen (1972).

The rate-determining step in this model is step b, the oxidation of

ferrous iron, which is extremely slow under natural conditions. By contrast

the oxidation of pyrite by ferric iron can be up to 1000 times faster

(Garrels and Thompson, 1960).

Stumm and Morgan (1970) believe that the sequence is initiated by

the direct oxidation (a) or dissolution and oxidation (a') of pyrite.

The ferrous iron formed is slowly oxidised to ferric iron (b) which in

turn is rapidly reduced in the presence of pyrite (c) releasing acidity

and more Fe 2+ f or use 1n

. step b.
Once the sequence is started atmospheric

oxygen is only indirectly involved.

Step d, the precipitation of ferric hydroxide, appears to be incidental

Goet hi1te may, h owever, act as a reserV01r

. f or soluble Fe 3+ • If step b
 145

becomes insignificant or if the pH falls dramatically ferric iron may

be supplied by the dissolution of goethite. At lower pH jarosite may

. . 3+ h
become the dom~nant sol~d Fe p ase.

The significance of microbial activity cannot be over-emphasised.

Organisms act as catalysts in the oxidation of both ferrous iron and

sulphur and are, therefore, responsible for the rate at which the whole

model operates.
Many workers reporting on the weathering of sulphide-rich rocks refer

to acid production without realising the true scale of the process. Water

with a pH of about 3.0 is not so uncommon as these authors would have us

believe; for example Hargreaves, et al. (1975) report many examples of

highly acidic streams in England and Wales. The most interesting point

concerning the acid waters at Mam Tor, therefore, is not that their acidity

is unusually great but that the rate of production of that acid is so rapid.

It is noted that pyrite oxidation and the accompanying release of H+

ions can proceed very rapidly in the presence of catalytic bacteria. Sub-

sequent reactions involving carbonates and silicates which consume H+

ions and release others into solution must proceed at striatly aomparabZe

rates to produce waters whose chemical composition varies so little.

The fact that the vast majority of H+ ions (99% at least) are consumed

in these additional reactions has not been stressed by other researchers.

Van Breemen (1972) is one of the few who acknowledge this fact - he quotes

a figures of 99.7% of the acid produced by pyrite oxidation being in-

activated by dissolved (carbonate) alkalinity or consumed by mineral

weathering.
The residual 1% acidity is sufficient to give the waters a pH of

about 3 to 3.5. Thermodynamic calculations suggest that this buffering

is a function of decreasing Fe2+ solubility as ferric hydroxide precipitates

and the redox potential rises. The effectiveness of this mechanism can be
 146

seen from the great many values of Appendix Bii, which fall in this

narrow pH range.

Wider consideration of the results of chemical analyses indicates

that different streams respond differently to changes in climatic vari-

ables. By using cluster analysis - a technique of hierarchical classi-

fication - the streams can be split into a number of distinct 'groups'

which derive their waters from different sources. Some are merely surface

runoff streams which have a composition close to that of rainfall while

others obtain their water from deep underground after a long residence

time in contact with pyritic shales. The majority of sites demonstrate

mixing of waters from both sources and as a result show marked dilution

when the surface contribution is large.

A series of chemical reactions, loosely termed a 'weathering model',

is suggested to account for the composition of waters emerging at one

of the spring sites - Site number 7 - which is least diluted by surface

runoff. Quantitative values can be fitted at each step in this model

and the resulting composition of the water can be explained. X-ray

analysis of the solid materials generally confirms the accuracy of the

weathering model, with a few minor qualifications for jarosite precipit-

ation.

Analysis of the precipitate minerals indicates that the first-formed

phase 'limonite' gradually converts to the more stable hydroxide goethite.

Jarosite precipitation appears to be restricted to sites above the zone

of permanent water saturation, and gypsum is only found at the surface.

The chemical alteration of clay minerals does not appear to be res-

ponsible for shale breakdown; only minor amounts of cations are derived

in this way. Surface weathering occurs principally as a result of the

growth of alteration products on bedding planes and in joints. At depth

such precipitation is still possible where aerated waters can penetrate

 147

the strata but processes of clay swelling (expandable fraction only)

and air breakage are also thought to be important. Long term chemical

alteration would appear to proceed much more slowly than the initially

rapid breakdown into angular fragments. Internal stresses produced by

solution and precipitation of minerals may be important locally.

Large scale and often rapid landslide movements occur periodically

at Mam Tor and appear to be triggered by changes in pore water pressures.

Such movements take place along pre-determined failure surfaces which

are maintained by deep chemical weathering.

The correlation between physical instability and extreme chemical

instability at this site is fascinating. The area is a vast 'geochemical

factory' ,and affords many excellent examples for teaching purposes; the

reactions can be 'seen' to be taking place. This thesis has done no more

than demonstrate and carefully document these chemical changes which could

conceivably contribute to the maintenance of instability; there is no

question of being able to link them with the original failure.

It is sincerely hoped that this research will prompt further work

into the links between chemical weathering and slope stability.

THE END
 148

11. REFERENCES

ALLEN, J.R.L. 1960. The Mam Tor Sandstones: A turbidite facies of

the Namurian Deltas of Derbyshire. Journ. Sed. Pet. 30, 193-208.

ALLEN, S.E. (ed.) 1974. Chemical Analysis of Ecological Materials.

Blackwell Scientific Publications, Oxford.

ALLEN, S.E., CARLISLE, A., WHITE, E.J. and EVANS, C.C. 1968. The plant

nutrient content of rainwater. J. Eco1. ~, 497-504.

AMIN, M.A. 1979. Geochemistry and mineralogy of Namurian sediments

in the Pennine Basin, England. Ph.D. Thesis, Univ. Sheffield.

ANDERSON, W.C. and YOUNGSTROM, M.P. 1976. Coal pile leachate quantity

and quality characteristics. 6th Symposium on Coal Mine Drainage

Research. NCA/BCR Coal Conference, Kentucky, 17-33.

ANDERSSON, A. and WIKLANDER, L. 1975. Release of crystal constituents

by chemical weathering of some soil minerals. Soil Sci., 120, 13-19.

ANON. 1960. Structures don't settle in this shale, but watch out for

heave. Engineering News-Record, 164, 46-48.

ANON. 1965. Mountain blamed for road closure. Article appearing in

The Sheffield Morning Telegraph, 30th December, 1965.

ANON. 1966. Mam Tor Road Closed. Article appearing 1n The Star

(Sheffield Newspapers Ltd.), 17th January, 1966.

APPALACHIAN REGIONAL COMMISSION. 1969. Acid Mine Drainage in Appalachia.

Washington, D.C.

ASHMEED, D. 1955. The influence of bacteria on the formation of acid

mine waters. Colliery Guardian, 190, 694-698.

AUSTIN, G.C. and LEININGER, R.K. 1976. The effect of heat treating

sedimented mixed layer illite-smectite as related to quantitative

clay mineral determinations. Journ. Sed. Pet., 46, 206-215.

 149

BAAS BECKING, L.G.M., KAPLAN, l.R. and MOORE, D. 1960. Limits of the

natural environment in terms of pH and oxidation-reduction potential.

J. Geol., ~, 243-284.

BADGER, C.W., CUMMINGS, A.D. and WHITMORE, R.L. 1956. The disintegration

of shales in water. J. Inst. Fuel. ~, 417-423.

BANERJEE, A.C. 1971. Mechanism of oxidation of iron pyrites. Chern.

Commun. 1006-1007.

BARNES, H.L. and ROMBERGER, S.B. 1968. Chemical aspects of acid mine

drainage. Jour. Water Pollution Control Federation, 40, 371-384.

BARNES, I. and BACK, W. (1964). Geochemistry of iron rich groundwater

of southern Maryland. J. Geol., ~, 435-447.

BARNES, I., STUART, W.T. and FISHER, D.W. 1964. Field investigations of

mine waters in the Northern Anthracite Field, Pennsylvania. U.S.

Geol. Surv., Prof. Pap. 473-B, 8 pp.

BARNHISEL, R.I. and ROTROMEL, A.L. 1974. Weathering of clay minerals

by simulated acid coal spoil bank solutions. Soil Sci. 118, 22 pp.

BERKOVITCH, I., MANACKERMAN, M. and POTTER, N.M. 1959. The shale break-

down problem in coal washing. Part 1 - Assessing the breakdown of

shales in water. J. Inst. Fuel. _:g, 579-589.

BERNER, R.A. 1964. Iron sulfides formed from aqueous solution at low

temperatures and atmospheric pressure. Jour. Geol. 1l, 293-306.

BERNER, R.A. 1969. Goethite stability and the origin of red beds.

Geochim. et Cosmochim Acta, 11, 267-273.

BERNER, R.A. 1970. Sedimentary pyrite formation. Amer. Jour. Sci.,

268, 1-23.

BIESECKER, J.E. and GEORGE, J.R. 1966. Stream quality in Appalachia

as related to coal mine drainage. U.S. Geol. Surv. Circ., 526,

Washington, D.C.
 150

BISHOP, A.W., HUTCHINSON, J.N., PENMAN, A.D. and EVANS, H.E. 1969.

Geotechnical investigation into the causes and circumstances of

the disaster of 21st October, 1966. Item 1, I-SO. A selection

of technical reports submitted to the Aberfan Tribunal. Welsh

Office, London. HMSO.

BLADH, K. 1978. Formation of goethite, jarosite and alunite during

250C weathering of felsic rocks containing pyrite and chalcopyrite.

Geol. Soc. Am. Abstracts with Programs, 10, p.368.

BLATT, H., MIDDLETON, G. and MURRAY, R. 1972. Origin of Sedimentary

Rocks. Prentice Hall, New Jersey, 634 pp.

BLOOMFIELD, C. 1972(a). The oxidation of iron sulfides in soils in

relation to the formation of acid sulphate soils and of ochre

deposits in field drains. J. Soil. Sci., ~, 1-16.

BLOOMFIELD, C. 1972(b). Acidification and ochre formation in pyritic

soils. Proc. Int. Symp. Acid Su1ph. Soils, Wageningen.

BOWEN, N.L., 1928. The evolution of the igneous rocks. Princeton Univ.

Press, Princeton, N.J.

BRAGG, W.L. 1913. X-ray diffraction and crystal structures. Proc. Roy.

Soc. A., 89, 248-277.

van BREEMEN, N. 1972. Soil forming processes in acid sulphate soils.

In: H. Dost (ed.), Acid Sulphate Soils. Proc. Int. Symp. on Acid

Sulphate Soils, Wageningen, Netherlands, 1972.

BRICKER, O.P., GODFREY, A.E. and CLEAVES, E.T., 1968. Mineral water

interaction during the chemical weathering of silicates. Am. Chern.

Soc. Advances, Chemistry Sere 73, 128-142.

BRINDLEY, G.W. 1955. Identification of clay minerals by X-ray

diffraction analysis. In: Pask and Turner, 1955.

BRITISH STANDARDS INSTITUTION, 1964. Methods of measurement of liquid

flow in open channels. Part 2A - Dilution Method, constant rate

injection. B.S. 3680, PE2A, BSI, London, 39p.

 151

BRITISH STANDARDS INSTITUTION, 1965. Methods of measurement of liquid

flow in open channels. B.S. 3680 4A.

BROCK, T.D. 1969. Microbial growth under extreme conditions. Symp.

Soc. Gen. Microbio. ~, 15-41.

BROPHY, G.P., SCOTT, E.S. and SNELLGROVE, R.A. 1962. Sulfate Studies II.

Solid solution between alunite and jarosite. Am. Min. ~, 112-126.

BROWN, G. 1961. The X-ray Identification and Crystal Structures of Clay

Minerals. Mineralogical Society, London.

BROWN, J.B. 1970. A chemical study of some synthetic potassium-hydronium

jarosites. Can. Mineral., 48, 703-706.

BROWN, J.B. 1971. Jarosite-goethite stabilities at 250C, 1 atm. Mineral

Deposits (Berl.), i, 245-252.

BROWN, R.D. 1977 •. Excursion itineraries for the 6th British Polish

Seminar, Sheffield 1977. Sheffield University, Geography Department.

BRYNER, L.C. and JAMERSON, A.K. 1958. Micro-organisms in leaching

sulphide minerals. Appl. Microbiol. i, 281-287.

CARPENTER, R.H. and HAYES, W.B. 1980. Annual accretion of Fe-Mo oxides

and certain associated metals in a stream environment. Chem. Geol.

~, 249-260.

CARUCCIO, F.T., GEIDEL, G. and SEWELL, J.M. 1976. The character of

drainage as a function of the occurrence of framboidal pyrite and

groundwater quality in eastern Kentucky. 6th Sumposium on Coal

Mine Drainage Research. NCA/BCR Coal Conference, Kentucky, 1-16,

CHAO, G.Y. 1969. 29(Cu) table for common minerals. Carleton Univ.,

Department of Geology, Geological Paper, 69-2, Ottawa, Canada.

CHRISTIAN, G.D. and FELDMAN, F.J. 1970. Atomic Absorption Spectroscopy.

Applications in Agriculture, Biology and Medicine. Wiley-Interscience,

New York, 490 pp.

 152

CLARK, W.M. and COHEN, B. 1923. An analysis of the theoretical relations

between reduction potentials and pH. Public Health Reps. Repr., 826.

CLEAVES, E.T., GODFREY, A.E. and BRICKER, O.P. 1970. Geochemical balance

of a small watershed and its geomorphic implications. Geol. Soc.

Am. Bull. ~, 3015-3032.

COLLINSON, J.D. and WALKER, R.G. 1967. Namurian sedimentation in the High

Peak. In: Geol. Excursions in the Sheffield Region. C. Downie and

R. Neves (Eds) Univ. Sheffield, 1967, 163 p.

COLMER, E.R. and HINKLE, M.E. 1947. The role of micro-organisms in acid

mine drainage. Science. 106, 253 p.

COOKE, R.U. and DOORNKAMP, J.C. 1974. Geomorphology in Environmental

Management. Oxford University Press, 413 p.

COOKE, R.U. and SMALLEY, I.J. 1968. Salt weathering in deserts. Nature,

220, 1226-1227.

COOMBS, D. 1971. Mam Tor: A Bronze Age Hillfort? C. Arch. ~, 100-102.

CORRENS, C.W. 1961. The experimental chemical weathering of silicates.

Clay Minerals Bull. i, 249-265.

CORRENS, C.W. 1963. Experiments on the decomposition of silicates and

discussion of chemical weathering. Clays Clay Minerals, Proc. Natl.

Conf. Clays Clay Minerals, 10, 443-59. Pergamon Press, New York.

COVENEY, R.M. and PARIZEK, E.J. 1977. Deformation of mine floors by

sulfide alteration. Bull. Assoc. Eng. Geol. ~, 131-156.

CRISP, D.T. 1966. Input and output of minerals for an area of Pennine

moorland: The importance of precipitation, drainage, peat erosion and

animals. J. Appl. Econ. 2, 327-348.

CROUSE, H.L. and ROSE, A.W. 1976. Natural beneficiation of acid mine drainage

by interaction of stream water with stream sediment. Proc. 6th Symp.

Coal Mine Drainage Res., Kentucky, 237-269.

CRUICKSHANK, J.G. 1972. Soil Geography. David and Charles Ltd., Newton

Abbot, 256 pp.

 153

CRYER, R. 1976(a) A study of the sources and variations of major solutes

in selected mid-Wales catchments. Ph.D. Thesis, University of Wales,

1978.

CRYER, R. 1976(b). The significance and variation of atmospheric nutrient

inputs in a small catchment system. J. Hydrol. 29, 121-137.

CUBITT, J.M. 1975. A regression technique for the analysis of shales by

X-ray diffraction. Jour. Sed. Pet. 45, 546-553.

CURTIS, C.D. 1975. Chemistry of rock weathering fundamental reactions

and controls. In: Geomorphology and Climate. E. Derbyshire (ed.)

Wiley, London.

CURTIS, C.D. 1976. Stability of minerals in surface weathering reactions:

A general thermochemical approach. Earth Surface Processes, ~

63-70.

CURTIS, G.D. and WILLIAMS, J.W. 1978. Dynamic demonstration of pore

pressure's role in slope instability. J. Geol. Educ., 26, 111-113.

DEPARTMENT OF THE ENVIRONMENT 1972. Determination of Sulphate. In:

Analysis of raw, potable and waste waters. HMSO, London, 78-79.

DEUL, M. 1974. Limestone for controlling acid mine drainage and for

treatment of acid mine water. 10th Forum Geol. Ind. Miner. (Columbus

USA), 43-46.

DOUGHERTY, M.T. and BARSOTTI, N.J. 1972. Structural damage and potentially

expansive sulfide minerals. Bull. Assoc. Eng. Geol. ~, 105-125.

DUGAN, P.R. 1972. Biochemical Ecology of Water Pollution. Plenum

Publishing Carp, New York, 159 pp.

