Management of Environmental Quality: An International Journal

Download as pdf or txt
Download as pdf or txt
You are on page 1of 17

Management of Environmental Quality: An International Journal

The application of Kruskal-Wallis technique for flood prediction in the Niger Delta,
Nigeria
A.E. Gobo T.K.S. Abam F.N. Ogam
Article information:
To cite this document:
A.E. Gobo T.K.S. Abam F.N. Ogam, (2006),"The application of Kruskal-Wallis technique for flood prediction
in the Niger Delta, Nigeria", Management of Environmental Quality: An International Journal, Vol. 17 Iss 3
pp. 275 - 288
Permanent link to this document:
http://dx.doi.org/10.1108/14777830610658692
Downloaded by University College London At 03:37 28 June 2016 (PT)

Downloaded on: 28 June 2016, At: 03:37 (PT)


References: this document contains references to 16 other documents.
To copy this document: permissions@emeraldinsight.com
The fulltext of this document has been downloaded 859 times since 2006*
Users who downloaded this article also downloaded:
(1997),"Increasing flood risk in Malaysia: causes and solutions", Disaster Prevention and Management: An
International Journal, Vol. 6 Iss 2 pp. 72-86
(1998),"Mitigating flood losses in the active floodplains of Bangladesh", Disaster Prevention and
Management: An International Journal, Vol. 7 Iss 2 pp. 113-123
(2005),"Restoring straightened rivers for sustainable flood mitigation", Disaster Prevention and
Management: An International Journal, Vol. 14 Iss 1 pp. 6-19

Access to this document was granted through an Emerald subscription provided by emerald-srm:368933 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.
*Related content and download information correct at time of download.
Downloaded by University College London At 03:37 28 June 2016 (PT)
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1477-7835.htm

The application
The application of Kruskal-Wallis of Kruskal-
technique for flood prediction in Wallis technique
the Niger Delta, Nigeria
275
A.E. Gobo and T.K.S. Abam
Institute of Geosciences and Space Technology, Rivers State University of
Science and Technology, Port Harcourt, Nigeria, and
F.N. Ogam
Mathematics and Computer Sciences Department, Rivers State University of
Science and Technology, Port Harcourt, Nigeria
Downloaded by University College London At 03:37 28 June 2016 (PT)

Abstract
Purpose – To evaluate and examine human perception of causes, frequency, duration, impact,
adjustment patterns and local attempts of control, protection and flood prediction in Niger Delta. This
will enable rural dwellers to appreciate some local flood control measures.
Design/methodology/approach – Through questionnaires administered in reclaimed areas of
urban centres and flood prone communities in the Niger Delta. The data abstracted from
questionnaires were then analyzed through Kruskal-Wallis Function. The Kruskal-Wallis approach
was used as it takes care of large data points, which consists of nominal or ordinal data.
Findings – Human perception of flooding as regards impact and local attempts at flood prediction
differ among Niger Delta States. The study highlights the socio-economic implications of flooding as
regards to causes, effects, control and predictive measures.
Practical implications – Assist rural dwellers on cheaper local and emergency measures such as
use of sand bars, opening up of creeks and assess, cleaning of drainages to allow free flow.
Originality/value – The paper suggests continuous enlightenment programs as a means to
encourage local and emergency measures to be adopted when flooding occurs. The work is original as
no such work or analysis had been carried out in the Niger Delta in the past. The paper has provided
raw data and knowledge, and adds to the limited literature in the Niger Delta. On the flooding cycles of
the Niger Delta. It should also raise the awareness of local dwellers on the requirements for flood
emergency response and adjustment.
Keywords Floods, Rural areas, Nigeria
Paper type Case study

Introduction
A flood is defined as any relatively high stream flow, which overtops the natural or
artificial banks in any reach of the streams. It is also considered as a large quantity of
water covering an area after held capacity is reached (Ward, 1978). According to
Faniran (1972), flooding results when an excessive quantity of water derived from
runoff and underground sources spreads beyond the banks of a stream channel.
While floods and related problems are giving the present generation much concern Management of Environmental
owing to their renewed intensities, they are not a new phenomena. According to Quality: An International Journal
Vol. 17 No. 3, 2006
Igbozurike (1993) the earliest history of flooding and attempts at predicting or pp. 275-288
controlling flooding and its effects, date back to Biblical times when God said to Noah q Emerald Group Publishing Limited
1477-7835
(Genesis Chapters 6-9) I shall send rain and their would be great flood. DOI 10.1108/14777830610658692
MEQ Sheeham and Hewit (1969) studied the major natural hazards (excluding droughts)
17,3 and concluded that flooding ranked first out of 16 disaster types responsible for either
million of dollars damages or for hundreds of persons injured or lost. While flood
disasters occur in all parts of the world, their impacts have been particularly very
severe in developing countries like China, India, Nigeria and the Niger Delta in
particular due to its low-lying flood plains, low level of technology and relatively under
276 funded rescue institutions.
Ekpete (1990) highlighted the lack of data on flow, water levels and floods in his
study on simulation of flow on the Bonny and new Calabar River system of the Niger
delta. This lack of data and insufficient studies as regards statistical prediction,
necessitated this work.
Floods in the Niger Delta have had a devastating impact on the quality of life.
It affects rural and urban residents, farm lands, infrastructures, residential/industrial
Downloaded by University College London At 03:37 28 June 2016 (PT)

buildings, roads, bridges, power plants, commercial and socio-economic activities.