DUNN, A. 1979. Mam Tor to get by-pass. Article appearing in The Guardian,

13th February, 1979.

EARLY, K.R. and SKEMPTON, A.W. 1972. Investigations of the landslide at

Walton's Wood, Staffordshire. Q.J1. Eng. Geo1., 1, 19-41.

 154

EDEN, R.A., ORME, G.R., MITCHELL, M. and SHIRLEY, J. 1962. A study of

part of the margin of the Carboniferous limestone massif in the Pin

Dale area, Derbyshire. Bull. Geol. Surv. ~, 73-118.

EMMONS, W.H. 1917. The enrichment of ore deposits. U.S. Geol. Surv. Bull.

625, 530 p.

FASSISKA, E.J., WAGENBLAST, H. and DOUGHERTY, M.T. 1974. The oxidation

mechanism of sulfide minerals. Bull. Assoc. Engineering Geologists,

!!, 75-82.

FORBES, H. 1947. Landslide investigation and correction. Am. Soc. Siv.

Eng., Tr. 112, 377-442.

FORD, T.D. 1976. Geochemical prospecting over Odin Rake. Chapter 6.

In: Ford T.D. and Rieuwerts, J.H., 1976. Bull. Peak Dist. Mines

Hist. Soc. ~, No.4.

FORD, T.D. and RIEUWERTS, J.H. 1975. Lead Mining in the Peak District.

Peak Park Joint Planning Board, Bakewell.

FORD, T.D. and RIEUWERTS, J.H. 1976. Odin Mine, Castleton, Derbyshire.

Bull. Peak Dist. Mines Hist. Soc., ~, No.4.

GARRELS, R.M. and CHRIST, C.L. 1965. Solutions, Minerals and Equilibria.

Harper and Row, New York, 450 p.

GARRELS, R.M. and NAESER, C.R. 1958. Equilibrium distribution of dissolved

sulphur species in water at 250C and one atmosphere total pressure.

Geochim et Cosmochim. Acta li, 113-130.

GARRELS, R.M. and THOMPSON, M.E. 1960. Oxidation of pyrite by iron

sulfate solutions. Am. J. Sci. 258A, 57-67.

GIBBS, R.J. 1965. Error due to segregation in quantitative clay mineral

X-ray diffraction mounting techniques. Am. Min. 50, 741-751.

GIBBS, R.J. 1968. Clay mineral mounting techniques for X-ray diffraction

analysis. Jour. Sed. Pet., 38, 242-243.

GOLDICH, S.S. 1938. A study in rock weathering. J. Geol. 46, 17-58.

 155

GOLTERMAN, H.L. 1975. Physiological Linmology. Elsevier, New York, 489 p.

GOLTERMAN, H.L., CLYMO, R.S. and OHNSTAD, M.A.M. 1978. Methods of physical

and chemical analysis of fresh waters (2nd Ed.). International Biological

Programme Handbook No.8.

GORE, A.J.P. 1968. The supply of six elements by rain to an upland peat

area. J. Ecol. 56, 483-495.

GOUDIE A., COOKE R.U. and SMALLEY, I. 1970. Experimental investigation

of rock weathering by salts. Area. 42-48.

GRAF, W.H. and ACAROGLU, E.R. 1966. Settling velocities of natural grains.

Bull. Intern. Assoc. Sci. Hydrology, ll, 27-43.

GRATTAN-BELLEW, P.E. and EDEN, W.J. 1975. Concrete deterioration and floor

heave due to biogeochemical weathering of underlying shale. Can.

Geotech. J., 11, 372-378.

GREGORY, S. 1963. Statistical methods and the geographer. (3rd Ed., 1973).

Longman, London, 271 p.

GRIM, R.E. 1968. Clay Mineralogy (2nd Ed.) McGraw Hill (International

Series In Earth and Planetary Sciences).

GUNEY, M. 1968. Oxidation and spontaneous combustion of coal: Review of

individual factors. Colliery Guardian, February, 1968, 137-143.

GUTSCHICK, K.A. 1974. Lime and limestone. Mining Engineering, 26,

113-114.

HANSHAW, B. 1974. Geochemical evolution of a goethite deposit. In:

Proc. International Symposium on Water-Rock Interaction, 70-75.

HARGREAVES, J.W., LLOYD, E.J.H. and WHITTON, B.A. 1975. Chemistry and

vegetation of highly acidic streams. Freshwat. BioI. ~, 563-576.

'HARRER, C.M. and TESCH, W.J. 1959. Reconnaissance of iron occurrences

in Colorado. U.S. Bureau of Mines, Information Circ. 7918.

 156

HARRISON, A.D. 1958. The effects of sulphuric acid pollution on the

biology of streams in the Transvaal, South Africa. Verh. Int.

Verein. theor. angew. Limnol., 13, 603-610.

HEM, J.D. 1960(a). Restraints on dissolved ferrous iron imposed by

bicarbonate, redox potential and pH. U.S. Geol. Surv. Water-Supply

Paper 1459-B, 33-55.

HEM, J.D. 1960(b). Some chemical relationships among sulphur species

and dissolved ferrous iron. U.S. Geol. Surv. Water Supply Paper 1459,

57-73.

HEM, J.D. 1961. Stability field diagrams as aids in iron chemistry studies.

Amer. Water Works Assoc. J. 53.

HEM, J.D. 1968. Graphical methods for studies of aqueous aluminium hydroxide

fluoride and sulphate complexes. U.S. Geol. Surv. Water Supply Paper

l827B.

HEM, J.D. 1970. Study and interpretation of the chemical characteristics of

natural water. U.S. Geol. Surv. Water Supply Paper 1473.

HEM, J.D. 1972. Chemical factors that influence the availability of iron

and manganese in aqueous systems. Bull. Geo1. Soc. Amer., 83, 443-450.

HEM, J.D. 1977. Reactions of metal ions at surfaces of hydrous iron oxide.

Geochim. et Cosmochim. Acta, il, 527-538.

HEM, J.D. and CROPPER, W.H. 1959. A survey of ferric-ferrous chemical

equilibria and redox potential. U.S. Geol. Surv. Water Supply Paper

1459A, 1-31.

HILL, R.D. 1974. Overview of use of carbonate rocks for controlling acid

mine drainage. 10th Forum, Geo1. Ind. Miner. Columbus, USA) 25-29.

HILL, R.D. and WILMOTH, R.C. 1971. Limestone treatment of acid mine

drainage. Trans. Soc. Min. Eng. 250, 162-166.

HODGE, R.A.L. and FREEZE, R.A. 1977. Groundwater flow systems and slope

stability. Can. Geotech. J., ~, 466-476.

 157

HOLLAND, H.D. 1979. Metals in Black Shales: A reassessment. Econ. Geol.

~, 1676-1680.

HUDSON, R.G.S. and COTTON, G. 1945. The Carboniferous rocks of the Edale

Anticline, Derbyshire. Quart. Jour. Geol. Soc. Lond., 101, 1-36.

HUTCHINSON, J.N. 1968. Mass movement. In: Fairbridge, R.W. (Ed.),

Encyclopedia of Geomorphology.

JACKSON, M.L. and SHERMAN, G.D. 1953. Chemical weathering of minerals in

soils. Advances in Agronomy, 2, Academic Press, N.Y., 219-318.

JACKSON, W. 1927. The succession below the Kinderscout Grit in North

Derbyshire. Manchester Geol. Soc. Jour. l, 15-32.

JENNE, E.A. 1968. Controls on Mn, Fe, Co, Ni, Cu, Zn concentrations in

soils and water: The significant role of hydrous Mn and Fe oxides.

in: Gould, R.F. (ed), Trace inorganics in water. Am. Chern. Soc.,

Advances in Chemistry Series, 21, 337-387.

JOHNSON, L.J., MATELSKI, R.P. and ENGLE, C.F. 1963. Clay mineral character-

isation of modal soil profiles in several Pennsylvanian counties. Proc.

Soil Sic. Soc. Am., 12, 568-572.

JOHNSON, N.M., LIKENS, G.E., BORMANN, F.H., FISHER, D.W., and PIERCE, R.S.

1969. A working model for the variation in stream water chemistry at

the Hubbard Brook Experimental Forest, New Hampshire. Water Resources

Research, 2, 1353-1363.

JOHNSTON, R.J. 1978. Multivariate Statistical Analysis in Geography.

Longman, London. 280 pp.

KAPLAN, I.R., EMERY, K.O. and RITTENBERG, S.C. 1963. The distribution and

isotopic abundance of sulphur in recent marine sediments off southern

California. Geochim. et Cosmochim. Acta, 12, 297-332.

KELLY, W.C. and GODDARD; E.N. 1969. Telluride ores of Boulder County,

Colorado. Geol. Soc. America Memoir, 109, 237 p.

 158

KHLYBOV, V.V. 1976. Yarozit iz zony okis1eniya piritizirovanykh porod

Zapadnogo Piritman ya (Jarosite from the oxidation zone of pyritised

rocks in the West Timan Region). In: Minera10giya rudnykh mestorochdeniy

Severa Ura1a i Pay-Khoya (Yushkin, N.P. - Ed.). Akad. Nauk SSSR,

Komi Fi1. Inst. Geo1. Tr, 20, 65-67.

KIELLAND, J. 1937. Individual activity coefficients of ions in aqueous

solutions. Am. Chem. Soc. Jour. ~, 1675-1678.

KINTER, E.B. and DIAMOND, S. 1956. A new method for preparation and treatment

of'oriented-aggregate specimens of soil clays for X-ray diffraction.

Soil Sci. ~, 111-120.

KITANO, Y., KATO, K., KANAMORI, S., KANAMORI, N. and YOSHIOKA, R., 1967.

Rockslides resulting from the geochemical weathering of parent

materials. Annals, Disaster Prevention, Res. Inst. Kyoto Univ. lOA,

557-587 (in Japanese).

KRAUSKOPF, K.B. 1967. Introduction to Geochemistry. McGraw-Hill, New

York, 721 p ,

KUBASCHEWSKI, 0. and EVANS, E.L. 1958. Metallurgical Thermochemistry. 3rd

Ed., Pergamon Press, New York, 226-343.

KUZNETSOV, S.I., IVANOV, M.V. and LYALIKOVA, N. 1963. Introduction to

Geological Microbiology. English Edition - C.H. Oppenheimer (Ed.),

McGraw-Hill, New York, 252 pp.

LACKEY, J.B. 1938. The fauna and flora of surface waters polluted by acid

mine drainage. U.S. Pub1. H1th. Rep., 53, 1499-1507.

LAGUROS, J.G., KUMAR, S. and ANNAMALA, M. 1974. Comparative study of

simulated and natural weathering of some Oklahoma shales. Clay, Clay

Min., 22, 111-115.

LANGMUIR, D. 1965. Stability of carbonates in the system MgO-C0 -H 0. Jour.

2 2
Geology, 73, 730-754.

LATIMER, W.M. 1952. Oxidation Potentials. Prentice Hall, New Jersey, 392 p.
 159

LAU, C.M., SHUMATE, K.S. and SMITH, E.E. 1970. The role of bacteria in

pyrite oxidation kinetics. Proc. 3rd Symp. Coal Mine Drainage

Res., Mellon Institute,Pittsburgh, Pa., 114-122.

LAZAROFF, N. 1963. Sulphate requirements for iron oxidation by Thiobacillus

ferrooxidans. J. Bacteriol. ~, 78-83.

LEATHEN, W.W., KINSEL, N. and BRALEY, S.A. 1956. Ferrobacillus ferrooxidans -

A chemosynthetic autotrophic bacterium. J. Bateriol. ~, No.5.

LEE, G.F. and HOADLEY, A.W. 1967. Biological activity in relation to the

chemical equilibrium composition of natural waters. In: Equilibrium

Concepts in Natural Water Systems. Am. Chern. Soc. Advances in Chemistry

Ser., ~, 319-338.

LIND, C.J. 1970. Specific conductance as a means of estimating ionic

strength. USGS Prof. Paper 700-D, D272-D280.

LOUGHNAN, F.C. 1962. Some considerations in the weathering of silicate

minerals. Jour. Sed. Pet., ~, 284-290.

LOVELL, H.L. and LACHMAN, R.I. 1970. An investigation of the natural

beneficiation of coal mine drainage. Special Research Report, SR-76,

Coal Research Section, Pennsylvania State University, 187 pp.

MAPSTONE, G.E. 1954. The weathering of pyrite. Chern. Ind. 577-578.

MASON, M.H. 1961. The Mam Tor sandstones: A turbidite facies of the

Namurian of North Derbyshire. Ph.D. Thesis, University of Manchester.

McARTHUR, J.L. 1970. A geomorphological study of the Upper Derwent Basin,

Derbyshire. Ph.D. Thesis, University of Sheffield.

MIELENTZ, R.C. and KING, M.E. 1955. Physical-chemical properties and

engineering performance of clays. Bull. Div. Mines Calif., 169, 196-254.

NAMBU, M. 1957. Hydrated Fe-oxides. 1. Hydrated ferric oxide minerals

altered from pyrite by weathering. Sci. Rep. Res. Inst., Tokoku

Univ., Sere A, ~, 215-226.

 160

NAKANO, R. 1967. On weathering and change of properties of Tertiary

mudstones related to landslide. Soil and Fdn., I, 1-14.

NONVEILLER, E. 1967. Shear strength of bedded and jointed rocks as

determined from the Sales ina and Vajont Slides. Proc. Geotechnical

Conference, Oslo 1967. Norwegian Geotech. Inst., 289-294.

NORDSTROM, D.K. and DAGENHART, T.V. 1978. Hydrated iron sulfate minerals

associated with pyrite oxidation. Field relations and thermodynamic

properties. Geo1. Soc. Am. Abstracts with programs, 10, p.464.

OBORN, E.T. 1960. A survey of pertinent biochemical literature. U.S.

Geo1. Surv. Water Supply Paper 1459F, 111-190.

OBORN, E.T. and HEM, J.D. 1962. Some effects of the larger types of

aquatic vegetation on iron content of water. U.S. Geo1. Surv. Water

Supply Paper 14591, 237-268.

OHIO STATE UNIVERSITY RESEARCH FOUNDATION 1971. Acid mine drainage formation

and abatement. Washington (US) Environmental Protection Agency Grant

14010, FPR, DAST-42, 82 p.

PEAK PARK JOINT PLANNING BOARD 1977. Mam Tor: The Alternatives. Peak

Park News (Journal of the Peak National Park), Autumn, 1977, p.4.

PERKIN-ELMER 1976. Analytical methods for atomic absorption spectrophotometry.

Perkin Elmer, September 1976.

PIERCE, J.W. and SIEGEL, F.R. 1969. Quantification in clay mineral studies

and sedimentary rocks. Jour. Sed. Set., ~, 187-193.

PITEAU, D.R. 1970. Geological factors significant to the stability of

slopes cut in rock. In: van Rensburg, P.W.J. (Ed.), Planning Open

Pit Mines. S. Afr. Inst. Min. Met.

PITTY, A.F. 1976. Water temperatures in the limestone areas of the central

and southern Pennines. Proc. Yorks. Geo1. Soc., 40,601-612.

 161

PITTY, A.F., HALLIWELL, R.A. and TERNAN, J.L. 1979. The range of water

temperature fluctuations in the limestone waters of the central and

southern Pennines. J. Hydrol., 41, 157-160.

POTTER, R.W. 1976. The weathering of sulfide ores in Shasta County,

California and its relationship to pollution associated with acid

mine drainage. Open File Report, U.S. Geol. Surv. ~, 18 p.

POURBAIX, M.J.N. 1949. Thermodynamics of Dilute Aqueous Solutions.

Edward Arnold & Co., London, 136 pp.

QUIGLEY, R.M. and VOGAN, R.W. 1970. Black shale heaving at Ottawa,

Canada. Canadian Geotechnical Journal, l, 106-115.

QUIGLEY, R.M., ZAJIC, J.E., McKYES, E. and YONG, R.N. 1973. Biochemical

alteration and heave of black shale: Detailed observations and inter-

pretations. Can. J. Earth Sci., 10, 1005-1006.

QUISPEL, A., HARMSEN, G.W. and OTZEN, D. 1952. Contribution to the chemical

and bacteriological oxidation of pyrite in soils. Pl. and Soil, ~,

43-55.

RAE, G. 1977. Computer analysis of information on acid mine-drainage

waters. In: Optimatl Development and Management of Groundwater

(Buchan, S., Chairperson). Int. Assoc. Hydrogeol. Mem., 11, B39-B49.

RAE, G. 1978. Mine drainage from coalfields in England and Wales: A

summary of its distribution and relationship to water resources.

G.B. Cent. Water Plann. Unit., Tech. Note, No.24, 21 pp.

RAINWATER, F.H. and THATCHER, L.L. 1960. Methods for collection and

analysis of water samples. A.S. Geol. Surv. Water Supply Paper, 1454.