As an aftermath, flooding, results in the submergence of farm lands which in most
cases leads to premature harvesting of crops with resultant poor yields and poor
market value.
Furthermore, access roads are made impassable by flood induced inundation, while,
power driven economic activities are brought to abrupt end as the floods lasts. Though
we may not be able to stop the occurrence of floods the impact could be minimized if
adequate and accurate predictions are available, which is the basis of this study.
In order to lessen the impact of floods in the Niger Delta region, it is important to:
.
determine causes, frequencies, duration and impact of flooding in the Niger Delta;
.
evaluate adjustment patterns to flooding in the Niger Delta;
.
examine local measures of control of flood problems in the Niger Delta;
.
examine local attempts at flood predictions; and
. evaluate the extent of damage to individual or community by floods.

The hypothesis for the effective study of the above socio-economic include:
Ho1. The human perception of causes, frequencies, and direction of flooding are the
same among sampled areas of the Niger Delta.
Ho2. The adjustment patterns during flooding remains the same in the Niger Delta
areas.
Ho3. Local measures of control of flood problems are similar in all areas of the
Niger Delta.
Ho4. Local attempts at flood predictions are the same among Niger Delta areas.
Ho5. The extent of damage to individual or communities by floods remain the same
among Niger Delta area.

Background information on the study area


The Niger Delta area is in the most southerly part of Nigeria. It is one of the world’s
largest wetlands with total land area of approximately 29,100 km2 but excluding the
continental shelf (NDES, 1997). Three of the states of Nigeria, namely, Delta, Bayelsa and
Rivers with a population of over 6.5 million are within the main Niger Delta complex. The application
The study also considered Anambra State (Onitsha) because of its strategic location of Kruskal-
along River Niger (Figures 1 and 2), by the Niger river system.
Wallis technique
Climate
The area displays climatic characteristics that could be classified as semi-hot
equatorial type with heavy rainfall most times of the year (Gobo, 1990). The wet season 277
lasts from March to October while the dry season lasts from November to February.
Nevertheless, the months considered as dry months are not even free from occasional
rainfall (Gobo, 1988, 1990). The mean annual rainfall ranges from 2,500 mm upland
locations to up to 4,500 mm in the extreme coastal locations. Typical mean monthly
rainfall amounts for some selected stations for high rainfall months could be as high as
Port Harcourt, July (361.2 mm) September (367.0 mm); Brass, July (447.9 mm)
Downloaded by University College London At 03:37 28 June 2016 (PT)

Figure 1.
Map of the study area
showing the stations and
towns used
MEQ September (570.1 mm); Bonny, July (492.3 mm) September (665.8 mm) and Warri, July
17,3 (464.6 mm), September (464.4 mm).
The areas are governed by the general circulation patterns in the equatorial tropics
associated with the apparent movement of the sun and the relative position of the
inter-tropical discontinuity (ITD). Two air masses also influence the climate of the area.
The first is the southwest monsoon (moisture laden) which originates from the region
278 of the tropical high-pressure belt over the southern Atlantic Ocean. From this region, it
undergoes some modifications and eventually arrives over West Africa as moist, cool,
air from a south-westerly direction.
The second is the northeast harmattan (dry wind) which has a continental origin
from the sub-tropical high pressure belt of the Sahara desert. It is a hot, dry air mass,
which also undergoes some modifications and arrives as a north-easterly wind.
The area also has high relative humidity, high insulation and associated high
Downloaded by University College London At 03:37 28 June 2016 (PT)

temperature.

Figure 2.
Flooding map of the Niger
Delta (FEPA/World Bank,
1998)
Vegetation The application
Generally, the vegetation in Nigeria is determined by climate particularly the mean of Kruskal-
annual rainfall and the severity of the dry season measured by minimum relative
humidity and length of the dry season. Accordingly, in the Niger Delta region tropical Wallis technique
forest is the climax vegetation. This vegetation of the Niger Delta area comprise the
following three major groups:
(1) 30 percent Mangrove forest and coastal vegetation; 279
(2) 20 percent Freshwater swamp forest; and
(3) 50 percent Dryland rainforest (NLNG, 1997; Teme 1990).