READ, H.H. 1970. Rutley's Elements of Mineralogy. 26th Ed., Allen and

Unwin, London, 560 pp.

READING, H.G. 1964. A review of factors affecting the sedimentation of

the Millstone Grit in the Southern Pennines. In: Deltaic and Shallow

Marine Deposits. L.M.J.V. van Straaten (Ed.), Elsevier, Amsterdam, 340-346.

 162

REED, C. 1977. An article concerning alternative routes to the Mam Tor

road. In: The Morning Telegraph (Sheffield Newspapers Ltd.), June

10th, 1977.

RICCA, V.T. and CHOW, K. 1974. Acid mine drainage quantity and quality

generation model. Trans. Soc. Min. Eng. AlME (New York), 256, 328-336.

RICKARD, D.T. 1969. The chemistry of iron sulphide formation at low

temperatures. Stockholm Contr. Geology, 20, 67-95.

RIPPON, J.E. 1980. Studies of acid rain on soils and catchments. NATO

Conference Series No.1, Ecology, ~, 499-

ROBIE, R.A. 1959. Thermodynamic properties of selected minerals and oxides

at high temperatures. U.S. Geol. Surv. Trace Elements Inv. Rept., 609.

ROBIE, R.A. and WALDBAUM, D.R. 1968. Thermodynamic properties of minerals

o
and related substances at 298.15 K (25 C) and one atmosphere (1.013

bars) pressure and at higher temperatures. U.S. Geol. Surv. Bull., 1259,

256 p.

RORISON, I.H. 1973. The effect of extreme soil acidity on the nutrient

uptake and physiology of plants. Acid Sulphate Soils, Proc. Int.

Symp. ~.

ROSSINI, F.D., WAGMAN, D.D., EVANS, W.H., LEVINE, S. and JAFFE, I. 1952.

Selected values of chemical thermodynamic properties. Natl. Bur.

Standards Circ. 500, U.S. Dept. Commerce.

ROYSTER, D.L. 1979. Landslide remedial measures. Assoc. Eng. Geo1.

Bull., 16, 301-352.

RUTTNER, F. 1953. Fundamentals of Limnology. University of Toronto Press,

Toronto.

SAAS, E., NATHAN, Y. and NISSENBAUM, A. 1965. Mineralogy of certain

pyrite concretions from Israel and their alteration products. Min.

Mag., 35, 84-87.

SAID, M. 1969. The Pleistocene geomorphology of the Burbage Basin.

Ph.D. Thesis, University of Sheffield.

 162

REED, C. 1977. An article concerning alternative routes to the Mam Tor

road. In: The Morning Telegraph (Sheffield Newspapers Ltd.), June

lOth, 1977.

RICCA, V.T. and CHOW, K. 1974. Acid mine drainage quantity and quality

generation model. Trans. Soc. Min. Eng. AlME (New York), 256, 328-336.

RICKARD, D.T. 1969. The chemistry of iron sulphide formation at low

temperatures. Stockholm Contr. Geology, 20, 67-95.

RIPPON, J.E. 1980. Studies of acid rain on soils and catchments. NATO

Conference Series No.1, Ecology, i, 499-

ROBIE, R.A. 1959. Thermodynamic properties of selected minerals and oxides

at high temperatures. U.S. Geol. Surv. Trace Elements Inv. Rept., 609.

ROBIE, R.A. and WALDBAUM, D.R. 1968. Thermodynamic properties of minerals

and related substances at 298.15 K (2SoC) and one atmosphere (1.013

bars) pressure and at higher temperatures. U.S. Geol. Surv. Bull., 1259,

256 p.

RORISON, I.H. 1973. The effect of extreme soil acidity on the nutrient

uptake and physiology of plants. Acid Sulphate Soils, Proc. Int.

Symp. ~.

ROSSINI, F.D., WAGMAN, D.D., EVANS, W.R., LEVINE, S. and JAFFE, I. 1952.

Selected values of chemical thermodynamic properties. Natl. Bur.

Standards Circ. 500, U.S. Dept. Commerce.

ROYSTER, D.L. 1979. Landslide remedial measures. Assoc. Eng. Geol.

Bull., ~, 301-352.

RUTTNER, F. 1953. Fundamentals of Limnology. University of Toronto Press,

Toronto.

SAAS, E., NATHAN, Y. and NISSENBAUM, A. 1965. Mineralogy of certain

pyrite concretions from Israel and their alteration products. Min.

Mag., 35, 84-87.

SAID, M. 1969. The Pleistocene geomorphology of the Burbage Basin.

Ph.D. Thesis, University of Sheffield.

 163

SAKSELA, M. 1952. Weathering of sulfide minerals. Geologi, 4, 23.

SATO, M. 1960(a). Oxidation of sulphide ore bodies. I. Geochemical

environments in terms of Eh and pH. Econ. Geol. 55, 928-961.

SCHINDLER, J.E. and HONICK, K.R. 1971. Oxidation-reduction determination

at the mud-water interface. Limnol. Oceanogr. ~, 837.

SCHRAMM, J.R. 1966. Plant colonisation studies on black wastes from

anthracite mining in Pennsylvania. Trans. Am. Phil. Soc. 56, 194 pp.

SCHULTZ, L.G. 1964. Quantitative interpretation of the mineralogical

composition from X-ray and chemical data for the Pierre Shale. U.S.

Geol. Surv. Prof. Paper, 39lC, 1-31.

SHARPE, C.F.S. 1960. Landslides and Related Phenomena. Pageant, New

York, 137 pp.

SILLEN, L.G. and MARTELL, A.E. 1964. Stability constants of metal-ion

complexes. Sec.l: Inorganic ligands. Chern. Soc. (London), Spec.

Pub. !Z_, 754 p.

SIMPSON, E. 1954. The Speedwell Cavern and some North Derbyshire drainage

problems. Cave Sci. 1, 267-273 •

.
SINGER, P.C. and STUMM, W. 1970. Acidic mine drainage: The rate determining

step. Geochim. et Cosmochim Acta, 40, 441-447.

SLACK, K.V. and FISHER, D.W. 1965. Light dependent quality changes in

stored water samples. U.S. Geol. Surv. Prof. Paper 525C. C190-C192.

SMITH, C.H. and DREVER, J.I. 1976. Control on the chemistry of springs at

Teels Marsh, Mineral County, Nevada. Geochim. et Cosmochim. Acta, 40,

1081-1093.

SMITH, K. and LAVIS, M.E. 1975. Environmental influences on the temperature

of a small uplant stream. Oikos, 26, 228-236.

SPEARS, D.A. and TAYLOR, R.K. 1972. The influence of weathering on the

composition and engineering properties of in-situ Coal Measures rocks.

Int. J. Rock Mech. Min. ScL, 2,., 729-756.

 164

SPEARS, D.A., TAYLOR, R.K. and TILL, R. 1971. A mineralogical investi-

gation of a spoil heap at Yorkshire Main Colliery. Q. J1. Eng.

Geo1.,1, 239-252.

SREBRODOL'SKIY, B.I. 1977. Mineral formation from sulfate solutions.

Acad. Sci. USSR, Dokl, Earth Sci. Sect. 226, 121-122.

STEGER, H.F. and DESJARDINS, L.E. 1978. Oxidation of sulfide minerals:

4. Pyrite, chalcopyrite and pyrrhotite. Chem. Geol. ~, 225-237.

STEINBACK, J.T. 1966. An ecological investigation of the alga genera

and other biota in waters polluted by mineral acid-drainage from coal

mines in Vintan Co., Ohio. M.Sc. Thesis, Ohio State University, U.S.A.

STEVENSON, C.M. 1968. An analysis of the chemical composition of rainwater

and air over the British Isles and Eire for the years 1959-64. Quart.

Jour. Met. Soc., ~, 56-70.

STEVENSON, I.P. 1967. The Castleton Reef Belt and adjoining areas. In:

Geological Excursions in the Sheffield Region. C. Downie and R. Neves

(Eds.) Univ. of Sheffield, 1967, 163 p.

STEVENSON, I.P. and GAUNT, G.D. 1971. Geology of the country around

Chapel-en-le-Frith. Mem. Geol. Surv. Gt. Br.

STOKKE, P.R. and CARSON, B. 1973. Variation in clay mineral X-ray

diffraction results with the quantity of sample mounted. Jour. Sed.

Pet., ~, 957-964.

STUMM, W. and MORGAN, J.J. 1970. Aquatic Chemistry - An introduction

emphasising chemical equilibria in natural waters. Wiley-Interscience,

New York, 583 p.

SYKORA, J.L., SMITH, E.H., SHAPIRO, M.A. and SYNAK, M. 1972. Chronic

effect of ferric hydroxide on certain species of aquatic animals.

Proc. 4th Symp. Coal Mine Drainage Res., Pittsburgh, Pa.

TANJI, K.K. and DONEEN, L.D. 1966. Predictions on the solubility of gypsum

in aqueous salt solutions. Water Resources Research, ~, 543-548.

 165

TAYLOR, R.K. and SPEARS, D.A. 1970. The breakdown of British Coal

Measure rocks. Int. J. Rock Mech. Min. Sci., i, 481-501.

TEMPLE, K.L. and COLMER, A.R. 1951. The autotrophic oxidation~ iron by

a new bacterium: ThiobaciUus ferroo::cidans. J. Bact., 62, 605-611.

TEMPLE, K. and DELCHAMPS, E. 1953. Autotrophic bacteria and the formation

of acid in bituminous coal mines. Appl. Microbiol. !, p.225.

TERZAGHI, K. 1950. Mechanism of Landslides. In: Application of Geology

to Engineering Practice. Berkey Volume, Geol. Soc. Am., 83-125.

TERZAGHI, K. and PECK, R.B. 1967. Soil Mechanics in Engineering Practice.

Wiley, New York, 729 p.

TREXLER, B.D., RALSTON, D.R., REECE, D.R. and WILLIAMS, R.E. 1975. Sources

and causes of acid mine drainage. Pam. Idaho Bur. Mines. Geol. 165,

129 p ,

VARNES, D.J. 1958. Landslides and Engineering Practice. In: Eckel,E.J.

(Ed.) Highway Research Board, Washington, Special Report ~, NAS-NRC

Pub!. 544, 323 p ,

VINE, J.D~ and TOURTELOT, E.B. 1970. Geochemistry of Black Shale Deposits:

A summary report. Econ. Geol. 65, 253-272.

VORA, K.H. and RAYMAHASHAY, B.C. 1978. Gypsurnrjarosite incrustation on

limestone; an unusual occurrence. Geo1. Soc. India J., ~, 228-230.

WALKER, R.G. 1966. Shale grit and Grinds10w shales: Transition from

turbidite to shallow water sediments in the upper Carboniferous of

northern England. J. Sed. Pet., 36, 90-114.

WALKER, R.G. 1967. Turbidite sedimentary structures and their relationship

to proximal and distal depositional environments. J. Sed. Pet. li,

25-43.

WARD, J.H. 1963. Hierarchical grouping to optimise an objective function.

Jour. Am. Stat. Assoc., 58, 236-244.

WARNER, R.W. 1968. Preliminary report on the biology of acid mine drainage

at Grassy Run and Roaring Creek, W. Virginia. Fed. Wat. Poll. Cont. Admn,

Cincinnati, Ohio (Nov. 1968).

 166

WENTZ, D.A. 1974. Effect of mine drainage on the quality of streams in

Colorado, 1971-1972. Colo. Water Resour. Circ. ~, 117 p.

WHALEY, P.W., CLARK, H.R. and WlRAM, V.P. 1977. Field assessment of acid

mine drainage potential of overburden materials encountered in strip

mining. Geol. Soc. Am. Abstr. Progr. ~, 665-666.

WHITTON, B.A. 1972. Environmental limits of plants in flowing waters.

Symp. Zool. Soc. Lond. ~, 3-19.

WILKINSON, P. 1950. Al10phane from Derbyshire. Clay Min. Bull. !, 122-123.

WILSON, M.J. 1975. Chemical weathering of some primary rock forming

minerals. Soil Sci. 119, 349-355.

WINMILL, T.F. 1915-16. Atmospheric oxidation of iron pyrites. Pt. IV.

Trans. Inst. Min. Engrs., ~, 500.

YAKHONTOVA, L.K., GRUDEV, A.P. and NESTEROVICH, L.G. 1979. Mekhanizm

okis1eniyapirita (The mechanism of pyrite oxidation). Geo1. Rud.

Mestorochd., ~, 63-68.

YOSHIOKA, R. 1975. Estimation of amounts of weathered products through

chemical composition of waters in the Kamenose landslide area.

Bull. Disaster Preven. Res. Inst. Kyoto Univ., 25.

ZoBELL, C.E. 1946. Studies on redox potentials of marine sediments. Bull.

Am. Assoc. Petrol. Geo1., 30, 477-513.

ZoBELL, C.E. and BROWN, B.F. 1944. Studies on the chemical preservation

of water samples. Jour. Mar. Res., ~, 178-184.

 APPENDIX A
Wet Chemistry Analytical Procedures
 167

Ai DETERMINATION OF CALCIUM

By Atomic Absorption Spectrophotometry.

Sample Treatment: 1% w/v lanthanum chloride was added to the filtered and

acidified field samples. These were then diluted 1 to 50 with distilled

water so that Ca2+ concentration fell within the range of the standard

solutions.

l
Standard Solutions: A commercial stock solution of 1000 Vg ml-
Ca was
l
diluted to give a number of standards ranging from 0 to 10 Vg ml-
l
In this analysis standards of 2.0, 6.0 and 10.0 Vg ml- Ca were used.

Operating Parameters:

Wavelength 422.7 nm
Slit Setting 0.7 nm
Light Source Hollow cathode lamp
Flame Type Air-acetylene flame, oxidising (lean, blue)

Sensitivity: For the standard conditions above, the sensitivity is about

-1
0.08 Vg ml Ca for 1% absorption.

Interferences: Silicon, aluminium, phosphate and sulphate depress the

sensitivity for calcium. This can be remedied by the addition of

lanthanum chloride or the use of a nitrous oxide-acetylene flame.

N.B. The acid concentration of the standards should be matched with that

of the samples.

Source: Perkin-Elmer, 1976.

 168

Aii DETERMINATION OF MAGNESIUM

By Atomic Absorption Spectrophotometry

Sample Treatment: As for calcium, 17.w/v lanthanum chloride was added

to the samples which were then diluted either 1 to 50 or 1 to 100 to

fall within the range of the standards.

-1
Standard Solutions: A commercial stock solution of 1000 Vg ml Mg was
-1
diluted to give standards ranging from 0 to 3 Vg m1 Mg. Standards
-1
of 0.5, 1.5 and 2.5 Vg ml Mg were used here.

Operating Parameters:

Wavelength 285.2 nm
Slit Setting 0.7 nm
Light Source Hollow cathode lamp
Flame Type Air-acetylene flame, oxidising (lean, blue)

Sensitivity: For the standard conditions above the sensitivity is about

0.007 vg ml-l Mg for 17. absorption.

Precision: About 0.57.

Interferences: Silicon and aluminium depress the absorption of magnesium

in an air-acetylene flame. Lanthanum chloride can be added or a

nitrous oxide-acetylene flame used.

Source: Perkin-Elmer, 1976.

 169

Aiii DETERMINATION OF SODIUM

By Atomic Absorption Spectrophotometry

Sample Treatment: The filtered and acidified samples were diluted 1 to 50

with distilled water.

l
Standard Solutions: A commercial stock solution of 1000 ~g ml- Na was
-1
diluted to give standards of 2.0, 6.0 and 10.0 ~g ml Na.

Operating Parameters:

Wavelength 589.2 nm
Slit Setting 0.7 nm
Ligh t Source Hollow cathode lamp
Flame Type Air-acetylene flame, oxidising (lean, blue)

Sensitivity: For the standard conditions above the sensitivity is about

0.015 ~g ml-l Na for 1% absorption.

Precision: About 2%.

Interferences: Sodium is partially ionised in an air-acetylene flame. This

may be overcome by the addition of another alkali in excess of 1000

or 2000 ~g m1-l, but this was not done in this case.

Source: Perkin-Elmer, 1976.

 170

Aiv DETERMINATION OF POTASSIUM

By Atomic Absorption Spectrophotometry

Sample Treatment: The original filtered and acidified samples were used

undiluted.

Standard Solutions: A commercial stock solution was diluted to give

-1
standards of 2.0, 6.0 and 10.0 ~g m1 K.

Operating Parameters:

Wavelength 766.5 nm
Slit Setting 2.0 nm
Light Source Hollow cathode lamp
Flame Type Air-acetylene flame, oxidising (lean, blue)

Sensitivity: For the standard conditions above the sensitivity is about

-1
0.04 ~g ml K for 1% absorption.

Precision: About 2%.

Interference: Potassium is partially ionised in an air-acetylene flame.

This may be overcome by the addition of another alkali in excess of

1000 ~g m1-l, but this was not done in this case.

Source: Perkin-Elmer, 1976.

 171

Av DETERMINATION OF TOTAL IRON

Reagents:

Colour-developing reagent: (a) Hydroxylamine hydrochloride, 25%

(b) Sodium acetate, 0.5 N
(c) Dipyridy1 solution: dissolve 0.2 g 2.2'-
dipyridyl in 100 m1 0.2 N hydrochloric acid.
Mix 2g (a), 6.8g (b), and 40 m1 (c) and
dilute to 200 m1 with distilled water. The
reagent is stable for at least one month.

Standard iron solution: Dilute connnercial 1000).lg ml-~e standard to

give a 100 ).lgml-1 standard solution.

Procedure: Into a 100 ml volumetric flask pipette 10 ml sample solution.

Add 5 m1 colour-developing reagent and make up the volume with distilled

water. For standard solution add 5 ml colour-developing reagent to 10 m1

of standard solution and dilute to 100 mI. A reagent blank should also

be made consisting of 5 ml colour-developing reagent diluted to 100 mI.

A deep red colour deve10pes if any iron is present in solution. Measure

the optical density of blank, standard and sample solutions in a 2 em

cell at 522 nm*. Determinations must be made within 24-hours of colour

development.

Calculation:
1-1 ° d 1 dO Dilution
).lgml
-1
Fe in sample
= ).lgm Fe l.n standar x samp e sp rea l.ngx factor
standard sp reading - blank sp reading
= -r1:.:0:.........:x=--:-:s:.:a:::m::li:p:..:1:.:e~s;J;p~r:..::e
x 10
(standard - blank) sp reading

sample sp. reading 100

(standard - blank) sp reading x

Precision: Approximately 1-2%.

Source: Sheffield University, Geology Department - Analytical Laboratory

Handbook.

*Using a Pye Unicam SP 500 spectrophotometer.

 172

Avi DETERMINATION OF FERROUS IRON

Reagents:

Potassium dichromate standard: Weigh 0.2728gdry potassium dichromate and

dilute to 1 litre in a volumetric flask.
2+
1 m1 of this solution = 0.31 mg Fe •

Indicator Solution: Dissolve 0.1 g sodium diphenylamine

sulphonate in 500 m1 distilled water.
Add 500 m1 orthophosphoric acid and mix
well.

Procedure: Add 5 m1 indicator to 50 m1 sample and titrate against K2Cr207

to purple end-point.

Calculation:

-1 2+
~g m1 Fe in sample = vo1umeof titre x 0.31 x dilution factor

= volume of titre x 6.10

Precision: 1-2% if dichromate indicator is carefully made up and checked

2+
against a Fe standard solution such as ammonium ferrous sulphate
-1 2+
(0.7021 g ~ 1 litre = 100 ~g m1 Fe ).

Source: Sheffield University, Geology Department - Analytical Laboratory

Handbook.
 173

Avii DETERMINATION OF ALUMINIUM

By Atomic Absorption Spectrophotometry

Sample Treatment: With the exception of sample No.7, which was diluted

I to 50, all samples were analysed undiluted.

-1
Standard Solutions: A commercial stock solution of 1000 ~g ml Al was

diluted to give standards covering the range 0 to 50 ~g ml-l AI.

Standards of 10, 20 and 50 ~g ml-l Al were used here.

Operating Parameters:

Wavelength 309.3 run

Sli t Setting 0.7 nm

Light Source Hollow cathode lamp
Flame Type Nitrous oxide-acetylene flame, reducing (rich,red)

Sensitivity: For the standard conditions above the sensitivity is about 1.0
-1
~g ml Al for 1% absorption.

Precision: 2-5%.

Interferences: Hydrochloric acid and iron greater than 0.2% decrease aluminium

sensitivity. Sulphuric acid has a similar effect. Aluminium is also

partially ionised in the flame unless total salts exceed 1000 ~g ml

-1 •

Source: Perkin-Elmer, 1976.

 174

Aviii DETERMINATION OF CHLORIDE

Reagents:

Silver nitrate, 0.1 N

Methyl orange indicator: dissolve 0.04 g methyl orange in
20 ml alcohol. Add 80 ml distilled
water and mix well.
Precipitated calcium carbonate.
Potassium chromate indicator: dissolve 5 g K2Cr04 in 100 ro1water.
Sodium carbonate 0.02 N.

Procedure: Pour 50 ml sample into a 100 m1 beaker. Add three drops methyl

orange indicator and neutralise acidity by the addition of 0.02 N

sodium carbonateCthe orange colour will disappear). Add approximately

5 ml K Cr0 indicator and titrate with 0.1 N AgN03 until yellow colour
2 4
is tinged with a reddish brown.

Calculation:

~g ml
-1
Cl 1n sample
= rol AgN03 x 3550
ml of sample

= ml AgN0 x 71
3
f 10 ~g ml -1. may 1nterfere.
I·t 1S poss1b
. 1e that Fe 3+ 1n
. excess 0
Interference:

The accuracy of samples high in ferric iron may, therefore,be con-

siderably less than stated below.

Precision: Approximately 5%.

Source: Allen, 1974.

 175

Aix DETERMINATION OF SULPHATE

Reagents:
Conditioning reagent: Dissolve 75 g NaC1 in 300 m1 distilled water.
Add 30 m1 conc. HC1, 100 m1 95% 1-propano1
(isopropyl alcohol), and 50 m1 glycerol.
Mix well.
Barium chloride: Sieved to produce crystals of uniform size
(300-500 ~).
Sulphate standard: Dissolve 0.1479 g anhydrous sodium sulphate
in water and make up to 1 litre (1 m1 = 0.1
2-
mg S04 ).

Procedure:

1. Measure maximum of 100 m1 sample containing not more than 4 mg SO~-

(40 ~g m1-1) into 250 m1 conical flask. Make to 100 m1 as necessary.

N.B. Considerable dilution was needed for many of the Mam Tor samples

(up to 100 x).

2. Add 5 m1 conditioning reagent and mix using magnetic stirrer.

3. Add 0.5 g barium chloride and stir for one minute.

4. Measure absorption on spectrophotometer at wavelength of 420 nm.

Continue to take readings at 30-second intervals for approximately 5

minutes, noting the maximum value (normally at c.2 min).

5. Subtract absorption due to colour and turbidity in original sample.

6. Prepare calibration curve using various dilutions of SO~- standard for

-1 2-
range 0.0 to 40 ~g m1 S04. Treat calibration samples exactly as

described above but use blank determination with distilled water to

set zero.

7. Read so~-concentration in sample from calibration curve and multiply

1
by dilution factor to obtain value in ~g m1- •
 176

Interferences:

(a) Silica in excess of 500 Vg m1-1

(b) Precipitation may be retarded by high concentrations

of dissolved organic matter.

(c) Extreme colour and turbidity in original sample.

Precision: 5-10%.

Source: Adapted from Department of the Environment 1972; Analysis of Raw,

Potable and Waste Waters. HMSO, London, p.78-79.

 177

Ax DETERMINATION OF TRACE METALS

By Atomic Absorption Spectrophotometry

Trace metals were determined using a high-temperature graphite

furnace in conjunction with the 'Perkin-Elmer 400' spectrophotometer.

In all cases the purge gas was argon and the gas interrupt 3 seconds

during atomisation, unless otherwise stated.

Copper (Sample dilution x 1)

Operating Parameters: Wavelength 324.7 nm

Slit Setting 0.7 nm
Drying 105°C for 50 seconds
Charring 10000C for 20 seconds
Atomising 27000C for 10 seconds
-1
Standards: 0.05 and 0.10 ~g m1 Cu.

Nickel (Sample dilution x 10)

Operating Parameters: Wavelength 232.0 nm

Slit Setting 0.2 nm
Drying 105°C for 50 seconds
Charring 10000C for 20 seconds
Atomising 27000C for 10 seconds
-1
Standards: 0.05 and 0.10 ~g m1 Ni.

Zinc (Sample dilution x 50)

Operating Parameters: Wavelength 294.0 nm

Slit Setting 0.7 nm
Drying 10SoC for 50 seconds
Charring 10000C for 20 seconds
Atomising 27000C for 10 seconds
-1
Standards: 0.5 and 1.0 ~g m1 Zn.
 178

Chromium (Sample dilution x 1)

Operating Parameters: Wavelength 357.9 nm

Slit Setting 0.7 nm
Drying 105°C for 50 seconds
Charring l3500C for 20 seconds
Atomising 27000C for 10 seconds
-1
Standards: 0.02 and 0.05 ug ml Cr.

Manganese (Sample dilution x 100)

Operating Parameters: Wavelength 279.5 nm

Slit Setting 0.2 nm
Drying 105°C for 50 seconds
Charring 1l00oC for 20 seconds
Atomising 27000C for 10 seconds
-1
Standards: 0.02 and 0.05 llgml Mn.

Lead: (Sample dilution x 10)

Operating Parameters: Wavelength 283.3 nrn

Slit Setting 0.7 nm
Drying 105°C for 50 seconds
Charring 500°C for 90 seconds
Atomising 27000C for 10 seconds
-1
Standards: 0.05 and 0.10 ug ml Pb.

N.B. Gas interrupt 7 seconds during atomisation.

Cadmium: (Sample dilution x 10)

Operating Parameters: Wavelength 228.8 nm

Sli t Setting 0.7 nm
Drying 105°C for 50 seconds
Charring 300°C for 90 seconds
Atomising 20000C for 10 seconds

Standards: 0.02 and 0.05 llgrnl-l Cd.

 179

Vanadium: (Sample dilution x 1)

Operating Parameters: Wavelength 318.4 nm

Slit Setting 0.2 nm
Drying 105°C for 50 seconds
Charring 13000C for 20 seconds
Atomising 27000C for 10 seconds
-1
Standards: 0.50 and 1.00 ug ml v.
 APPENDIX B
Results of the Water Analyses
 180

·. . . .. .
NNooC7'l~oo~ .....

.....
o
·... .... ..
~C7'IC7'IooOoo~C7'IN~ ooN
·......... .....
OO~~OO .....

.....

o
o
.....
~
·...... ~. · . . . . .
O~~OOOO N NO~
·.... . . ... .....
~OOO~OO

00
.....
M
·.............
.....
MooOC7'l~MOO~~~N~
OOC7'looC7'lMN~NN .....
N~NI~IOONM
o
·.........
~OM~O~MOOO~
NM Noo
.......... .....

· . . . . . . . . . . . ..
OOOON~~N
C7'1C7'IC7'IooooN.....
~O
.....
O~OO~~O
.....
OOMNI~IO~~NM~NN
0
. O~ O~ooNooO
.
C7'I
.............................................

·.. .. .. . . . . . . .
~OOOC7'l
.....
.....
NM~~ooM
OOOooC7'lM~~~M~~~~I~ION~O
.....
~OO .....
·.... .....
C7'1OOOoo~C7'IOO~~M
MO
.......... ..... ~

M
.....
~O~
· ..
OOOOloo
ooOO~ooOO~M
~
OOO
~ I~IO~~
.
..... OONM~NOO~C7'I
M N
.
.......... ..... oo

·... . .. ... ·.... . ....

-
~OOO ooO~~MMC7'IOO~C7'I ~~ooooC7'lNO .....
NOO
.....
..... OOOOloo~~~~MMN~ .....
I~IO~~ M N
u
o
.......... ..... oo
'-"

00
·..
..... 0\

OO\OOloo
~

..........
~ .....
O~~~ .....
o
I~IN~NMN~N~
·.........
~OO~ONOOO~~OO
C7'I
..... N .....

·.. . .. . . . .
~NC7'IN ..........
~~NO o ·... . .....
~OOOC7'lONONOC7'l
C7'1C7'IC7'IIC7'IC7'IOOC7'lC7'l~OOOOC7'lO\IC7'IIC7'IC7'IOC7'lOOOOOC7'l
..... ..... ....................

.....
~ · ..
~C7'I"'" MOOOOOC7'l~~M~oo
OC7'lC7'lIC7'l~M~~~NM~MI~IONNOO
~
. OC7'lC7'lOooC7'l ~NO
.
..... ..... ~
~
· . . . . . . . . . . . .. . ....
OOOOOOOO
.....
~~OO~ .....
C7'IC7'I
..........
~O
..........
.....
OO~N
ONMNI~IO~~N~~~~~~
C7'I OO~~O

.......... ..... ...............

M
·.........
OC7'lC7'lIC7'l~~~~~~~~~I~ION~OO
·.........
OOooO .....
~~M~~~
O OOO
..... .....

N OOC7'llooN
·.........
~00~0\~00~00

.....
..... ~

~ ....................
~NI~IN~~N~~NN O
.......... .................... ..... .....

..... ·. .. . ..... . ...

~oo~M""'~ .....
CX)C7'IC7'IC7'I~~O .....
·. .. . ... . .
NCX)C7'IONCX)~O~O

Cl)
"0
o
U

·.. . . . . . . . . . . . . . . . . . . . . . . . .
oooooooooooooooooo~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~

·. . . . . . . . . . . . . . . . . . . . . . . . . .
OO ....................
NNN .....
NNMM~~~~~~~~OOOO~~O
..........................................
00000000000000000 .....
~ .....
~~ .....
OO~N~~~~M~ ~OON~~M~~ .....
~OON
NMO .....
NNO ..........
OON .....
N NONO .....
ONONO .....
O
 181