Soils
The Niger Delta is underlain by soils that are developed in recent (Holocene) alluvium
and beach deposits. The soils can be broadly classified into three physiographic
Downloaded by University College London At 03:37 28 June 2016 (PT)

units/regions.
(1) An area of sandy beaches and ridges along the coast.
(2) The tidal creeks and mangrove swamps with a distinct root-mat in the upper
50-100 cm depth characterized by generally poorly drained soils, sparse
population and very low intensity agriculture.
(3) The river plains/floodplain reaches in flat relief and permanent hydromorphism
including the back swamps.

The soils comprise the Entisols, Inceptisols and Ultisols (Teme, 2002) with textures
varying from mostly silty clay loam for the alluvium to coarse sands for the beach
rides. The swamps contained mostly interbedded silt and clay. The muds which are
flooded diurnally, are found mainly in the intertidal areas of the Niger Delta while the
sandy areas include the levee crest, levee slope, saline sand, beach ridge and areas of
the Sombreiro-Warri Deltaic Plains.

Topography
The areas have flat topography, high water table and acidic soil moisture regimes, the
high and frequent precipitations in the region coupled with the impermeable soils. The
soils are mostly weakly formed (with mainly A and C horizons), with weak granular
structures, and are associated with long but gently sloping land areas.

Study methodology
The research questionnaires were administered to mostly people between the ages of
36 and 50 years and those who have lived in the area or community for over 15 years.
Most of the respondents are also well educated with at least the West African School
certificate. The questionnaire technique was used to supplement other sources of
information and data. The method was useful in tapping respondents knowledge on
past flood events through the recall technique. More so, it is the most convenient means
of appreciating or gauging respondents perception of the pertinent socio-economic
variables as they relate to the study.
Ho1 was tested by using the Kruskal-Wallis test separately on data which cover
human perception on the causes, frequencies and duration and impact of flooding in
the Niger Delta area.
MEQ The Kruskal-Wallis test is used for the following reasons:
17,3 .
The data points in the contingency table are large especially the number of rows
which makes the Kruskal-Wallis test more appropriate.
.
Kruskal-Wallis like Chi-square test is used mainly on normal or ordinal data, but
since the data we analysed was ordinal, we were free to use Kruskal-Wallis
formula.
280 .
For large sample data points the use of Kruskal-Wallis test is faster and more
convenient than the Chi-square since we only need to get the sum of rankings for
each column of the contingency table and substituted into the test function.
The Kruskal-Wallis test statistic is given by:

12 X
K
R2i
Downloaded by University College London At 03:37 28 June 2016 (PT)

W¼ · 2 3ðn þ 1Þ ð1Þ
nðn þ 1Þ i¼1 ni

where: n is the size of the sampled data and K is the number of states sampled. Ri/ni is
the mean rank of the values of ith sample. This test is usually based on the
large-sample theory that the sampling distribution of W can be approximated closely
with a chi-square distribution with K 2 1 degree of freedom (Ebdan, 1985; Ogam,
2000):
.
We reject the hypotheses if the value of W exceeds X 2a;k21 ; where a is the level of
significance.
.
Ho2 and Ho3 shall also be tested by using the Kruskal-Wallis test on data, on
adjustment pattern and extent of damage, respectively.
.
Ho4 is tested using also Kruskal-Wallis test and data on remedial measures.
.
Ho5 is tested by using the Kruskal-Wallis test on data on predictive
measures.

Analyses of data
The analyses of data are shown in Tables I-VIII. We wish to test that:
Ho1A. Perception of causes of flooding is the same among the Niger Delta
states.

Anambra Delta Rivers Bayelsa

Heavy rainfalls 45 (6) 70 (3) 75 (1) 68 (4)


Absence of drainage system 73 (2) 56 (5) 38 (7) 33 (8)
Blockage of drainages 20 (13.5) 25 (12) 30 (9) 28 (10)
Unmaintained drainages 18 (16) 11 (25) 18 (16) 15 (20.5)
Insufficient drainages 12 (23.5) 16 (19) 13 (22) 20 (13.5)
Blockage of River channels 17 (18) 10 (26.5) 08 (28) 26 (11)
Location of community 15 (20.5) 12 (23.5) 18 (16) 10 (26.5)
Total respondents 200 200 200 200
Table I.
Causes of flooding Note: Figures in bracket are the joint rankings (R)
The application
Anambra Delta Rivers Bayelsa
of Kruskal-
Never 08 (25) 02 (27) 03 (26) 01 (28) Wallis technique
Every year 66 (4) 78 (3) 84 (1) 80 (2)
Every 2 years 40 (8) 53 (5) 46 (7) 50 (6)
Every 3 years 35 (9) 27 (11) 28 (10) 26 (12)
Every 4 years 20 (13) 16 (16.5) 18 (14.5) 15 (18) 281
Every 5 years 18 (14.5) 14 (19) 12 (21.5) 16 (16.5)
Above 5 years 13 (20) 10 (24) 09 (23) 12 (21.5)
Total respondents 200 200 200 200
Table II.
Note: Figures in parentheses are the joint rankings (R) Frequency of flooding
Downloaded by University College London At 03:37 28 June 2016 (PT)