....... ~,......._,...-.-...,-.....~~.,.......,_,,......,_,~,........_,...-..~~...-.....-.....-..
...-.....-.....-.....-.....-.....-.....-.....-.....-..cn ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
~~~~~~~~~~ ~~~~~~~~~~~~~~~~
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
...-l...-l...-l...-lMMM...-l...-lM f"'-NNNNNNNNNC"'INNNNN
'-''''''_'''-'''-'''-''-''-''-''-''-' ....._".......,"-"'-'''-''.......,..._,,'-''-''-''-''-''-''''''-''-''-'

·. . . .... . . ·. ... ... . . . . . .. .

OOI.OLI"I...-lO\MO...-l-::t OLl"lOI.Of"'-OO-:tf"'-C"'IOI.OI.O-::tf"'-I.OO'I
OOOf"'-OOO\C"'I-:t-:tLl"lI-::tOLl"lf"'-I.Of"'-...-lOOLl"lOI.OON-:tOO-::t
LI"I...-lM-:t-:tM OO\LI"IOOM-:tLl"lLl"l-::tC"'l...-lMO-:t N
...-l M ...-l

OO'lNN-::tNLI"I -::t ,....f"'- 1.0 ~O LI"I I.O-::t

I • • • • • • • I • I I • • • • • I • • I

O'IO'II.OOOI.ONNN 00 -::tLl"lI.Of"'-I.ON I.Of"'-OOO f"'-O'IO'I

• • • • • • • • I • • • • • • I • • • • • I • • •
1.01.01.01.01.01.01.01.0 I.OLI"ILI"ILI"ILI"II.O 1.01.Of"'-f"'-f"'- 1.01.01.0

M
o
o
OLl"l-::tI.OC"'IN-::tLl"l
• • • • • • • • I
Ll"lLI"ILI"ILI"ILI"ILI"ILI"ILI"I
I
O\OOf"'-Of"'-
• • • • • I • • • • • I
LI"I-:t-:tLl"lLl"l 1.0 1.0 1.0 1.0 1.0 I.OLI"ILI"I
...
00
...-l
C"'IO'INLI"II.ONO...-l
• • • • • • • • I
Ll"lLI"ILI"ILI"ILI"ILI"ILI"ILI"I
o
• I
N...-l-:tO-::tN
• • • • • • I
Ll"lLI"ILI"ILI"ILI"ILI"I
Ll"lOO...-lf"'-OOOOI.ONO'I
·........
NONNNLI"ION
• • • • • •
C"'IC"'IC"'IC"'IMC"'IMC"'I
• • I
C"'I
• I
I.O-::tLl"lC"'lOC"'l
• • • • •
C"'IMC"'IC"'IC"'IC"'I
• I ·........
NON...-lNONC"'ILI"I
C"'IC"'IC"'IC"'IMC"'IC"'IC"'IM

OOOONON...-lC"'l
• • • • •
NC"'IC"'IC"'IMMC"'IM
• • • I
N

M
-::t -::t N N M C"'I
• • • • • • I
• I MMC"'IMC"'IC"'I
-:tOC"'lC"'lNNN-:tC"'l
C"'IMC"'IC"'IC"'IMC"'IC"'IM
·. . . .. ...
OO'lOOOOI.OI.OI.O...-l
• • • • • • •
I.OLI"I-:t-:tLl"l-::tLl"lI.O
• I
1.0
M
• I
MO\...-lM-::t-::t
• • • • • • I
LI"I -::t LI"I LI"I -::t LI"I
...-l1.Of"'-OOO'lO'lOLl"lLl"l
Ll"lLI"I-:t-::t-::tLl"lI.OLI"ILI"I
·. . . .. .. .
.......
,....,
+
::t::
LJ
OC"'l...-lN-::tMMN
• • • • • •
C"'IC"'IC"'IC"'IMMMC"'I
• • I
M
M
• I
1.OC"'If"'--::tN-::t
• • • • •
MMMMMM
• I ·. . . .....
-:tNNC"'INC"'IC"'INM
MMC"'IC"'IC"'IMC"'IC"'IC"'I
o
eo
o
M
00
I.ONN-:tMf"'--:t-:t
• • • • • • •
L/)LI"ILI"ILI"ILI"I-::t-::t-:t
• I
00
.,
-::t
I
OOI.Of"'-NI.O
• • • • • I
LI"I-:t-::tLl"lLl"l
NN...-l...-lN-:tO'l...-lO\
·. .. .....
...-l
I
.._.,
::t::
c,
-:tLl"lC"'lLl"lMMNOO
• • • • • •
C"'IC"'IC"'IC"'IMMMC"'I
• • I
M
M
• I
1.0 LI"I -:t N -::t -::t
• • • • • • I
MC"'IC"'IC"'IC"'IM
f"'-I.OI.OOOOONf"'-1.01.O
C"'IC"'IC"'IMC"'IMC"'IC"'IC"'I
·. .. .... .
.......
Nf"'-LI"IO'I-::tI.O-::t-:t
LI"I-::tC"'lC"'lLl"l-::tC"'lLl"l
-,., -::t
-::t
LI"I...-lNNNI.O
Ll"lLI"ILI"ILI"ILI"IC"'I
, ·. . . . . .. .
OOOOI.OM...-lOOOOLl"lM
Ll"lLI"ILI"I-:t-::tLl"lLl"lLl"lLl"l

........
MI.O-:tI.OI.OLI"I-::t1.O
C"'IMMMMMMC"'I
'.,NM OOOOOLl"lI.Of"'-
-::tMMMMM
, • • • I I I I I I

O'I...-lNOOOMO-:t 00 LI"I-::t-:tN-::tO -:tN-:tM-:tN-:t-:tLl"l

("I"') ••••••• -, -, ••••• -, •••••••••
NC"'IMMNMMM N MMMMMM MMMMMMMC"'IM

N
NOC"'l-:tOI.OO'IC"'I
• • • • • • •
C"'IC"'IC"'IC"'IC"'IC"'INC"'I
• I
M
M
• I
""'LI"ILI"IN-::tO
• • • •
MMC"'IMMM
• • I ....
...-lLl"lC"'l-::t
MC"'IMM

...-lNN-:tLl"lNO'IN 0\ MNOO\NI.O O'II.ON-::tf"'-I.Of"'-OOOO

t""""I ···.····1.'···.··,·
Ll"lLI"ILI"ILI"ILI"ILI"I-:tLl"l -::t Ll"lLI"ILI"I-:tLl"lLl"l
.. ····
Ll"lLI"ILI"ILI"ILI"ILI"ILI"ILI"ILI"I
..
~
'"Cl
o
CJ

·... .... . .... . . . .... . . .. . . .

0000000000000000000\0\0\0\0'I0\0\0\0'I0'I0'I0'If"'-0\0'I0'I0'I0\
f"'-f"'-f"'-f"'-f"'-f"'-f"'-f"'-f"'-f"'-f"'-""'f"'-f"'-f"'-""'f"'-f"'-f"'-f"'-f"'-0'If"'-f"'-f"'-f"'-"'"

·.. . . ... . .... . ..... . . ... . . .

OO...-l...-l...-l...-lNNNMNNMM-::t-::tLl"lLl"lI.OI.Of"'-f"'-OOOOO'lO'lO
...-l...-l...-lM...-lMMM...-lOOOOOOOOOOOOOOOOOM
-:t...-lf"'--::t...-lOOLl"lNO'IO\I.O""'Mf"'-M-::tOONLI"IO'IM-::t,.......-l-:tOON
NMOMNNOMMOONMNMNONOMONONOMO
 182

o
101
~
..-l

o 0000 OOOOMOOLl"lLl"lLl"lO
o
..-l 1000000001000\ ......
0\000 ......
0000M
..-l ..-l

00

III
00
.~
~
'"(j
til
,_.
Q)

.u
o
Q)
p...
III
;::)
III

-e
..
Q)
N 00

0 ~
Z '"(j
,_. ,_.
Q) Q)
.u .u
Q) Q)
~ ~
iC +
iC +

Q) iC iC +iC iC iC iC iC iC iC
'"(j ole iC +oIe ole ole ole ole ole ole
o ~~~0~H~~H~ZO~O~OOH~>~~~N~~8§
u

...........................
OOOOOOOOOOOOOOOOOOO\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\O\~
..................................................................................................................................................................

OO..-l~~..-lNNN~NNMM~~LI"ILI"I~~ ............
OOOOO\O\O
~~~~..-l~..-l~~00000000000000000~
~~~~~~~~~~~~M~~~~~~~M~~~~~~
NMO~NNO~~OON~N~NONO~ONONO~O
 183

~~NN~OO~OOO~I~~~NMNMOOI I ~
IN~MM~~~NO~ O~~OO~N~ oICoIC0 I
NNN~~~ ~~~ ~ N~~~~ N

~ OO~~~~~OO~MM~ON~OOOO~O~~~~ O~~
o ~~OO~~~~~~OO~~OO~OO~~~~~OOolCoICOO~~
~ ~~

o ~~~~~~~~~OO I I OONN~~~~~OO ~~OO

o ~~ oICoIC
~

+ MOM~~~MOOOO~ I I ..:tNNO~OOOO..:t..:t~ ~O~

N ~~N~~~M~~ ~~ ~ ~~~oICoIC~N~
cO
U

§
H
U

~
U

OO~~~~~~~~..:t~~~~MOOM~~MM
~~~OO~OO~~~~OOO~OOOMOOO~~~~oICoIC I I I
~~~~~~~~~~~N~N~MN~~NN

Q)
"0
o
U

................... ........
OOOOOOOOOOOOOOOOOO~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~
OO~~~~NNN~NNMM..:t..:t~~~~~~OOOO~~O
~~~~~~~~~OOOOOOOOOOOOOOOOO~
~~~~~OO~~~~~~M~~~OON~~M~~~~OO~
NMO~NNO~~OON~N~NONO~ONONO~O
 184

m oo~ON~oo~N~~ ~ON~~~~~ ~
M I ~~\ON~~~MNM I NNN~N~MM I I I I ~ I I

00 MrlMM\OooO~oo~OO~~~~N\O~ooON~~~O~~
~ ~~ ~ ~ ~ ~~~~ ~~~

~ ~MM~oo~NN~~~NMO~~~~ooM~~~~Noo~
rl MMrlMNMMNNNNMNNNNNNNNMMMNM~~

M ~ooO\O\OM~MooNoo~~~~rl~NrlNM~N~oo~\O
rl MN~M N~~NNN N ~ NMM~~~N~M~

~~ooooO~NNNM~~oo~~oo~~~ooOoo~~oo~rl
~OM~Moo~~Ooo~~ooMM~~~O~NMMoo~MN
rl~ ~ rl ~ rlrlrlrl~ rlrl

~~O~~O~~~oo ~~MOO~M~oo~O~~~~~
~ N~ ~rl ~ I I ~~ rlrl
~

~~N~~~~N~MNM~~~MO~ooO~~~oo~~oo
,,0 NNooN NNNNMNNN~~NNNNMNNNNNMN
~

~~NNOO~~O~~~oo~OOooM~~O~O~~M~
~~~~~~~~~oo~~~~~~~~~~~~\O~~~~

Cl)
"'tj
o
u

...........................
...........................
OO~~~~NNNONNMM~~~~~~~~oooo~~O
rlrl~rlrlrl~rl~~OOOOOOOOOOOOOOOO~
~~~~~oo~N~~~~M~rl~ooN~~M~~~~ooN
NMO~NNO~~OON~N~NONO~ONONO~O
 185

~ IOONMO~~N~O ~NM~OONM~ N
M Ll)LI)~M~~~MNC") I NNN~N~MM I I I I ~ I I

~ ~OONM~~~~LI)MOOOLl)~~~~LI)LI)NLI)C")~OO~~~
o ~~~~~~~NNN~~~~~~~~~~~~~~~
~

o ~MOO~OM~OO~O O~~O~ooOOOO~O~N~O
o ~~~ ~ I I ~~ ~~ ~ ~~~~~~~
~

co OMNN~OO~~~~OOO~~~~COOOOC")C")M~~CO~
~ N~NM~~N~~~~N~~~N~NNNNNMNNNN

M M~M~~O~~~OM~~~~M~O~~N~~~CO~~
~ N~NNNM~MN~~~NMNMMC")NNNNMC")NNN

~~~MCONN~~NNNN~~~N~MNN~~~N~M
~~ ~~~~~~~~~ ~~~~~~~~~~~~

~NOO~NMO~O OO~OO~~~~CON~NNO
co ~~~~~~~~ I I ~~ ~~ ~ ~ ~~~~~

+~
z ~COMM~CO~~~~~CO~~~~~~~~~~~~~N~
~ ~~~~~~~~~~~~~~~~~~~~~~~N~
~
H
Cl
o
tI)

MOMN~M~N~OO~COM~~~N~~~~~~~LI)~M
M ~~LI)~~~LI)~~M~~~~~~~~LI)~~~~~LI)~~

<l)
'"Cl
o
Co)

...........................
...........................
OO~~~~NNN~NNMM~~~~~~~~COOO~~O
~~~~~~~~~OOOOOOOOOOOOOOOOO~
~~~~~OO~N~~~~M~~~CON~~M~~~~CON
NMO~NNO~~OON~N~NONO~ONONO~O
 186

· · · · · · · · · · · · · · · ·~·~· ·
CIj O\\olf"IM MN~ O\O\LI"l NLI"l -::t 0"1 MN -::t 0\ \0
M I I I I I I I I
If''Ilf''I\O-::t LI"l\0 -::t MNM MN N -::tNM ~

~ ~ -::t ~ M-::tMLI"l-::t ~ ~ NLI"lM ~ N ~ 0 N M

· · · · · ·"~·~·~·~·~~· ·.....·~·~·~· ~·~· ~·~·~· ·
-::t N 0 M 00 0\

~·~·~· ~· ~· ~~~~~~
0
~ 0

\0 00 \0
·"· · ·"·"·"· · ·"·"·"· · · ·
\0 \0 \0 \0 \0 \0
· ·"· ·"·"·"·"· ·"·
0 CO 0\ 00
0 I I
~ 0000000000 000000000000000

~ 00 0 \0 LI"l 0 0\
·~· ·"·~· · · ·"~·~~·"·~~· ·~·"~· ·~·~·~·~·~·~· · ~~· · ·~·
00 N 0\ 0 \0 \0 \0 \0 LI"l M N tI'I M 0 M
~ " M
M
N NN ~N~ N N

~ 0 N 0 ~ ~
"·~·N-::to\ ~
· · · ·"· · · ·
M N LI"l\0 0\
· ·-::t·MN·
-::tOO
· · · · · ·LI"l-::t
Lf'\oo LI"lMN\O
"
~
· · · ·~0·00\0· ·LI"lLf'\LI"l-::t-::t-::t
"" " 0\ "
r-,
~
NM M LI"lLI"l LI"lLI"l

~ -::t ~ 0\ 00 ~
···························
\0 0 0\ 00 0\ 0 M 0"1 -.:t N N M -::t -::t 0\ 0 M 0 \0 -::t-::t
~
\OLI"llf"l\oLI"l\O\o\o\oLI"l\O\oLI"lLl"llf"l\oLI"lLl"lLl"llf"l\o\o\o\oLI"l LI"lLI"l

-e LI"l\0 r-, \0 M
· · · ·~·"~· ·~·~· · · · · · · · · ·"·"· · · · · · · ·
M NO\M\O 0 \0 00 0\ 0 LI"lCOMMN MM 00
~ NN N NN
N~NN N r-I N r-I r-I r-I ~ N r-I r-I r-I r-I r-I r-I
.......
r-I
I

· · · · ·"· · ·"· · . . · · · ·
~ ~ ~ 00 N \0 0
·· ·
0 -::t LI"l~ NLI"l
· ·
0\ ONooMCOOLl"l\OMO NO\"
S ~
Q)
~
· · · · · ·
LI"l -::t-::t-::tMM-::tMMM-::tMMNN-::tNMMM-::t-::t-::tMM-::t-::t
~;j
~ ~
· · ~·~·"~·"~·~·"~· · · · ·"·
00\0 LI"l-::t -::t ooO"looN
· · · · · · · · ·~· ~·"·~~
Z N N\oM \OM r-I 00 0"1
+ 00 I I
N
:::.:: Q) N NNNN NNN N r-I ~ r-I r-I r-I
~
S .,-l
Cl)
~
· · · · · · ·"· · · · · · · · · ·"·"· ·LI"l·"Lf'\·LI"l·If''I·-::t·
Lf'\-::tN \0 M M N Lf'\ -::t
LI"l-::tLf'\Lf'\ If''I\0 00 00 LI"l N 0\

" Lf'\·If''I
·LI"llf"lLl"lLl"lLl"lLf'\LI"lLl"lLl"lLl"lLl"lLl"lLl"lLl"l-::tLl"lLl"lLl"lLl"lLl"l
H 0\
Cl)
Cl)
-<
~
o ~ ~ ~
~~ ~ -::t ~ M
· · · · · · · · · · · · · · · · · · · · · · · · ·"· ·
p.., 0 00 00 N 0\ 0 ooO\LI"l-::t-::tM N 0 00 -::tM 0\
\0
-.:t MM -::tM -.:t-.:t-.:t MMMMMM M M MM M M MM -::t-.:t-::t M

~ -::t ~ LI"l
· · · · · · · · · · · ·"·M · I
0 \0 0 \OLI"lLl"lLl"llf"lM
-::t "·-::t·0·00·-::t·"·N ·LI"l-::tLl"l-::t-::t-::t-::t-::tLl"l-::t-::tMM I I I I I
-::t-::t-::t-::tLl"l-::t
"
M -.:t M ooLl"l LI"l ~ N
· · · · · · · · · · · · · · · · ·"· · · · · · · · · ·
0 \0 0\ M
-::t LI"lLI"l 0\ \0 Lf'\Lf'\00 If''I0 000"l-.:t 0\
M
\0 \0 \0 \0 \0 \0 \0 \0 \0 \0 LI"lLI"l\0 \0 If''I\0 LI"lLI"lLI"l\0 \0\0\0\0\0\0 LI"l

-::t ~
"·~·\0\00000 ····
O"INlf"Ioo
· ·
-::t-::t\OooLl"looN
N
""
· · · "
"·0
~
· ·" ·
0 N If''I00 NLI"l
· · · · "· · · · · ·
\O\OLI"lLl"l-::t-::tLl"l-::t-.:t-::t-::tLl"l
0"1
I I
\0"\0\0

00 0 N M ~ N 0 ~~
· · · ·NN· ·~N· ·~·~N· ·~·NNN · · · · · ·"· · · ·
\000 OMLI"lOO-::t
~
NNN· · ·N ·N NNNN
0\
" 0\
· " " LI"l
" " 0"1
MMMNNNN

Q)
· !Xl U t=l
'"
0
U
Or>:l~c'!)::CH~:::'::H~ZOp..,O'~CI) ~~>~:X:~N~ !Xl U t=l

00 ooooCOooooooooooooO\O\O"IO"IO\O\O\O\ 0"1 0"1 0\ 0\0\0\0\0"1 0\ 0\

s::
.,-l
~ ~
Q) """""~"""""""""""
·~~ ~ ~ ~ · · · · · · · ·"·"·"·"·"·"·"·"·"·"
· · · · ·NNN· · ·~NNMM-::t-::tLl"lLl"l\O\O""ooCOO\O\
00 0·

· ·"· ·~· · · · · · · · · · · · · · · · · · · · · · ·
r-I r-I r-I r-I r-I ~ ~ r-I 00000000000000000 r-I
~~ -::t ooLl"lNO\O\\O"M"~-::tooNLf'\O\M-::t"~-::tooN
CIj -::t ~
Cl) NM 0 r-I N NO~~OON~N~NONO~ONONO~O
 187

·· ··. . .. · ·· ·····
~ o.nN-::t~o.n -::tr-4 I' N -::tr-4 0 M
M
til
MOO~
I ~r-40-::t~1'000o.nO
0 0
I
N N 0 M\ON I' 0
r-4
I I I I N · I I
NMM

~
·············· ·············
r-4 MN 00 N o.n-.;t-::tOO~\O 0 o.n M r-4 I' 00 o.n I' MNOOM r-4 -.;to.n
0 r-4 r-4 N r-4 r-4 o.n r-4 r-4 r-4 MOOr-4N r-4 r-4
r-4 N r-4 00 r-4o.n-.;to.nNN 0

~ \O-.;t o.n
· ··········· ···
OO-.;t I' 0 M M I' O..;to N
··········
0 I' M\O r-4 \0 r-4 0 M~ 0 0
0 I I
r-4 0 Mr-4 r-40-::t r-4 N r-4 N NM r-4 0 o.n r-4 OONOOO r-4 r-4 r-4

~ ~
·· · · · ··· ·· · ·· · ··· ·· ··· · · · ··
00 0 r-4 0 o.n o.nN N N o.nN OO~ I' I' o.n \0 o.n M N r-4 0 0 I' -::tr-4
r-4
r-4 N M N N o.n o.n \0 -::t-::tON r-4 r-4 o.n r-4 o.n 0 -::tr-4 -::tM M N-::t \0 \0

~ ~ \0 000000000000
· · · · · · · · · · · ·~· · · · · · · · · · · · · · ·
00 r-4 ~N r-4 I' 0 N I' NOO ~
I'
r-4 N N o.n\OOOM N -.;t~M 0 M OOOOo.n\OOOO-::to.n\O\O-::tM\OOO
NN NNr-4N N r-4 r-4 r-4 r-4 r-4r-4 r-4N Mo.n\O\O-::to.no.no.n

··················· ······. ·
M -.;tN N o.n o.n o.n o.n ~ 00 N r-4 000000000000000
\0 I' \O\OM o.nr-4o.n~ ~M o.n \0 I'l'o.n 00\0 Mo.nNM-::tOO-.;tOMN
r-4 I' 000000 1'0000\0 o.n \0 I' \0 ~I'\O 000 OO~Oo.nOOMo.n~l'o.n
r-4 r-4

~ ~
··········· ········ ········
N MNNo.n~\O r-4 o.n N M 0 NO I' 0 r-4 I' 0 ~M M ~OO\O
M
r-4 N No.n-::tO~M OO~N o.n r-4 o.n r-4 M o.n -.;to.n o.n N o.n I' I' NNOO
..;to
r-4 r-4 r-4

~ ~
· · · · · · · ·~· · · · · · · · · · · · · · · · · · ·
I-< 00 00 \0 -::t00 I' \0 I' 00 000000000000000
Q)
r-4 I'-::tM-::t-::t ~N M 00 r-4 o.n \0 r-4 NOO~~~MI'-::t-::t r-4 \Oo.n-.;tN
il::l r-4 OOr-4r-4o.n
r-4 r-4 r-4 r-4
00
r-4 r-4
I' r-4 Oo.n N -.;tM~N-::t1'1'
r-4 r-4 r-4
r-4 r-4 0 1'1'0r-4
r-4 r-4 r-4 r-4 r-4
Z
+
M Q)
. ,

~ r-4 ~ 0
Q)
r:<.< .~
+J
tI)
00
M
··········
r-4 r-4 N
r-4 N I' 0
r-4 N
o.noo M
00 -.;t\0 N
M
M
I I
0
···············
r-4 N M
M
r-4 \0 N
o.n 00 M
N
\O\Oo.nOO
NNOMr-4
M
M
I' 0
M r-4
+
+ ~ 00 \Oo.n I' \ON~ 000000000000000
· · · · · · · · · · · · · · · · · · · · ·~~· · · · ·
N r-4 00 N
Q)
r:<.< ON 0 M~ 0 r-4 l'o.nl'~r-4-.;t 0 -.;t-::t\O\O~N~ M I' o.n o.n\O 0
I' r-4 r-4 r-4 r-40 r-4 r-4 0000r-4r-4 r-4 r-40r-4000r-4 r-4 00 r-4
r-4 r-4 r-4 r-4 r-4 r-4 r-4 r-4 r-4 r-4 r-4 ~ r-4 r-4 r-4 ~ ~ r-4 r-4 ~ r-4 ~ r-4 r-4 ~
Z
g
H

· · · · · · · · ·~~· · · · · · · · · · · · · · · . · ·
OO-.;tO-.;t\O N 00 r-4 r-4 N o.n o.n N r-4 r-4 M I' -::tM M r-4 00 ~ \0 00
\0 ..;tor-4-.;tMM~-::tOO\O \0 0 \0 o.n I' N\ONM-::tOO M I' OO-.;tC'-lo.n
r-4 r-4 r-4NM

~ I' r-4
·· ·· ······ ·· ·· · ····· ·
00 \O~ 0\0 NNN CON 0 M 0 o.no.nMM
-.;t I I I I I I
~N Nr-4 N-.;t-.;to.nM M ON r-4 r-4 M r-4-::tr-4o.no.n-.;t

· · · · · · · · · · · · · · · · · · · · · · · · · · ~.
\0 00 -.;tr-4\O~OO 0 \000\0 000000000000000
N 0r-4 r-4 o.n r-4 \000 I' ~~OOOOONo.nOO\Oo.n M o.nM\O~\O 0
M M 0000 ~ \000 00\0 I' \OI'OOOOI'\O~\O\OI'OO r-4 ~O~OO
r-4 r-4 ~ r-4 r-4
r-4

~
··············· ······ ···
I' 000 M-::t o.nN M o.n r-4 000000000 0000
I' MI'OOM~l'o.nOOM
•
I' ~ o.n r-4
M r-4 -.;tr-4 M I' o.n-.;t\O~o.n
N ~~ 00 o.n~ 1'-.;t-::to.n\O Mr-4r-4o.n~MI'~N
r-4
I I 00 r-4 r-4 N
r-4 r-4 r-4
r-4 r-4

···· ··.····················
I' ~OO-.;t-.;t M 00 N 00 o.nN 0 o.nM\O I' 0 r-4 00000 -.;to.n~ I'
r-4 \000 r-4 N r-4 CO-::t o.n o.n o.n-.;t 0 OONN I' N -::t0 I' OONo.nN o.n 0
r-4 r-4 r-4 r-4 r-4 r-4 r-4

Q)
'"Cl ::3~UO
0 O~~~=H~~~EZO~a~tI)H=>~~~N ~UO
u

00 00 00 00 0000000000 ~ ~~~ ~~~ ~~~~~~~~~~ ~

··············· ·······.. . . ·
00
~
.~ Q)
1'1'1' I' 1'1'1'1'1' I' 1'1'1' 1'1'1' 1'1'1'1'1'1'1'1'1'1' I'
r-4~C'-IC'-IN ~C'-IC'-IMM-.;t-.;to.no.n\O\OI'I'COOO~~O

··········. . . . . . ···········
~+J 00r-4r-4
~~ ..... ..... ..... ~ ..... ~ ..... 00000000000000000r-4
~~ -.;t.....
I'-.;t .....
COo.nC'-l~~\OI'MI' .....
-.;tOOC'-lo.n~M-.;tI' .....
..;toOON
t1l
tI) NMO .....
NNO~r-400Nr-4Nr-4NONOr-40NONOr-40
 188

..-I 00000
o .... .
..-I 00000

o
o 00000
.... .
.-I 00000

00
~ .....
00000
00000

......MO\O'\O'"
r-I •••••
N.-IONN

1.1')0'\0000
\0 •••• •
.-I N.-I .-I1.1')
1.1')
.-I.-I.-I.-I.-I

·....
-:tM ......
O'\I.I')
.-I

00 ·. . ..
00000
00000

·.. . .
1.1').-1.-100
OO",C""I.-IOO
'" ............
00 00

·... .
OM-:tOM
OO.-lMO

..j" 0011 I
..-I
.-I

000..j"000

N •• I I •

r""'4 •••••

Q)
"C
o
u

·....
·. .. .
"' ............
00000
0000

0\C""I..j"
......
..-I
..-IONON
 189

.....NOO .....
OOooNNOOooMOOO~~~~~N~NNoo .....
~O~~
o ~NNNNM~NN~~~~~~~~~~~~NM~~NN
.....
o ..........
OO~~oo~OOoooo ~ ...............
OO~~OO~N~~NNN
o NN I I NNNN NNNNNN
.....
00 ~~~ .....
~~ON~~~OO~~~ .....
~~~~~O~~~~~
.....~~~~M~~~~~~~~~~~~~~~~~~~~~~

~ ooOON~ ..........
OO~~OO~~~~O ..........
~~O~~OO~ ~
.....~~OOOO~~~~~~~~~~~~~~~~oooo~oooo ..........O

~ ~~~OO ..........
~~ .....
~N~ .....
~NN~~~~oo~O .....
~~N
.....M~~NNN~ .....
NO .....
OO~~ OO ~ OO~ .....
................................................................. .....

~ ~~~O~ON~N~OOO~ .....
OO~~~ .....
~~~~~~O
.....~~~~~~~~~~O~~~~~oo~~~~~~~~~~
...............

~ ...............
~ ...............
OO~ ..........
~M .....
~OO ..........
~~~~~N
I 00 NNNN .....
NNN ..........
I I NN .....
NN ..........
NN .....
NNNNN
.....
u
w ..........
~~~~ .....
oo~ .....
O~~~ .....
~OO~ .....
Q~OOO~~ .....
~
o ~~~OON .....
NN~~~M~NN~NN~.·,~N~~~~~
H
p::; .....
o
Ea oo .....
ooOO~ ..........
~O~oo~~~~~~~~~~ .....
N~~ooOO
u NNNNONN ..........
N~N .....
~~OO NNN~~NN~
................................................................. .....

~~~~~~~~ .....
~ ~~~~~ ..........
OOO
~ ~~~~~M~~NNN OOO ..........
NN~NI I I I I I
.........................................................................................................

~~ .....
~~~~~~ ~~~~~~ooNOOO~NOooOO~
~ ONN ....................
O N ..........
OO .....
O ....................
O ..........
OOO
.......................................................................................................................................

~NN~N~~~N~~~~~~~~~~~O .....
MNO
N ~~~~OOOOooOOOO~~~~~~~~~~~OO I I ~oo~ .....
.....

~~NN~OO~~ .....
O .....
~~~~N~ ..........
~~ .....
~~~OO
oooo~~~~OOoo~~~~~~~ .....
~~~~~OOOO~~~oo
............... .....

Q)
'0
o
U

...........................
ooOOooOOOOOOooOOoo~~~~~~~~~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~~~
OO
...........................
....................
NNN .....
NN~M~~~~~~~~OOoo~~O
00000000000000000
.............................................
~""'~~""'OO~N~~~~~~""'~OON~~~~~""'~OON
.....

NMO .....
NNO ..........
OON .....
N .....
NONO .....
ONONO .....
O
 190

OOO~O~ONO~~~OOOO~~NN~~~~~~~OOO
~ ~NOOOO~~~oo~oo~~~~~N~~~~~~~~~~~
o ~ ~
~

~O~~OOOO~O ~OOOOO~~~~~N~~OO~O
o ~Ooo~~~~~~~ I I ~~~~NNNNN~NNN~~
o ~
~