Anambra Delta Rivers Bayelsa

Some hours in a day 30 (9.5) 26 (14) 20 (18.5) 18 (21)


One whole day 46 (5) 29 (11) 28 (12) 24 (16.5)
Two to three days 51 (4) 40 (6.5) 36 (8) 40 (6.5)
Four to 5 days 25 (15) 52 (3) 58 (1) 55 (2)
Six day to one week 20 (18.5) 24 (6.5) 27 (13) 30 (9.5)
One to two weeks 18 (21) 14 (26.5) 17 (23) 18 (21)
At least 3 weeks 10 (28) 15 (24.5) 14 (26.5) 15 (24.5)
Total respondents 200 200 200 200
Table III.
Note: Figures in parentheses are the joint rankings (R) Duration of flood

Anambra Delta Rivers Bayelsa

The whole household is flooded 12 (21.5) 20 (15) 24 (14) 41 (6)


Most parts of the community is flooded 36 (9) 40 (7.5) 46 (5) 68 (4)
Cause blockages on our road 27 (12) 78 (2) 72 (13) 33 (10)
Most of the farmlands are destroyed 16 (17.5) 40 (7.5) 30 (11) 25 (13)
Premature harvesting of farm produce 09 (24) 12 (21.5) 13 (20) 16 (17.5)
Leads to rapid gully/landslide formation 100 (1) 10 (23) 15 (19) 17 (16)
Total respondents 200 200 200 200
Table IV.
Note: Figures in parentheses are the joint rankings (R) Impact of flooding

Anambra Delta Rivers Bayelsa

Temporarily change accommodation 14 (20) 20 (18) 22 (17) 31 (11)


Evacuate your properties 18 (19) 24 (16) 27 (14) 35 (10)
Temporary clean up when flood recedes 50 (5) 46 (6) 40 (7) 29 (13)
Create flood channels 88 (1) 80 (2.5) 73 (4) 80 (2.5)
Others 30 (12) 36 (9) 38 (8) 25 (15)
Total respondents 200 200 200 200
Table V.
Note: Figures in parentheses are the joint rankings (R) Adjustment patterns
MEQ
Anambra Delta Rivers Bayelsa
17,3
Less than N20,000 06 (23.5) 03 (28) 01 (26.5) 04 (25)
N20,000-N29,000 13 (18.5) 09 (22) 06 (23.5) 01 (26.5)
N30,000-N39,000 21 (15) 16 (10) 11 (20) 13 (18.5)
N40,000-N49,000 33 (10.5) 28 (13) 20 (16) 01 (21)
282 N50,000-N59,000 27 (14) 30 (12) 33 (10.5) 44 (8.5)
N60,000-N69,000 44 (8.5) 50 (7) 58 (4) 53 (6)
N70,000 and above 56 (5) 64 (3) 71 (2) 75 (1)
Table VI. 200 200 200 200
Extent of damage by
flood Note: Figures in parentheses are the joint rankings (R)
Downloaded by University College London At 03:37 28 June 2016 (PT)

Anambra Delta Rivers Bayelsa

Nothing 14 (20) 20 (15) 29 (11) 19 (17.5)


Opening of blocked drainages 67 (4) 58 (7) 60 (6) 51 (8)
Construction of new drainages 77 (1) 70 (3) 73 (2) 63 (5)
Construction of temporary sand banks 20 (15) 33 (10) 22 (12.5) 47 (9)
Any other 22 (12.5) 19 (17.5) 16 (19) 20 (15)
Table VII. 200 200 200 200
Remedial or local
measures of flood control Note: Figures in parentheses are the joint rankings (R)

Anambra Delta Rivers Bayelsa

Through warning signals 54 (7) 65 (5) 60 (6) 50 (8)


Looking at the weather 68 (4) 70 (3) 74 (2) 80 (1)
Through guessing 43 (9) 38 (11) 40 (10) 33 (14)
Any other 35 (13) 27 (15) 26 (16) 37 (12)
200 200 200 200
Table VIII. Note: Figures in parentheses are the joint rankings (R)
Predictive measures Source: Data derived from the Socio-economics questionnaires administered

Using Kruskal-Wallis test function (x2a probability table in Pearson and Hartley, 1966)
on data on Table I, we have:

R1 ¼ 6 þ 2 þ 13:5 þ 16 þ 23:5 þ 18 þ 20:5 ¼ 99:5

R2 ¼ 3 þ 5 þ 12 þ 25 þ 19 þ 26:5 þ 23:5 ¼ 114:0

R3 ¼ 1 þ 7 þ 9 þ 16 þ 22 þ 28 þ 16 ¼ 99:0

R4 ¼ 4 þ 8 þ 10 þ 16 þ 20:5 þ 13:5 þ 11 þ 26:5 ¼ 93:5

With n ¼ 28. We substitute in equation (1) we obtain:



12ð5; 919:93Þ
¼ 3ð28 þ 1Þ ¼ 0:49
The application
28ð28 þ 1Þ of Kruskal-
Decision rule. Reject Ho1 if W is greater than X 2a ðK 2 1Þ; (K ¼ number of states Wallis technique
sampled) ¼ 4, otherwise, accept where a is the level of significance, at a ¼ 0.05,
X 2a ðK 2 1Þ ¼ X 20:5;3 ¼ 7:81; (from the X 2a table of probability Appendix).
Inference. We accept Ho1 since W ¼ 0.49 is less than X 2a ¼ 7:81 and infer that Niger 283
Delta states agree on their perception of causes of flooding in the Niger Delta area.
Ho1B. Perception of frequencies of flooding is the same among Niger Delta states.
To test the above hypothesis, we shall derive the following from Table II
R1 ¼ 25 þ 4 þ 8 þ 9 þ 13 þ 14:5 þ 20 ¼ 93:55
Downloaded by University College London At 03:37 28 June 2016 (PT)

R2 ¼ 27 þ 3 þ 5 þ 11 þ 16:5 þ 19 þ 24 ¼ 105:5
R3 ¼ 26 þ 1 þ 7 þ 10 þ 14:5 þ 21:52 þ 23 ¼ 103
R4 ¼ 28 þ 2 þ 6 þ 12 þ 18 þ 16:5 þ 21:5 ¼ 104
With n ¼ 28 and substituting in equation (1)
12ð5; 900:98Þ
W¼ 2 3ð28 þ 1Þ ¼ 0:21
28ð28 þ 1Þ
At a ¼ 0.05, X 20:5;3 ¼ 7:81 [from the X 2a table of probability Appendix]
Inference. We accept Ho1B since W , X 2a (that is W ¼ 0:21 , X 2a ¼ 7:81 (table of
probability Appendix) and conclude that Niger Delta states agree on their perception of
frequencies of flooding in Niger Delta.
H1C. Perception of Duration of flooding is the same among Niger Delta states.
Test statistic is the Kruskal-Wallis function equation (1) with sums of ranks for each
state as follows; from Table III.
R1 ¼ 9:5 þ 5 þ 4 þ 15 þ 18:5 þ 21 þ 28 ¼ 101
R2 ¼ 14 þ 11 þ 6:5 þ 3 þ 16:5 þ 26:5 þ 24:5 ¼ 102
R3 ¼ 18:5 þ 12 þ 8 þ 1 þ 13 þ 23 þ 26:5 ¼ 102
R4 ¼ 21 þ 16:5 þ 6:5 þ 2 þ 9:5 þ 21 þ 24:5 ¼ 101
With n ¼ 28 and substituting in equation (3.11), we have:
12ð5; 887:14Þ
W¼ 2 3ð28 þ 1Þ ¼ 0:002
28ð28 þ 1Þ
Inference. At a ¼ 0.05, we accept Ho1c since W , X 2a (that is since W ¼ 0:002 ,
X 2a ¼ 7:81 (table of probability Appendix) and conclude that Niger Delta states agree
on their perception of duration of flooding in Niger Delta:
Ho1D. Perception of impact of flooding is the same among Niger Delta States as
derived from Table IV.
MEQ Below are the sums of ranks on the data for each state:
17,3 R1 ¼ 21:5 þ 9 þ 12 þ 17:5 þ 24 þ 1 ¼ 85
R2 ¼ 15 þ 7:5 þ 2 þ 7:5 þ 21:5 þ 23 ¼ 76:5
R3 ¼ 14 þ 5 þ 13 þ 11 þ 20 þ 19 ¼ 82
284 R4 ¼ 6 þ 4 þ 10 þ 13 þ 17:5 þ 16 ¼ 66:5
With n ¼ 24 and substituting in equation (1), we obtain:
12ð5; 241:42Þ
W¼ 2 3ð24 þ 1Þ ¼ 29:8
24ð24 þ 1Þ
At a ¼ 0.05, X 20:05;3 ¼ 7:81
Downloaded by University College London At 03:37 28 June 2016 (PT)