~~~~~~OON~~~N~OO~~OO~~OO~~~
~~0~~0~~~~~~~~~~00~~~~0~~~~00
~~~~~NN~N~~~N~~~~N~~~~~~~~~

~
I
- ~
S
~
~
000000
OON~~~
~~~N~OO
~ ~ ~ ~
~ OOOON
N ~~N~~
~ OOO~~~
~~~
~ OO~~~~
~ O~~OONN
~ ~~~~~~
~ ~ ~~
~~
~~
O~
~~
00
~~
~~
~~
~ 000
~ ~OO~
~ ~~~
~~
;::l.
'-' 0 00 OOOONOOO
00 ~ ~~~O~~O
N ~~~00~00~~ OOOO~ ~~~
~ I I ~~~~~~~~ ~~~ ~~~~
I ~ 00 ~ ~~~~~~~
N~ Q)

0
CIl
~
~ Z
~ Q)

~
p.., .~
~
CIl

CIl
S
.~
>:
!Xl

Q)
-e
o
u

...........................
OO~~~~NNN~NN~~~~~~~~~~OOOO~~O
...........................
~~~~~~~~~OOOOOOOOOOOOOOOOO~
~~~~~OO~N~~~~~~~~OON~~~~~~~OON
N~O~NNO~~OON~N~NONO~ONONO~O
 191

0 0 0 0 0 0 0 0
0
...-I
'"
.j"
0
0
0
0
...-I
0
0
0
U"\
00
0'1
0
0
0
0
0 · · · · · · · ·
0 0 0 ...-I 0 0 0

U"\ 0 0 ...-I 0 0 0 ...-I

...-I \.0 .j" 0 0 ...-I \.0
'"
-:t
· · · · · · · ·
\.0 0 C'I"l 0 0 0 0 0
...-I ...-I
0 0 0 0 U"\ 0 0 0 I
oj( ...-I
S
00
;:L

00 0 0 C'I"l 0 0 0 0'1 o
...-I
0'1
0
0'1
C'I"l
C'I"l
0
0
0
0
0
C'I"l
0
C'I"l
0
<r
0
o .
........
...-I
...-I
0 · · · · · · ·
0 0 0 U"\
oj(
...-I 0 0
I
...-I
S
1-1
ClJ 0 0 0 ...-I 0 0 0 0
0 0 U"\ 0 0 U"\ ...-I 0
1
Z
00 0
0 · · · · · · · ·
0
0
0
0
0
0
N

N
\.0

0
0
0
0
0
ClJ
...-I

~
CIj U"\ 0 0 0 0 0 0 \.0
Cl) \.0 0 0 0 N
r-,
U"\
'" '"
· · · · · · · ·
N 0'1 .j" 0 U"\0 \.0
'" 0 ...-I ...-I 0 U"\
oj(
0 0 0

0 0 0 ...-I 0 0 0 -:t
0 0 0'1 0 0 C'I"l N ...-I
C'I"l
· · · · · · · ·
0 0 0 0 \.0 0 ...-I
C'I"l
0 0 0 0 U"\ 0 0 0
oj(

0'1 0 0 ...-I 0 0 0
N 0'1 0 0 0 0 ...-I '"
0

· · · · · · · ·
0 0 0'1 0 0 00 0 0
N
0 0 C'I"l 0 U"\ 0 0 0
oj(

0 0 0 0 0 0 0 .j"
0 N U"\ 0 0 0 0 N
...-I C'I"l
· · · · · · · ·
0 0 0 N 0 0 0
0 0 0 0 C'I"l 0 0 0

~ + + + + + + N +
0
N N N C'I"l N N +
;::l .,-l 1-1~ ..c "0
H
U Z N U f p., U
C'I"l"
:>
 APPENDIX C

Mineral Analysis
 192

M
M M I M M
(I)
.j.J M
.~
I
c::
I
c::
M
I .~c:: I
.~ s::
I
.gs::
cV
p::
I
.~ c:: 13 M
Is::
'a M
I .~ M
I
s:: 13 M
I
M
I
M.-4
s:: 1s::1c::
.-4
I 13
.-4 .-4.-4
I I I
.-4 .-4
I
s:: s:: s:: Is:: s::
.-4
I
s:: I3CD
N
.~ CD .~
s:: 13 S::CD
.~ N
S
.~
s::
.g.g .g .g ~
s::
.~.~ .~ .g .g c::
.~ CD
cV SN SCD 13 13 13 13 SN
t) CDO 0 0 0 0
tf.) NLI"'I CDI.I'\ CD~ CD 1.1'\ CO CO CO CO col.l'\ CO CO CO CO CO CO 1.1'\
N NN NN NN N NNN NN NNN NN NN
0 0 0 0 0 0 0 0 0 0 o 0 0 0 0
.-40 .-40 .-40 .-40 .-4 .-4 .-4 .-4 .-40 .-4 .-4 .-4 .-4 .-4 .-40

..c::
.j.J CO CO CO CO CD CD
.~
"0
~ 0
CDN
NO
M 0
CDN
NO
M 0
CDN
NO
M 0
coN
NO
M
N
0
CO CO CD CD
NNN
0 0 0 0
coN
NO
M
CO CD
NNN
000
CO
0
CD CO
NN
0
CON
NO
0 M
1.1'\ M I.I'\M 1.1'\ M 1.1'\ M 1.1'\ I.I'\OM LI"'IM I.I'\OM OM LI"'IM
s:: --r I --r I --r I --r I --r --rC"1.-4 --r I --rM.-4 MM --r I
cV 10 10 10 10 I I I I 10 I I I I I 10
(J 0 0 <o 0 0 o 0 <o 0 o 0 0 0 0
tf.) --rN '" --rN --rN '" 0
--rN '" 0
--r --r--r--r
0
--rN --r --r --r --r --r --rN '"

(I) ,-... ,-... ,-...

~
.._, ~
.._, o
.._,
~ 1-1 1-1 1-1 1-1 1-1 1-1
~ .j.J
cV cV cV cV cV cV
Z
§ 1-1 1-1 1-1 1-1 1-1 1-1 1-1 1-1 1-1 1-1 ~ ~ 1-1 1-1 1-1 ~ ~ ~ ~ 1-1 1-1
~ 0
(I)
"0"0
(I) (I)
"0"0
(I) (I)
"0"0
(I) (I)
"0"0
(I) (I) (I) til til (I)
"0"0
(I) (I) til til til til (I) (I)
"0 '0 "0 "0"0
~ ::E: )) ))
~ ~
) ) ) ) ;::l. ;::l. )) ) ;::l. ;::l. ;::l.;::l.

ga o 0
p..p..
o 0
p..p..
0 0 0 ON N o 0 ON N NN ~ ~
p..p.. p..p.. p.. p..V V p..p.. p.. V V V V p..p..
~
A
13
~ 0
1-1 (I)
Z ~ (I)
0 "0 1-1
H (I) >. (I) (I) t) (I)
~ s:: s:: (I) .j.J (I) .j.J .j.J 1-1 til 13 til (I) t)
p..
H
p::
M
cV
0 0
.j.J N
tIl·~
1-1
00
cV cV
C::N
.j.J
cV
(J (I)
cV .j.J
.j.J cV
.~ c:: (I)
..c:: 0
.j.J 1-10
til t) (I)
t)
(I)
t) (1)"0
.-4 (I)
.j.J
cV~
cV

u .~ "0 1-1 .j.J

0
,.c c:: 0
s::
c:: c::
0
•..., c:: cV (JM ~~
cV cV
cV 1-1
s::
0 p..
tf.) 1-1
~] .~ 1-1 cV .0 o 0 (I) cV 1-1 ..c:: (!) .0 1-1
(I) fo cV..c:: 1-1 (J,.c 13 ~ )
0
(J ) ::l til > 1-1 cV
~
r:<l
.j.J
cV
::E:
. ..c::
"0
(I)
.-4
~
t) .j.J

>.
cV
t) s:: cV
.~
1-1
t)
o
4-l _
1-1.0
1-1
~
o
0
13
1-1 til
o
.-4
(I)
til
)
0
s:: t)
cV t)
t) til

H til .j.J ~ 0 (I) ~ >. ~ s:: .0 0 ~ til ~

4-l (I) c:: 0 1-1 0 (I) (l)N (I) ~ (I) 0 .-4 "0 til "0 0
0 1-1 (I) "0 c:: 00 1-1 M .-4 .-4"ON M (I) cV (I)
~
tf.) s::
~ 13
(I)
s:: s::
o cV .~
0
1-1
s:: 00
0
~~
..c::
cVM
..c:: (I)
~ (l)o~
p..1-1
(I)
(J .~13
cV.o s:: 1-1
(I) 00
1-1 p..
(I) 0

.~ 0
>'1
t) .~ .0
.j.J
.j.J (!)
I-I~
.~ .j.J
.j.J,.d
tIl~
0
til >. cV cV 0
tIll-I..c::~
cV 1-1
(I) t)
~
e~
,l:l.j.J
.j.J
.~ .~
U
.j.J M>.
(!)M
>M
1-1 (I)
o .j.J
p..cV
o
p..cV
1-1
"0
1-1 00
O·~
p..M
"0
(!) p..
1-1 0
"0
1-1
(I) 0"

(I)
00
c::
• t)
(!) til til ::l
00
1-1
::l til
.j.J
~~
.oM
(I)
cV 13
(I) 0
) 1-1
1-1
t)
til
• .-1 cV
.j.J .~
cV .j.J
1-1 0
(!).o
s:: (!)
1-1 0"
.j.J'O
1-1
(!)'O
.. (!) .j.J
,.d
.j.J,l:l
(!)
M
p..
(I)
,.d
.j.J
t).~
cV c::
~.~
cV 0" 0
1-1 (!) 1-1 )
(!) M (I) 0
(!)
(!) 13
t)
M
>'.0
..c:: 13
c::
)'0
~
(I) M 1-1 p..1-1 c:: s:: ) c:: cV .j.J 13 cV 1-1 cV >cV13M 1-1 0 000 o s::
A (!) cV
p:: p..
p..cV
p
(!) cV o cV (!).~ cV (!) ::l .j.J <,.d::l(!) (J 1-1 .~ 1-1 1-1 cV
t) U.o H.o ~ ) til ~ til til tIlS::.o tf.)~ ::c~ P::).o

(!)
t)
(!)
t)
(!)
t)
(!)
t)
(!)
t)
(!)
t)
(!)
t) .
~
cV
~
cV
~
cV
~
cV
~
cV
~
cV cV
~ .~
.j.J

. 1-1 1-1 1-1

.1-1
. 1-1 1-1 1-1
. 1-1
. 1-1
§
s:: 0 0 0 0 0 0 0 0 0 (!)-
.~0
.j.J
~ E-l ~ E-l E-l E-l E-l ~ E-l
.~
> ......

cV § ~
13
cV ~ ~
13
cV ~
13
Cl!
til (!)
til .j.J
t) ::E: ~ ::E: ::E: ::E: ::E: ::E: ::E: ::E: (!).~
0 I-Itf.)
H (!) (!) (I) (!) (!) (!) (!) (I) (!) 00
.-4
.j.J
M
.j.J
M
.j.J
M
.j.J
M
.j.J
M
.j.J
M
.j.J
M
.j.J
M
.j.J
o (!)
1-1 >
.j.J
.~ .~
.j.J
.~
.j.J
.~
.j.J
.~
.j.J
.~
.j.J
.~
.j.J .j.J
.~.j.J .j.J 0

.
H
.H
.
H
.
H
....:I
.
H
.
H
o~

. H
. .~~.._,
H

(J
N
E-l
E-l
 193

~ ~
I~ I~
7~ 'g ~ .~ ~
I
~
I
~
I
~ ~~
I I I
~
I~ El I
.~
aN
CD .~
aN
<D .~
~
aN
<D .~
El
~
'g
~ ~ ~ ~
'8'g 'g 'g
~
o o o
CDLt') CDLt') CD Lt') CD CD CD <D <D
NN NN NN N N N N N
o o o o o o o o
~o ~o ~o ~ ~ ~ ~ ~

..c:
.j..J CD <D <D
.~
"0
~
CDN
NO
0 C"")
CD N
NO
0 C"") 0
CD N
NO
C""l 0
CD
N
0
N
<D

0
CD
N
0
CD CD CD
NNN
0 0 0
CD
N
<D CD
NNNNN
0 0 o
<D <D <D

0 0 0
N
<D

Lt') C""l Lt') C""l Lt') C""l Lt') Lt') Lt') Lt')000 Lt') Lt')Lt')0000 Lt')
~ ...:t I ...:t I ...:t I ...:t ...:t ...:t ...:tC""lN ...:t ...:t...:tC""lNC""l ...:t
til I 0 10 10 I I I I I I I I I I I I I
(J 0 \0 0 \0 0 \0 0 0 0 o 0 0 0 0 o 0 o 0 0
tI) ...:tN ...:tN ...:tN ...:t ...:t ...:t ...:t...:t ...:t <r ...:t...:t...:t '...:t...:t ...:t

QJ
c, ,... ~
til cO
-o
.._,
~
cO
QJ
atil .._,--
t!>::c.._,
,... ,... ~
til til til
E-t QJ QJ ~
.j..J a a 0 ~ ~ ~
,... ,... til til
,... ~ til til til
~ ~ ~ ,...
~ ~ ~ ~ ~ ~ ~
.
;:l QJ QJ QJ QJ QJ QJ QJ QJ QJ QJ ;::l.;::l. QJ QJ QJ ;::l. ;::l.;::l. QJ
0 "0"0 "0"0 "0"0 "0 "0 "0 "0 Lt') Lt') "0 "0 ~ Lt') Lt') Lt') "0
~ )) )) )) ) ) ) ) 0 )
o
p.,p.,
0 o
p.,p.,
0 o 0 0 0 0 ~OO ~ O§OOO 0
p.,p., p., p., p., Po< v v p., p., v v v p.,

til "0
QJ
(J
.j..J til
~ ,...QJ
a ~ .~
... .~
til "0 .j..J M QJ
~ ~ ~
~
~
~
0
)
til
..c:
til
M
! til
.j..J
~ til
"0
~
til til
.~
~
~
~
QJ
"0
~
til 0 ~ til .j..J .!:<: ~
.~ "0 ~ >..~ ~
~
0
(J ~ QJ (J.Q
QJ
cd
cO
~ ~ QJ ~ >. ~ 0.j..J '2cd QJ
QJ
.j..J
cO
cO
,.0
"0,.0
,...
QJ
QJ
QJ~
>
a §
QJ
r-l ~
cO cO ~
0
til !
cd
,...
cd
..c:
til .
~
,.0 QJ
cd
) r-l
..c:
til

>.
~
cd
..c:
~ QJ
.j..J
QJ QJ
~
.Q .~ <: .~
cO QJ til
..c: o .~ til ~ ~ ~ QJ QJ ~
~ cO .j..J (J ~ .j..J QJ r-l (J ~ M "0
~
0
~
0
,.0 QJ
cd
QJ
)
til

a
~

~ (J
~
QJ
&~a
o
•
...:t
r-...
QJ
~
0
~
~
,.0
til cO
..c:~
til ~
QJ
.j..J
cO ,...
.!:<:
~
QJ
.j..J
~
.~
0
.j..J
cO cd
(J
~ ~
0 QJ ~ QJ
.!:<: 0 QJ
0,... (J ~
~
,.0>'
_"0
::c
~
.j..J
til
I "0
"0 ::l
a
QJ
::l
til
a"O cd
"0
QJ

So
(J4-I 0 4-1 "0 QJ
~ > >.~
.~ .j..J ~
(J >."0 0 'M ) "0 "0 cO
>.~ til QJ
.j..J .j..J
,.0 QJ til cd QJ til0
§~ QJ ~ QJ
~ .j..J ~ ~ ~ ~
(J
QJ
~
0
~ cO til cO~ M ~
til
~.~
"0 ~ tIl~
m~ til ~
.j..J
~ 0

A
til
QJ
MQJ
~
.!:<: .j..J
QJ

So]
~ QJ
r-l
~,.c ,...
~ 0
0
QJ
~.~ .~
~
~ .j..J
til QJ
~ til,...
cO~
>.>. QJ
>. ~
M

!
>..j..J
r-l
QJ
til
.j..J
~ ~
cO
A
QJ
(J
cO
,... '"'cO
~ (J
til cd cO
tI) (J,.c
til
m
tf.l
.~
El QJ
til tI)
cO
0
(J
~QJ
t!> >
'"'
QJ
~
"""0
o ~
cO
~~
~
~"O
cO
cO
~'"'
~~
."", -,-4
~ til

.~ .j..J .j..J .j..J

-
'M 'M
~ ~
~
'M
0
.j..J
cO
.~
>
§
QJr-... - -
til .j..J
QJ
>
'M
::l
QJr-...

til .j..J
QJ .~
::l
QJ
> r-... -. -. - - -
a r-...
a r-...
a r-...
El
LI') r-... a r-... - El r-...
Lt')
N
a

til 0'\ 0 0
(J tIl'M tIl'M til 0 ...:t LI') r-...
0 QJtf.l
~ ,...
QJtf.l
,...
QJ QJC""l QJ\O QJO'\ QJr-l QJ~ QJr-l QJ ~
~ O()QJ
r-l r-l r-l r-l r-l r-l ~
MQJ M o..c: o..c: o..c: o..c: o..c: o..c: 0 ..c:
o >
~ 0 ,...o 0> ,...0 ..c:.j..J
QJ p..
..c:.j..J
QJ p..
..c:.j..J
QJ p..
..c:.j..J
QJ p..
..c:.j..J
QJ Po
..c:.j..J
QJ p..
..c: .j..J
QJ p..
.j..J,.c .j..J,.c .j..J QJ ~ QJ ~ QJ ,... QJ

~'"' .._,
QJ

~'0"0
"' .._,
QJ cd QJ cO '"' QJ '"' QJ
QJ 0"0 0"0 0"0 0"0 0"0 0"0
p:::.._, p:::.._, p::: ~.._, ~ .._, ~ .._, ~.._, ~.._,

QJ 0 0 ...:t Lt') r-...

~ 00 0'\ ~ ...:t \0 0'\ ~ ~ ~ ~
r-... r-... r-... r-... r-... r-... r-...
~ ~ ~ ~ ::c ::c ::c ::c ::c ::c ::c
cO ~ ~ ~ !X:l !X:l !X:l !X:l
tf.l
 194

r-I ..-I r-I r-I r-I r-I

Ir::lr::lr::lr::lr::lr::
'.-1 • .-1 -.-I -.-I -.-I -.-I
13 e 13 13 13 13
CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD
NNN NNN NNN N NNN NNNNN NNNNNN
o 0 0 000 000 o 000 00000 000 0 0 0
r-I r-I r-I r-I r-I r-I r-I r-I ..-I r-I r-I r-I r-I r-I ..-I r-I r-I r-I ..-Ir-Ir-Ir-Ir-I..-I

CD CD <D CD CD CD CD CD CD <D CD CD <D <D CD CD CD CD CD CD CD CD CD CD

NNN NNN NNN N NNN NNNNN NNNNNN
o 0 0 o 0 0 000 o,.... o 0 0 0 000 o 00000 0
00Cl0Cl 00Cl0 000 000 ~ONOO ~ONOOO
C")NN C")NC") C")C")C") ~ C")C")C") ~C")C")NN ~C")C")NNN
I I I I I I I I I I I I I I I I I I I I I I I I
o 00 o 0 0 o 0 0 o 000 o 0 000 000 0 0 0
~~~ ~~~ ~~~ ~ ~~~ ~~~~~ ~~~~~~
,... ,...
et! et!
~ ~
til
""' til ""'
r:: r::
o o
• .-1
.jj 1-1 -.-I 1-1
,!:G CJ et! ~ et!
CJ
o
,... ~ ~ et! ~
44 til .t: til

Q)
. ..
;:l.;:l.;:l. ;:l.;:l.;:l. ;:l.;:l.;:l. ;:l.;:l.;:l. ~;:l.;:l.;:l.;:l. ~;:l.;:l.;:l.;:l.;:l.
If''\lf''\lf''\ If''\lf''\lf''\ If''\lf''\lf''\ r-I If''\ If''\ If''\ :> If''\lf''\lf''\ :> If''\lf''\lf''\lf''\
o et! • ct:J •••
000 000 000 000 Q)NOOO Q)NOOOO
V V V V V V V V V $ V V V =z::V V V V =z::VVVVV

g
1-1
44
Q)
r-I
et!
,..c::
til

"t:I
Q)
.jj
,.c:: r::
til

1-1
44
Q)
!
et! 1-1
1-1 Q)
44 >,
.jj et!
CJ >,r-I
et! r-I
Q)44
~ r:: ,...
o • .-1 ::l
u ~.jj

til til
.jj .jj
-.-I -.-I
r:: r::
Q) ;:l ;:l
r:: CJ Q) Q)
0 ""'
13 I'd :> Q) :> al
-.-I >, 44 o :> 0 :>
.jj tOtO to ""'
13 to ""'
13 .jj 1-1
,... 1-1 ,0 -.-I ,0 -.-I
;:l et!
et!
CJ \0 N ,.... r::
et! til
til
til
I'd til
til
0 al..-l al N alN ~
Q)
;:l .jj Q) .jj al
..,.:I r-I ..-I r-I 0- al -.-I 1-1 -.-I I-a
O,.c:: 0 ,.c:: O,.c:: U I-a P-OO P-OO
,..c::.jj ,.c:: .jj ,.c::01-1 til ;:l 0 0
Q) P- al P- al p.. Q),!:G ..-I r-I I-a ..-I I-a

~- ~- ~-
I-a Q) I-a Q) I-a al p..,... -.-I -.-I 01-1 -.-I 01-1
O~ O~ O~ o 0 I'd 0 Q) 0 al
=z::~ ~ Ul I-a Ul I-a
 195

..-I ..-I ..-I ..-I

I~ I I~ I~
.,-l
~
.,-l .,-l .,-l
S s S S
CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD
NNNNN NNNNNN N N N N N N N N
o 0 0 0 0 o 0 0 000 o o o o o o o o
..-1..-1..-1..-1..-1 ..-1..-1..-1..-1..-1..-1 ..-I ..-I ..-I ..-I ..-I ..-I ..-I ..-I

..c
+J
CD CD CD CD ID CD ID
'0 CD CD CD CD ID CD CD CD ID CD CD CD
N N
.,-l NNNNN NNNNNN N N N N N N
t3 0 o 0 0 0 0 0 0 000 0 0 0 0 0 0 0 0
..;tONOO ..;t..;tNOOO 0 LI"'I N LI"'I <r 0 LI"'I 0
~ ..::tMMNN ..;t..;tMNNN M M M M ..;t M ..;t LI"'I
t1l 'I 1 I 1 1 1 1 1 I 1 1 I 1 1 1 1 I 1 1
t) 0 0 0 0 0 0 0 0 000 0 0 0 0 0 0 0 0
CI.l ..;t..;t..;t..;t..;t ..;t..;t..;t..;t..;t..;t ..;t ..::t ..;t ..;t ..;t -e ..;t -e

_.-
~
~
t'd

~
Cl)

~ c.!>::r:
t'd

~
Cl)

~
_.- -
c.!>::r:::r:
.....................
Q) 0 .............. 0
• ,-l .,-l
+J ~ ~ ~ ~ +J ~ ~ ~ ~ ~
t) t1l t'd t'd t1l t) t'd t'd t1l t1l t'd
E-t
t'd Q) Q) t'd Q)
+J ~ ~ S S ~ ~ S ~ ~ ~ ~
~ 1.1-1 {/) Cl) {/) {/) 1.1-1 Cl) Cl) {/) {/) Cl)

::l
~ ~ ~ ~ ~ ~ ~ ~
0
...
~;l.;l.;l.;l. ~;l.;l.;l.;l.;l.
~ t'd
LI"'I LI"'I LI"'I
t'd
LI"'I LI"'I LI"'I LI"'I t'd
<1.1
t'd t'd t'd t'd
<1.1
t'd
<1.1
t'd t'd

<1.INOOO <1.INOOOO S ~ ~ ~ S S ~ ~
::r: V V V V ::r: V V V V V tr.l tr.l CI.l CI.l CI.l tr.l tr.l tr.l

~
<1.1
>..
t'd
btl
~ § ~
~ btl btl ..-I .,-l <1.1 0
..-I 1.1-1 0 ~ ~ +J ~
t1l 0 .,-l .,-l .,-l Q) '0 t'd +J btl
.,-l I S +J +J +J t) ~..-I 0 {/) ~
~ <1.1 .,-l .. 0 Q) t'd t'd t'd t'd t'd t) .,-l
Q) Cl) {/) <1.1 Cl) ~ ~ 0 0 4-1 .,-l S +J
+J t'd_ p.t) <1.11.1-1 U U U ~ Q) ~ Q) 0 t'd
t'd ,.0 ~ >..t'd {/) ~ ::l t) <1.1 t) ~ 0
~ btl4-1 Cl) ~ 0 <1.1 <1.1 {/) t1l +J t'd 1.1-1 t)

1.1-1
0
+J ~
t'd'O .. S
'0 0
t'd Q)
S+J
Q)
U
Q)
(J
t'd
1.1-1
(J
t'd
1.1-1
'0
Q)
t:::l m 4-1
~
::l
<1.1
+J
<1.1
~
<1.1 S Q) ~ S
<1.1 btl ~ ~ ~ {/) ,.!Il {/) t'd .,-l
~ ..-I t) ~ 4-1 +J +J t'd ~ ::l ::s I (J +J ..-I
0 t'd Q) .,-l .,-l.,-l t1l Cl) Cl) ..c Q) t'd >.. .,-l ..-I
.,-l ,.dLl"'l ,.d <1.1 P. ..c'O..-I {/) +J
.,-l
~
(J
~ .,-l
Po t'd
+J {/)
+J {/)..::t +J..-I <1.1 .,-l Q)
....... t'd t'd..-l )~4-I ~ btl .-I
.; '0 t) Cl) s
.,-l '0 Q) ..c .,-l .-I .,-l 0 ..-I ~ P. {/) <1.1
~
>"Q)
~ <1.1 +J ) Cl) 0 t'd ..-I t'd ~ .,-l ~ ~ +J
(J ~ .,-l {/) Q) <1.1 ~ ::s {/) Cl) Po Po (J Cl)

{/) <1.1 Po >"CI) .~ m u >.. 0 p. ::s <1.1

o ~
<1.1 ..c ..-I ::s ) ~ t1l Q) <1.1 <1.1 +J
~ t'd ~
. ~::s
Q +J..-I ..c 0 0 <1.IN 4-1 +J "C '0 +J +J +J
t'd ',-l btl~..-I ..c 1.1-1 ~ ~ ,.0 ~ .,-l .,-l .,-l t'd
..-I
~
<1.1 0
Cl)
.,-l
::r:4-I,.o
<1.1 <1.1 Po
tr.l (J {/)
0
tr.l
t'd
::r: ::r:
t'd 'r-4 >..
~ Cl) $ $ $ Po

~ ~ ~ ~ ~ ~
{/) {/) 0 0 0 0 0 0
+J +J .,-l ',-l .,-l .,-l .,-l .,-l
.,-l .,-l +J +J +J +J +J +J M
~ ~ <1.1 <1.1 <1.1 <1.1 <1.1 <1.1
::s ::s ~ ~ ~ ~ ~ ~ <1.1
~ <1.1 (J (J (J (J (J (J
~
+J
0 ::- Q) <1.1 ~ ~ ~ ~ ~ .,-l
.,-l
+J
0
,.0 .,-l
::- ::-
.,-l
0
(J
0
(J
0
(J
0
(J
0
t)
0
CJ
tr.l

t'd t'd Cl) {/) ,.d +J

(J Cl) {/) <1.1 <1.1 <1.1 Q) <1.1 <1.1 (J t'd
0 +J Q) <1.1 +J +J +J +J +J +J +J
~ .,-l ~ ~ t'd t'd t'd t'd t'd t'd .,-l <1.1
PobO bO ~ ~ ~ ~ ~ ~ CI.l btl
0 0 0 0 0 0 0 0 t'd
..-I ~ ~ ,.0 ~ Po
.,-l +J +J of:t'd ~
t1l
of:t'd of:t'd of:t'd of:t'd .,-l
'0
<1.1
0 <1.1 <1.1 <1.1
tr.l ~ ~ u u u u u u 0 CI.l
 196

QJ
-1-1
t'd ...-I ...-I ...-I
p::: I I I
t:l s:: t:l
~ -.-I -.-I -.-I
t'd S S S
()
Cl) CD CD CD
N N N
0 0 0
...-I ...-I ...-I

,.c::
-1-1
"0 CD CD CD
-.-I N N N
:3 0 0 0
LI"'I LI"'I LI"'I
t:l ..:r ..:r ..:r
t'd I I I
0 0 0
~ ..:r ..:r ..:r
,.c::
-1-1
-,-4
~
III
"0 (I)
QJ -1-1
-1-1 ::I
QJ t'd ~
~-.-I ~ ~
?-> ::I S ::l ::I
~ -1-1 0 0
-1-1 t'd0 ,.c:: ,.c::
t:l IIlN
::l ~ ~ ~ ...-I ...-I
0 t'd t'd t'd QJ ~
::<:: QJ ~ 0 ~ ~
~
Cl)
S
Cl)
~
Cl)
(1)1.14
,.c::
0
1.14
0
1.14
o..U
1Il0 U U
"0 0 0
QJ ~~ 0 0
.0 -1-1 LI"'I LI"'I
t'd -1-1 N LI"'I
t'd
~ 13 t:l -1-1 -1-1
...-I QJ () .,-4...-1 t'd t'd
t'd ~ 0
.,-4 -1-1 0 "0 () "0 "0
~ III ...-I 13 QJ
()...-I
:>. QJ
-1-1
QJ
QJ C) -1-1
-1-1 t:l S t'd co t'd t'd
t'd 0 0 0 ...-I QJ QJ
;:E: ~ ..:r p..QJ ,.c:: ,.c::
QJ 1.14 QJ QJ t:l
1.14 ~ ~ 13 ~ QJ QJ QJ QJ
0 ,.c:: QJ 0 0 0 r-I r-I r-I r-I
() ~ C) ~ ()
t:l 0 ,.c:: 1.14 ~~ ~ ~
0 () ~ ~ t'd -1-1 t'd t'd
• .-1 "0 0 QJ QJQJ Cl) QJ Cl) Cl)
-1-1 (I) co ~ co
~ '"Cl ::l ,.c:: ::I
,-..
,-..
,-..
I QJ t'd tU
.,-4
~
()
,.c::
III
~
I ~
()
0
~
~
........
r::'
........ ::r::
........
III .,-4 QJ.,-4 QJ.,-4
QJ ...-I co co
~ p..
~§ t:l t:l
~ t'd ~
::l ~ 0 ~ 0 III
p., 0"0 0"0 r-I
0

i:>.
Cl)

1.14
0
~ ~ ~
t:l
QJ QJ QJ 0
t:l -1-1 -1-1 -1-1 .,-4
0 • .-1 .,-4 .,-4 -1-1
• .-1 Cl) Cl) Cl) t'd
-1-1 t:l
t'd -1-1 -1-1 -1-1 t'd
() t'd t'd t'd ...-I
0
...J QJ QJ QJ ~
00 co co ~
t'd t'd t'd
p.. 0.. p..
QJ QJ QJ
QJ QJ QJ
Cl) Cl) Cl)
 197

)l
>4
r
I SnORcnIOWV XW-X
I I I ... ,
R >4>4>4
~ e >4>4><
Q) ,[
P- >4>4>4
... 3
a
><>4><
~ >4 >4
..... J
...-l ><
en H ><
>. V >4>4><
'~"'
Q) IS'[ >4 >4 >4 >4 >4>4 >4 ...
...
H1JR >< >< >< >< ...
68RH >4 >< >4 >< >< ><
g9RH >4 >4 >4 >4 ...
~ 98HH >4 ... >4 >4
Q)
P-
Z!) >4 >4 >4 ><
CIl
~
't:I
0
0
cas ... >< >< >< >< >4 ><>4><
H OD ZdS >< >< >< >< >< ><
~ >4 IdS >4 >4 >4 >< >< >4
CIl
><
~
~
L HHH >< >< >< ><>4 >< ... >4
0
cd
Q)
£"HHH >< ... >< >4>< ><>< >< ... ... ><
~
O
P- f]HHH >< >< >< ><><>< >< ... ><
OD
~ OHHH >< >4>< >4 >4 >4 >4
~ 0
...en'"'
6LHH >< >< >< ><>4 ... ... ><
~ LLHH >< >4 >4 ><>4 ... ... ... >4
~ 9LHH >< >< >< ><><>< >< ... ><
.....
>4>4 f]LHH >< >< >< ><>< >< ... ><
.....
~
U
cd
Q)
OUM
6.LW
... ... ><>< >4>< >< ><><
>< ><
P-
>< >< ><
OD
S.LW
...
>< ><>< >< >< >< ... ><
~ L.LW
... ><
...
>< ><><>4 >< >< ><>< ... ><
0
...'en"' 9.LW ... >< >< >< ><><>4 >4>< >< >< ... ><
£".LW >< >< >< ><~ ... >< >< ><
>. f].LW >< >< ><>< >< >< >< ><
'~"'
Q)
£.LW >< >< >< >4>< ... >< >< >< ><
xZ.LW >< >< >< >< ><>< ><
>4>4>4 ;)Z.LW ... ... >< ><>4>< >< ><>< ... ><
QZ.LW ... >< >4 ><><>4 >< >< ... >4
ez.LW ... ... >< >< ><>< >< ><>< >< ><
1.LW >< >< ><>< >< ><>< ><>< >4 ><

........
Q)

........ 'P-"' ........ ... ........ ....en....

Q)

........
Q) Q) Q) Q)
~ Q)

........
cd Q)

... .....
........
e-.
.....~
........ ,..c::...
...-l Q) N
u en S
Q)
S
'"'
0 ...-l 't:I '"'Q) '"'
Q) 0 C.J
;:l
~
'ell;::I"' 0 en
...-l
,..c::
u
~
N
...-l
...-l
...-l
H
0
ell
::.:: et
...-l

~
Q)
't:I
.....
CIl ~
...-l

~
0
...-l
ell
U
'>."'
P-t ~
Q)
0 '"'
~
cd
~
p.
>.
0
N
.
 198

H I ..::to I ...-I I I
Q) QJ QJ QJ ...-I
<) <) <) <)
H C1l C1l C1l C1l
I-< I-< I-< I-< I I
~ E-<E-<E-<E-< 0\
N 00
::c
~ r:Q
,..., 0\ 00 00

oc-
'-'
'0
M...-INN
...-I.......::tN
····
MNNN
I I
<
'-'
'0
I
100
·· ·
Lr'I Lr'I
I
0
M
I I

QJ I 10 OMLr'I I I
<) H ...-I
H C1l Lr'I Lr'I 0
I-<N..::t 0 I I
E-< ...-I

-
<)
N
...-I
...-I ~ p...

-
\0 Ul
0\
...-I ...::tON...-I ,..., ...-100...-100

··. ·
oc <
····
...-I........::tN .....0...-10
'-' I I '-' I I I
'0 MNNN '0 Lr'ILr'IMM
~
~
r:Q

p:: Lr'IOO
~ H I NLr'lO I I
E-< ...-I
I Lr'IO I \.0 I I
~ .c H ...-I
'-'

Ul
~
....:l
p::
0
E-<
N
~
-
O<l:l
O\NO
\O...::t N
-
,...,\.0
::r:
N
p...

,...,
'-'
'0
I
NNN ··· I I 0
'-'
Ul

~
Ul ~
~
N
Ul
QJ
-
0
N

...::t ,...,
<
...-I00
.....0
00
0
~
~
~
~
~
'-'
H
000
I M\.OO I I
Cl)
'-'
M
'-'
'0
I
·· ·
Lr'I Lr'I
I
M
I I

Z QJ ...-I QJ
H Ul +J ~
::.:: ~ -.-I C1l ~
....:l I-< N '-'

-
>..
p... ~ QJ
~
p...
~
Ul . O<l:l
...-IN...-I
I ........::tN
+J

··· I Lr'I 0 I 00 N...::t

-.-I
...-I ..._"
I I til H ...-I
~ 0 '0 N NN 0
0 ~ I-<
E-< C1l
Z U I-)
0
Ul
H
~
....:l
~ 0000 N
. N
o
p:: Ul H ...::to\.OLr'I I I
< ::c ...-I
,...,
·· ·. .
...-I00 00\.0 00
~ E-< ...-I < .....0 000 N
0 H '-' I I
U ~ ~ '0 Lr'I Lr'I MN N

'.-1
• .-1
-
O<l:l
'-'
'0
...::t...-lN...-I
...-I........::tN
····
MNNN
I I
-.-I
U

\0 ...::t\.0 NOO
H "-- Lr'IOOOO 000
...-I
00

H MOO\.O Lr'I...::tO r::

...-I ~
~ 0
I-< I-<
r:Q r:Q

-
O<l:l
'-'
'0
M...-IN
·. . . . .
...-INM
...-I........::tNO\\.O
MNNN...-I...-I
-
<
'-'
'0
.......
MOO\...-I .....Lr'lOO
O\ .....O...-lOOON
Lr'ILr'ILr'IMMNN

,--. ~ ~ '""' ~ .,.......,. ~,.......,......._,.........._. ........... ,......,......

...-I ...-100...-10...-1 ...-IMN...-IM\.O .....
~ ...-IO...-l...-lN...-I OO...-lN...-IOO
.c ...-INNNNM
...._, ...._, '-'...._,...._,...._, ...-1000...-10...-1
...._,...._,...._,'-"...._,'-''-''
 199

(lJ (lJ
() ()
01/') I/') (1j (1j
H
ON
0
C""l M ,... ,...
M E-t E-t
(1j
N

~ ,...._ r-, ,..._ 00

·· · · . ·
MOO 0
<
.._, \0 N 00 000 \0

"0 ,..._<r C""l C""lN N

0 M I/') N
H 0 I I
M

t!>
,...._ \0 ,..._ 0'1 \0
,..._
-< I/')N
.._,
"0 ·· · ·
,..._..;t C""l C""l
0
I I

01/') 0 00
H 0 M I I
M

-
0
N
::x::
)::I:..

-<
.._,
\0 ,..._ 0'1
I/')N
·· · ·
,..._ \0
0
N
.
-:t
"0 ,..._..;t C""l C""l
I I

0
til
(1j
I/') I/')
.._,
U
H
01/')
OM N N I I
M

~
III
c, -<
o
>.

C""l
. -
-<
.._,
"0
\000
I/')N
·· · ·
,..._..;t
0'1
,..._
C""l C""l
,..._
0
I I

00 I/') 01/') I/')

H ON C""l C""l
M

::x::

-
-<
.._,
"0
\000

·· · ·. ·
I/')N
,..._..;t
0'1
,..._ 0000
000
C""l C""lN
00
\0
N

M ,..._r-,
H 8~
M
N I/')N
00
N

~
,...
0
~
--<
.._,
"0
\0 ,..._ 0'1

·· · ·. ·
I/')N
,..._..;t
,..._ \0 ,..._ 00
000
C""l C""lN
\0
N

M
~~""'~""'~,-...."'"
OIM M OIM N N M
~ NNC""l-:t-:tONI/')
..c:: OMOOMOOO
'J'-""'_''-''''_''''_'''_''''_'