Inference. We reject Ho1D since W . X 2a (that is since W ¼ 29:8 . X 2a ¼ 7:81


(table of probability Appendix) and conclude that Niger Delta states differ on their
perception of impact of flooding in Niger Delta:
H2. The adjustment patterns during flooding remain the same among Niger
Delta states as derived from Table V.
Applying Kruskal-Wallis test-function equation (1) we have:
R1 ¼ 20 þ 19 þ 5 þ 1 þ 12 ¼ 57
R2 ¼ 18 þ 16 þ 6 þ 2:5 þ 9 ¼ 51:6
R3 ¼ 17 þ 14 þ 7 þ 4 þ 8 ¼ 50

R4 ¼ 11 þ 10 þ 13 þ 2:5 þ 15 ¼ 51:5
with n ¼ 20 and substituting in equation (1), we have:
12ð2; 212:76Þ
W¼ 2 3ð20 þ 1Þ ¼ 0:22
20ð20 þ 1Þ
Inference. At a ¼ 0.05, X 2a ¼ 7:81 we accept H2 since W ¼ 0.22 is less than X 2a ¼ 7:81
(from the X 2a table of probability Appendix) and conclude that Niger Delta states agree
on their perception of adjustment patterns of flooding in Niger Delta:
H3. States of the Niger Delta agree on the extent of damage by floods.
Using equation (1) on data in Table VI we have:
R1 ¼ 23:5 þ 18:5 þ 15 þ 10:5 þ 14 þ 8:5 þ 5 ¼ 95
R2 ¼ 28 þ 22 þ 17 þ 13 þ 12 þ 7 þ 3 ¼ 102
R3 ¼ 26:5 þ 23:5 þ 20 þ 16 þ 10:5 þ 4 þ 2 ¼ 102:5
R4 ¼ 25 þ 26:5 þ 18:5 þ 21 þ 8:5 þ 6 þ 1 ¼ 106:5
So that,

12ð5; 896:79Þ
2 3ð28 þ 1Þ ¼ 0:14
The application
28ð28 þ 1Þ of Kruskal-
At a ¼ 0.05, X 20:05;3 ¼ 7:81 Wallis technique
Inference. We accept H3 since W ¼ 0.14 is less than X 2a ¼ 7:81 (from the X 2a table of
probability Appendix) and conclude that states of the Niger Delta agree on the extent
of damage by flood. 285
H4. Remedial or local measures of control of flood problems are similar in all
states of Niger Delta as derived from Table VII.
Using Kruskal-Wallis test-function we have:
R1 ¼ 20 þ 4 þ 1 þ 15 þ 12:5 ¼ 52:5
Downloaded by University College London At 03:37 28 June 2016 (PT)

R2 ¼ 15 þ 7 þ 3 þ 10 þ 17:5 ¼ 52:5
R3 ¼ 11 þ 6 þ 2 þ 12:5 þ 19 ¼ 50:5
R4 ¼ 17:5 þ 8 þ 5 þ 9 þ 15 ¼ 54:5
Substituting in equation (3.11), we obtain:
12ð2; 212:76Þ
W¼ 2 3ð20 þ 1Þ ¼ 1:6
20ð20 þ 1Þ
At a ¼ 0.05, X 2aðK21Þ
Inference. We accept H4 since W , X 2a ¼ 7:81 (from the X 2a table of probability
Appendix) and conclude that remedial measures of control of flooding problems are
similar in all states of Niger Delta.
H5. States of the Niger Delta agree on their perception of local Attempt at flood
predictions in the Niger Delta (Table 4.52).
Using Kruskal-Wallis test-function we have the sum of joint Ranking for each State as:
R1 ¼ 7 þ 4 þ 9 þ 13 ¼ 33
R2 ¼ 5 þ 3 þ 11 þ 15 ¼ 34
R3 ¼ 6 þ 2 þ 10 þ 16 ¼ 34
R4 ¼ 8 þ 1 þ 14 þ 12 ¼ 35
and substituting in equation (3.11), we have:
12ð1; 378:5Þ
W¼ 2 3ð16 þ 1Þ ¼ 60:82 2 51 ¼ 9:82
16ð16 þ 1Þ
At a ¼ 0.05, K 2 1 ¼ 4 2 1 ¼ 3, we have from the X 2a table of probability (Appendix)
that X 2aðK21Þ ¼ 7:81
Inference. We rejects H5 since W ¼ 9.82 is . X 2a ¼ 7:81 and conclude that states of
the Niger Delta differ on their perception of local attempts at flood predictions in the
Niger Delta.
MEQ Results and discussion
17,3 The data analyses showed the following deductions:
Human perception of flooding in the Niger Delta especially with regards to causes,
frequencies and duration is the same among the Niger Delta communities. They all
argued that the causes range from heavy rainfalls, absence of drainages, blocked,
unmaintained and insufficient drainages to location of community in relation to river
286 channels. Also that flooding is an annual event and that when flooding occurs it lasts
for about 4-5 months. Their consideration of flooding lasting 4-5 months may be
related to the fact that right from the onset of the rains say (June) to the month of
October it is considered as flooding period.
Human perception of flooding in the Niger Delta differ among Niger Delta states or
committees in the aspect of impacts. The differences in terms of impacts differ in terms
of upland and coastal communities. While most of the upland communities such as
Downloaded by University College London At 03:37 28 June 2016 (PT)

Onitsha and upland areas of Port Harcourt, considered the impacts to be more in
terms of causing of blockages on our roads and causing formation of gullies and slides.
The coastal communities especially in Bayelsa considered the impact more on the loose
or destruction of farmlands and consequently premature harvesting of farm products
followed flooding of most homes.
The adjustment pattern to flooding are similar in all the states of the Niger Deltas.
In all the states, the main adjustment pattern is to temporarily clean up the house when
ever the flooding resides and evacuation of properties on higher places such as
temporarily improvised alters.
Local measures of control of flood problems are similar among the states of the
Niger Delta. Most of the rural communities in the coastal areas of the Niger Delta are
lost to what local measures to apply. They most often each year allow themselves to
faith; others in these areas attempt and come together and provide or construct
temporary sand banks. While in the upland locations and urban cities, they resort to
opening up of blocked drainages and construction of cheap and temporary drainages.
States of the Niger Delta differ in their perception of local attempts at flood
prediction in the Niger Delta. While some argue for warning signs such as thundering
or the likelihood of heavy downpour, others through looking at the weather while the
other category depended on guess work.
The states in the Niger Delta agree on the perception of the extent of damage by
flooding when it occurs. They all agree that the damage is great, enormous but differ
on the extent of such damage and in most cases are unable to determine the financial
implications of such destructions and damages caused by flooding.

Conclusion
The study has highlighted the importance of statistical technique analyses through the
Kruskal-Wallis approach in attempting to predict flood phenomenon in the Niger Delta,
Nigeria. Also that for cheap and local attempts at flood control and predictive measures
to be effective and successful, in the rural Areas, programmes on flood control
measures should be mounted to educate rural dwellers to avoid dumping of refuse in
the creeks, streams and rivers. This is most cases, causes the problem of river channel
improvement, which, makes navigation nearly impossible and river transportation
very difficult.
Enlightenment programmes should be mounted or introduced in flood prone local The application
communities on cheaper and local methods of effective flood control techniques such as
use of sand bars, opening up of creeks to divert course of flow and encourage
of Kruskal-
community participation in flood control effects. Wallis technique
References
Ebdan, D. (1985), Statistics in Geography, 2nd ed., Billing and Sons Ltd, Worcester, p. 232. 287
Ekpete, B.O. (1990), “Simulation of flow in the Bonny and New Calabar River System”,
unpublished MPhil thesis, submitted to Post Graduate School Rivers State University of
Science and Technology, Port Harcourt.
Faniran, A. (1972), River Basins as Planning Units for Nigeria, Ibadan University Press, Ibadan.
FEPA/World Bank (1998), Report on Flooding Study for the Niger Delta, Federal Government
Protection Agency, Mosunmolu Nig., Lagos, p. 253.
Downloaded by University College London At 03:37 28 June 2016 (PT)

Gobo, A.E. (1988), “Relationship between rainfall trends and flooding in the Niger-Benue Basins”,
The J. of Meteorology, Vol. 13 No. 132, pp. 318-24.
Gobo, A.E. (1990), “Rainfall data analysis as an aid for designs for maximum drainage and flood
control works in Port Harcourt”, J. of Discovery and Innovations African Academy of
Science, Vol. 2 No. 4, pp. 25-31.
Igbozurike, U.M. (1993), Soil Erosion Prevention and Control Manual, NEST, Ibadan, pp. 1-12.
NDES (1997), Niger Delta Environmental Survey: Phase I NDES Vol. 1. Report, Environmental
and Socio-economic Characteristics, Environmental Resources Managers, Lagos, pp. 1-4.
NLNG (1997), Gas Transmission System Environmental Inventory Report Vol. 2, Nigeria LNG,
Bonny, pp. 8-12.
Ogam, F.N. (2000), Fundamentals of Statistics and Probability, Springfield Publications Ltd,
Owerri, p. 210.
Pearson, E. and Hartley, H.O. (1966), “Biometric tables for statistians”, in Douglas, C. (Ed.),
Montomery Design and Analyses of Experiments, 3rd ed.,Vol. I, Cambridge University
Press/Wiley, Cambridge/London, p. 418.
Teme, S.C. (1990), “On cheap and local methods of flood control measures in the Niger Delta”,
IFERT, R/S UST. PH., pp. 18-20.
Teme, S.C. (2002), “The physical environment, engineering, hydrology and water resources in
Rivers State Ministry of Environment and Natural Resources, (RSMENR)”, Interim
Guidelines and Standards, pp. 1-23.
Sheeham, L. and Hewit, K. (1969), “A pilot survey of global natural disasters of the past twenty
years”, Natural Hazard Research Working Paper No. 11, Dept. of Geography, University of
Toronto, Toronto, pp. 258-60.
Ward, R.C. (1978), Floods – A Geographical Perspective, Macmillan Press, London, p. 326.

Further reading
Subramanya, K. (1991), Engineering Hydrology, Tata McGraw-Hill Publications Ltd, New Delhi,
p. 382.
MEQ Appendix
17,3
a
n 0.995 0.990 0.975 0.950 0.500 0.050 0.025 0.010 0.005

1 0.00 þ 0.00 þ 0.00 þ 0.00 þ 0.45 3.84 5.02 6.63 7.88


2 0.01 0.02 0.05 0.10 1.39 5.99 7.38 9.21 10.60
288 3 0.07 0.11 0.22 0.35 2.37 7.81 9.35 11.34 12.84
4 0.21 0.30 0.48 0.71 3.36 9.49 11.14 13.28 14.86
5 0.41 0.55 0.83 1.15 4.35 11.07 12.38 15.09 16.75
6 0.68 0.87 1.24 1.64 5.35 12.59 14.45 16.81 18.55
7 0.99 1.24 1.69 2.17 6.35 14.07 16.01 18.48 20.28
8 1.34 1.65 2.18 2.73 7.34 15.51 17.53 20.09 21.96
9 1.73 2.09 2.70 3.33 8.34 16.92 19.02 21.67 23.59
10 2.16 2.56 3.25 3.94 9.34 18.31 20.48 23.21 25.19
Downloaded by University College London At 03:37 28 June 2016 (PT)

11 2.60 3.05 3.82 4.57 10.34 19.68 21.92 24.72 26.76


12 3.07 3.57 4.40 5.23 11.34 21.03 23.34 26.22 28.30
13 3.57 4.11 5.01 5.89 12.34 22.36 24.74 27.69 29.82
14 4.07 4.66 5.63 6.57 13.34 23.68 26.12 29.14 31.32
15 4.60 5.23 6.27 7.26 14.34 25.00 27.49 30.58 32.80
16 5.14 5.81 6.91 7.96 15.34 26.30 28.85 32.00 34.27
17 5.70 6.41 7.56 8.67 16.34 27.59 30.19 33.41 35.72
18 6.26 7.01 8.23 9.39 17.35 28.87 31.53 34.81 37.16
19 6.84 7.63 8.91 10.12 18.34 30.14 32.85 36.19 38.58
20 7.43 8.26 9.59 10.85 19.34 31.41 34.17 37.57 40.00
25 10.52 11.52 13.12 14.61 24.34 37.65 40.65 44.31 46.93
30 13.79 14.95 16.79 18.49 29.34 43.77 46.98 50.89 53.67
40 20.71 22.16 24.43 26.51 39.34 55.76 59.34 63.69 66.77
50 27.99 29.71 32.36 34.76 49.33 67.50 71.42 76.15 79.49
60 35.53 37.48 40.48 43.19 59.33 79.08 83.30 88.38 91.95
70 43.28 45.44 48.76 51.74 69.33 90.53 95.02 100.42 104.22
80 54.17 53.54 57.15 60.39 79.33 101.88 106.63 112.33 116.32
90 59.20 61.75 65.65 69.13 89.33 113.14 118.14 124.12 128.30
100 67.33 70.06 74.22 77.93 99.33 124.34 129.56 135.81 140.17
Table AI.
III Percentage points of Notes: n – degrees of freedom
the x 2 distributiona Source: Adapted with permission from Pearson and Hartley (1966)

To purchase reprints of this article please e-mail: reprints@emeraldinsight.com


Or visit our web site for further details: www.emeraldinsight.com/reprints
This article has been cited by:

1. Mohamed Abdallah Gad DARWISH. 2013. Geochemistry of the High Dam Lake sediments, south
Egypt: implications for environmental significance. International Journal of Sediment Research 28:4,
544-559. [CrossRef]
2. Nelson Oly Ndubisi. 2011. Factorial and discriminant analyses of environmental sensitivity and initiative
of Nigerian firms. Thunderbird International Business Review 53:1, 19-36. [CrossRef]
3. P.N. Ranasinghe, G.W.A.R Fernando, C.B. Dissanayake, M.S. Rupasinghe, D.L. Witter. 2009.
Statistical evaluation of stream sediment geochemistry in interpreting the river catchment of high-grade
metamorphic terrains. Journal of Geochemical Exploration 103:2-3, 97-114. [CrossRef]
4. Uzoechi NwagbaraStakeholder Interaction for Sustainability: 185-198. [CrossRef]
Downloaded by University College London At 03:37 28 June 2016 (PT)

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy