Elsevier Editorial System(tm) for Earth Science Reviews Manuscript Draft Manuscript Number: EARTH1381 Title: Subaerial salt extrusions in Iran as analogues of ice sheets, streams and glaciers Article Type: Review Article Keywords: glaciers; salt sheets; namakiers; structures and fabrics; material properties. Corresponding Author: Dr. Christopher Talbot, PhD Corresponding Author's Institution: Uppsala University First Author: Christopher J Talbot, PhD Order of Authors: Christopher J Talbot, PhD; Veijo Pohjola , PhD Abstract: Ice (H20) and salt (halite, NaCl) share many physical properties and resemble each other in hand specimens and subaerial gravity-driven flows. However, while most significant bodies of ice accumulate in cold highlands and gravity spread where and soon after they form, most significant bodies of salt accumulate in tropical marine basins and have to be buried by >1 km of other rocks before they flow. Buried salt is driven by differential loading into various categories of piercing structures known as diapirs. Many diapirs extrude onto the surface as sheets of allochthonous (out of place) salt. Thousands of sheets of allochthonous salt have been interpreted in over 35 basins worldwide in the last 25 years, mainly in the toes of passive continental margins and in orogenic belts where some are >10E3 km2 in area. Most former salt sheets are now submarine or subsurface and have to be studied by seismic profiling ± drilling but several active examples are beautifully exposed in Iran. These were compared to ice glaciers soon after they were introduced to western science, a comparison that has been neglected since. Here we update this analogy and use modern understanding of flowing ice and salt to examine the similarities and differences that might be mutually beneficial to both fields of study as well as extraterrestrial scientists. The profiles, internal structures and fabrics in flowing bodies of ice and salt are sensitive gauges of the histories of their budgets of supply and loss. However, whereas ice caps merely compact where snow accumulates, salt sheets are fed by already deformed salt from below. When salt diapirs first emerge on land they extrude domes that mature to the profiles of viscous fountains that often feed glacier-like flows known as namakiers. After locally exhausting their deep source layers, salt fountains spread to the profiles of viscous droplets normal for ice caps. Ice typically deforms at >80% (usually > 90%) of its absolute melting temperature while most salt deforms at <50% of its homologous temperature; as a result, grain shape fabrics in salt are clearer and have longer strain memories than in ice. Foliations in deformed salt map streamlines that aid understanding of how internal folds develop. Salt sheets seldom erode their channels like flowing ice and internal debris accumulates on their tops rather than their bases. Ice sheets float on water but, as salt sheets are twice as dense as ice, rain that falls onto the top surface of namakiers tends to stay there. Both glaciers and namkiers surge but the association between surges and changes in boundary conditions are much clearer for namakiers than glaciers. Because the rate of delivery of land ice to the oceans is such an important control on sea level, we end by considering how the implications of surging salt might help studies of surging ice and converge on recent glaciological findings that pont to changes in boundary conditions other than their bases. Suggested Reviewers: Ian Alsop gia@st-andrews.ac.uk Expert in salt tectonics Wendy Lawson wendy.lawson@canterbury.ac.nz Expert structural glaciologist Garry KC Clarke clarke@eos.ubc.ca International expert in glaciology JK Warren jwarren@brunet.bn International expert in Salt Author Agreement Cover letter, Dear Editors, Subaerial salt extrusions in Iran as analogues of ice sheets, streams and glaciers On behalf of both authors, I enclose an entirely re-written version of a script that was submitted as EARTH1192 with a slightly different name over a year ago. We were asked by our reviewers and editor of that script for a major reworking in what was a reasonable time were it not for the fact that I retired and moved my household from Sweden to England soon after. I apologise for the delays beyond the original and subsequently-negotiated deadlines and fully understand that you wish to treat this as a new script. However, we wished to produce a paper we would be proud of rather than to rush to meet your schedule. This script is much longer than the original version because every one of the reviewers’ and editor’s comments on the previous version have been taken into account. We have taken to heart their encouragement to add more coverage on ice and have added two significant sections, one on micro-organisms in ice and salt, the other summarising the hot topic of surging ice (up to September 2008) and relating it to surging salt. We now have a clearly stated aim and theme. We have also added to and improved out Figures and cited primary references (as well as a few text books). I have had great difficulty uploading the Figure files to your web site and submit all the figures and plates in black and white jpegs to save space. If the script is accepted we would like the figures on the web to be in colour. Sincerely, Christopher Talbot * Manuscript Click here to download Manuscript: Talbot-ice-final.doc Click here to view linked References 1 1 Subaerial salt extrusions in Iran 2 as analogues of ice sheets, streams and glaciers 3 Christopher J. Talbot*, Veijo Pohjola 4 Department of Earth Sciences, Uppsala University, Villavägen 16, 752 36 UPPSALA, Sweden. 5 Received February 2007; returned November 07 Resubmitted 15 September 2008 Accepted 6 7 Abstract: Ice (H20) and salt (halite, NaCl) share many physical properties and resemble each 8 other in hand specimens and subaerial gravity-driven flows. However, while most significant bodies 9 of ice accumulate in cold highlands and gravity spread where and soon after they form, most 10 significant bodies of salt accumulate in tropical marine basins and have to be buried by >1 km of 11 other rocks before they flow. Buried salt is driven by differential loading into various categories of 12 piercing structures known as diapirs. Many diapirs extrude onto the surface as sheets of 13 allochthonous (out of place) salt. Thousands of sheets of allochthonous salt have been interpreted 14 in over 35 basins worldwide in the last 25 years, mainly in the toes of passive continental margins 15 and in orogenic belts where some are >103 km2 in area. Most former salt sheets are now submarine 16 or subsurface and have to be studied by seismic profiling ± drilling but several active examples are 17 beautifully exposed in Iran. These were compared to ice glaciers soon after they were introduced to 18 western science, a comparison that has been neglected since. Here we update this analogy and 19 use modern understanding of flowing ice and salt to examine the similarities and differences that 20 might be mutually beneficial to both fields of study as well as extraterrestrial scientists. 21 The profiles, internal structures and fabrics in flowing bodies of ice and salt are sensitive 22 gauges of the histories of their budgets of supply and loss. However, whereas ice caps merely 23 compact where snow accumulates, salt sheets are fed by already deformed salt from below. When 24 salt diapirs first emerge on land they extrude domes that mature to the profiles of viscous fountains 25 that often feed glacier-like flows known as namakiers. After locally exhausting their deep source 26 layers, salt fountains spread to the profiles of viscous droplets normal for ice caps. 27 Ice typically deforms at >80% (usually > 90%) of its absolute melting temperature while most 28 salt deforms at <50% of its homologous temperature; as a result, grain shape fabrics in salt are 29 clearer and have longer strain memories than in ice. Foliations in deformed salt map streamlines 30 that aid understanding of how internal folds develop. Salt sheets seldom erode their channels like 2 31 flowing ice and internal debris accumulates on their tops rather than their bases. Ice sheets float on 32 water but, as salt sheets are twice as dense as ice, rain that falls onto the top surface of namakiers 33 tends to stay there. Both glaciers and namkiers surge but the association between surges and 34 changes in boundary conditions are much clearer for namakiers than glaciers. Because the rate of 35 delivery of land ice to the oceans is such an important control on sea level, we end by considering 36 how the implications of surging salt might help studies of surging ice and converge on recent 37 glaciological findings that pont to changes in boundary conditions other than their bases. (485 38 words) 39 40 Key words: Ice sheets, glaciers, salt sheets, namakiers, structures and fabrics, material properties. 41 * Corresponding author: Now at 14 Dinglederry, Olney, Bucks, 46MK 5ES, UK (Tel: +44 1234 714 42 43 44 45 46 47 48 140) E-mail addresses: Christopher.Talbot@geo.uu.se and Veijo.Pohjola@geo.uu.se 1. Introduction Water ice and salt (consisting mainly of halite, NaCl) can resemble each other in both hand 49 specimen and in surface bodies (Figs 1 & 2) despite forming by entirely different processes in 50 different and mutually exclusive environments. Most significant bodies of ice form at high latitudes 51 near to maritime air masses and soon gravity spread from their high inland accumulation areas as 52 lobes and outlet glaciers to more continental and drier areas. By contrast, most significant bodies of 53 salt crystallise by the evaporation of seawater in marine basins between latitudes 10-40º N & S. 54 Autochthonous (in place) salt has to be buried to depths of km and squeezed back to the surface in 55 intrusive bodies known as diapirs before some extrude* sheets of allochthonous (out of place) salt 56 that gravity spreads over the surface in smaller analogues of ice caps and glaciers (Fig. 3). Large 57 ------------------------------------------------------------------------------------------------------------------------- 58 *Only footnote: Except where signalled in section 5.6, we use extrusion here in the geological 59 sense of an elasto-plastic solid being driven upward through vents in other rocks, not in the 60 glaciological sense (Demorest, 1942; Gudmundsson, 1997a) of an increase in horizontal flow 61 velocity with depth. ----- 3 62 bodies of both ice and salt on the surface are temporary because they are gravitationally unstable 63 and vulnerable to the climate. Significant bodies of ice seldom survive for more than a million years 64 and, while extruded sheets of allochthonous salt are also short-lived, the salt in them can be 5-500 65 million years old. 66 Here we update early comparisons between glaciers of ice and salt (e.g. Harrison, 1930) and 67 explore how the comparisons and contrasts between flows of ice and salt might be of mutual 68 benefit to studies of both materials on Earth and analogue materials on extra-terrestrial bodies. 69 Bodies of both water ice and salt are being recognised or suspected on increasing numbers of 70 extraterrestrial bodies where the different environments may lead to ice flowing rather like salt on 71 Earth (Schenk and Jackson, 1993; Pappalardo and Barr, 2004, and Schenk and Pappalardo, 2004) 72 or methane flowing like water on Titan (Lunine and Atreya, 2008). To understand the processes 73 that give rise to extraterrestrial landforms, it is becoming increasingly important to study their 74 analogues on Earth. It is also advantageous to be able to distinguish by remote means such 75 phenomena as snow fields from salt playas, cryo-volcanoes from salt diapirs, ice sheets from salt 76 sheets and glaciers from namakiers. However, here we will focus mainly on Earth-bound studies as 77 structural glaciologist and geologists. 78 Early hominids probably used ice as “land bridges” to spread through high latitudes and must 79 have learned to cope with the harsh climates of the ice age. They probably also learned that salt is 80 a necessary for their health and have probably “mined” it for many millennia. The first known 81 solution mining was in Poland at least ~3000 years ago and salt mining began in the middle east 82 ~2000 years before science began to take an interest in salt (for a historical review of salt tectonics, 83 deformation involving salt flow, see Jackson, 1995). 84 Glaciology began slowly when glaciers invaded valleys farmed in the Alps during the little ice 85 age in the 17th century (Walker and Waddington, 1988). However, the subject really took off after 86 1837 when Louis Agassiz, accepting earlier suggestions that the huge erratic boulders of granite 87 high on limestones in the Jura Mountains were emplaced by ice rather than the “Deluge”, went on 88 to suggest that Earth had been subject to a past ice age. 89 Svein Pálsson, an Icelander, first described in his diary at the end of the 18th century how 90 glaciers flowed under gravity like viscous tar, a theory proved in 1846 by James Davis Forbes 91 using experiments on the Mer de Glace in the French Alps (Walker and Waddington, 1988). We 4 92 guess that Pálsson had in mind a recent Icelandic lava eruption as another analogue of viscous 93 flow in rocks and ice, an approach we follow in this review. 94 After a century of relatively slow progress in theoretical glaciology during the exploration of 95 polar geography, glaciology kicked off again after WWII, boosted by the Habbakuk project – one of 96 Churchill’s pet projects attempting to build aircraft carriers using ice and wood pulp (Perutz, 1948). 97 This project inspired British ice physicists to start post-war glaciological field projects to test their 98 theories about glacial dynamics (Clarke, 1987). In fact, exploration of the ice sheets of Antarctica 99 and Greenland had begun in the 1930s (Robin and Swithinbank, 1988) but blossomed as a result of 100 the large geophysical campaigns and >50 Antarctic stations installed for the International 101 Geophysical Year (IGY) of 1957-58 by the 12 original signatories of the Antarctic treaty that came 102 into force in 1961. 103 The literature that has accumulated on glaciology since the work of Agassiz in mid 19th century 104 is rather larger than that on salt extrusions since Ville (1856) first described a mountain salt of salt 105 extruding geyser-like in the Sahara Atlas of Algeria. This imbalance reflects snow and ice covering 106 >10% of the Earth’s surface whereas analogous extruded salt covers much less (but see maps in 107 Hudec and Jackson, 2006 for regions underlain by salt). 108 Shumskii (1964) is a definitive text on structural glaciology and Maltman et al., (2000) reviewed 109 modern developments. These authors summarised how structural glaciologists developed new 110 insights by applying structural geological concepts, first to the small individual valley glaciers and 111 ice caps that form over centuries, and then to the larger ice sheets that form over millennia 112 (Maltman et al, 2000; Hambrey and Lawson, 2000). Such studies accelerated in 1988-1998 after 113 the structures and deformation histories of surge-type glaciers and ice streams (fast flowing rivers 114 of ice that drain most of an ice sheet) were distinguished from those subject to slow penetrative 115 creeping flow. 116 Glaciology advanced further when short-wave surface radar identified new categories of folds 117 well above the beds of ice streams (Jacobel et al., 1993) and when high-resolution satellite images 118 became available (Merry and Whillans, 1993; Bindschadler, 1993, 1998; Unwin and Wingham, 119 1997). It surged again when ice cores found unexpected complications in near-basal ice strata 120 beneath the current divides of ice sheets (e.g. Souchez et al., 2000). Since then the recognition of 5 121 how the drainage of subglacial lakes relate to ice streams promises another significant advance 122 that will be considered later (section 9). 123 Hudec and Jackson (2007) summarized current ideas about the mechanics of salt flow, the 124 processes of diapir growth (Fig. 4), and the ways that these processes interact with regional 125 deformation to produce salt structures with a huge variety of shapes (Fig.5). They correct the past 126 emphasis on the buoyancy of structures of viscous salt in ductile surroundings to the current picture 127 where the location and shape of most salt bodies depend on how the stronger surrounding rocks 128 deformed. There has always been discussion about the relative importance of gravity and lateral 129 tectonic forces as drives for particular suites of salt structures, a tradition that we will continue here. 130 Hudec & Jackson (2006) reviewed the literature on the thousands of sheets of allochthonous salt 131 now known in orogens and passive continental margins and distinguished them into 4 different 132 categories. We will focus here on one of these, the subaerial salt extrusions in Iran that are 133 relatively accessible and provide the closest analogues of flowing ice. 134 Although past studies of ice have been funded by the hydro-electric power industry keen to 135 exploit melt water, today’s focus is on the concern about climate change and how the rate of 136 delivery of land ice to the oceans affects sea-level changes. By contrast, the data and funding for 137 studies of salt have come largely from industry. In the 1930s, hundreds of mines in the salt domes 138 of central Europe were extracting potash and sodium salts, gypsum, sulphur, borates, nitrates, and 139 zeolites as the feedstocks for the chemical and construction industries (Warren, 1999). In the 1970s 140 and early 80s, salt, was a target for nuclear waste isolation because excavating or dissolving 141 underground space in it is comparatively easy; by the late 1980s deformed salt was recognised as 142 too mobile for this use (the WIPP depository for military nuclear waste in New Mexico is in bedded 143 salt). From its beginning, the oil industry has always been interested in salt because salt affects 144 most aspects of hydrocarbon systems by creating structural traps, influencing reservoir distribution 145 and blocking fluid flow. Within 5 years of the first hydrocarbon blow-out in Texas in 1901, oil or gas 146 were being produced from sands in the flanks of ~50 salt diapirs. Something like 60% of 147 hydrocarbon fields are associated with salt structures (Halbouty, 1979). Many of the world's great 148 hydrocarbon provinces lie in such salt basins as the Gulf of Mexico (Nelson and Fairchild, 1989; 149 Worrall and Snelson, 1989, Wu et al., 1990, Diegel et al., 1995; Peel et al.,1995; Schuster, 1995, 150 Rowan et al. 1999), Persian Gulf (Kent, 1979; Talbot and Alavi, 1996; Sherkati and Letouzey, Eo, 6 151 Volgeo (ELS)2004), North Sea (Coward and Stewart ,1995; Kockel, 1998), Lower Congo Basin 152 (Rouby et al., 2002) the Campos Basin off Brazil (Cobbold et al., 1995) and the Pricaspian Basin 153 (e.g. Volozh et al., 2003). 154 Like water, salt is vital to most life forms. Just as a little salt flavours larger volumes of food, 155 a little salt exaggerates so many geological deformation processes (Talbot, 1992) that, as for ice, 156 salt structures can be treated as natural analogues models of many deformation processes in more 157 usual rocks Thus thin layers of salt can detach overlying rocks in regional extension and shortening 158 and simulate decoupling by shear along weak zones in and beneath the lithosphere. Salt often 159 plays a role in sedimentary basins similar to that of weak and buoyant granitoids in the crystalline 160 continental crust (Talbot, 1992). 161 This comparative review explores some of the analogies between ice and salt. As processes 162 involving salt are generally less well known than glacial processes, we will focus more on salt than 163 ice. Section 2 starts by contrasting how ice sheets and analogue salt bodies form. Section 3 then 164 explains how salt is buried and expelled back to the surface to form salt diapirs, some of which 165 extrude analogues of ice sheets. We then compare the shapes and rates of salt extrusions with ice 166 before discussing the dynamics of salt diapirs and extrusions. Hudec & Jackson (2006) recently 167 suggested that the emplacement of salt sheets require lateral compression but we will describe 168 evidence that questions this. Section 4 describes the layering within ice and salt and the debris they 169 generate, carry and deposit. Section 5 discusses their grain-scale shape fabrics and deformation 170 mechanisms before distinguishing their macroscopic internal structures into kinematic and dynamic. 171 We then outline kinematic folds in ice and salt before considering dynamic folds near the bottom 172 and top boundaries. Section 6 contrasts the fractured carapaces of ice and salt bodies before 173 section 7 dicusses microscopic their fluid inclusions and biological contents. Section 8 considers 174 the mechanical effects of water on ice and salt and leads to section 9 discussing current 175 understanding of surging flows in not only ice and salt but also of laboratory experiments on the 176 gravity spreading of viscous fluids. The conclusions highlight the areas where comparative studies 177 of bodies of flowing ice and salt might be mutually beneficial. 178 179 180 7 181 182 2. Formation and deformation of ice and salt It usually takes decades to centuries for snow sedimented on topographic highs in cold climates 183 to be buried to depths where firn compacts to crystalline ice that can gravity spread or glide down- 184 slope to levels where it ablates (Fig. 3A). By comparison, rock salt is a comparatively rare rock that 185 precipitates wherever seawater evaporates beyond 11 times its general salinity in restricted basins 186 between latitudes 10 & 40º N and S with or without additional input from volcanic brines (Fig. 3B) 187 (Warren, 1999; Hardie, 1991; Hovland et al., 2006). Unlike ice and most other rocks that compact 188 on burial, rock salt precipitates as a crystalline rock with a grain size of cm that typically loses any 189 initial porosity by 40 –70 m burial (often over millions of years) as halite precipitated from 190 subsurface brines infills the pores (Warren 1999, p. 22). By comparison, compaction and 191 crystallisation (e.g. to cm grains) at depths of 10-100 cm in a snow pack involve vapours rather than 192 solutions. 193 We will follow a classical approach and consider both glacier ice and rock salt as essentially 194 mono-minerallic metamorphic rocks that deform comparativelt rapidly at low stresses and low 195 temperatures and therefore model more common rocks deformed more slowly at much higher 196 stresses and temperatures. Table 1 lists some of the properties of water ice and halite. 197 The strengths of ice crystals are strongly anisotropic so that dislocation glide on the basal 198 plane is comparatively easy until hindered by kinks (Shumskii, 1964; Wilson and Marmo, 2000) 199 Compaction with depth increasingly rotates the initially random c-axes of snow through broad 200 clusters of c-axis about the vertical in firn that tightens over 103-4 years to a narrow girdle in ice at 201 an ice depth ~1.5 km (Wang et al., 2002). Compaction therefore progressively aligns the basal 202 shear planes in ice to the orientations where they are increasingly vulnerable to the bed-normal 203 layer thinning, along-bed extension and bed-parallel shear that increases toward the bed. Shear 204 along the margins of ice streams and glaciers imposes narrow girdles much faster (e.g. ~102 years). 205 Large strains are aided by diffusion, grain rotation and boundary translation processes (Alley et al., 206 1997; Wilson and Marmo, 2000) that may lead to dynamic recrystallisation. The crystallography of 207 halite, is very different from ice and its face-centred cubic symmetry with 3 orthogonal potential slip 208 planes oblique to the faces ensures that, in bulk, salt is close to mechanically isotropic. 209 210 The cycle from snow to glacier ice through gravity-driven compaction and flow to water by melting or ablation usually takes <<1 Ma on Earth today. The equivalent cycle, where brine 8 211 evaporation forms bodies of crystalline rock salt that are buried and squeezed back to the surface 212 where they gravity spread in flows that share visual and mechanical similarities to ice sheets, caps, 213 streams and glaciers usually lies somewhere between 5 and 500 Ma. 214 Significant salt sequences often begin accumulating rather slowly in cyclic beds of other 215 evaporites (carbonates and calcium sulphates) by evaporation of ephemeral brines in rift valleys 216 (Warren, 1999). The brines in rift valleys are likely to have high components of volcanic input and 217 Hovland et al., (2006a, b) have recently suggested that brines heated to supercriticality by volcanic 218 heat may deposit sizeable salt deposits. If the rift valleys open to narrow seaways and then oceans, 219 any initial cyclic sequence of subaerial evaporites can be buried by purer salt that can crystallise 220 much more rapidly (cm a-1) by epitaxial growth on pre-existing halitic substrates beneath permanent 221 brines (Talbot et al., 1996; Sage et al., 2005). Because salt is usually one of the first sediments to 222 form in a new basin it is usually buried by rocks of other origins (clastic sediments ± volcanic rocks). 223 Such basal salt beds may be >1 km thick over Mediterranean-like areas the but are still 224 volumetrically small compared to current and past ice sheets. 225 Old dirty ice with strong grain-shape fabrics are comparatively weak and can decouple cleaner, 226 younger and stronger overlying ice (Hooke and Hanson, 1986). Similarly, basal salt sequences are 227 so weak that their ductile flow can mechanically decouple the overlying cover rocks from their 228 crystalline basement during subsequent “thin skinned” tectonics Thus cover sequences tens of km 229 thick may undergo enormous lateral extension when they spread and/or slide off “passive” 230 continental margins (such as the Gulf of Mexico, Brazil, West Africa, etc: Duval et al., 1992; Hudec 231 and Jackson 2004) or enormous lateral shortening in foreland basins and thrust-fold belts beside 232 mountain chains where continents converge after intervening oceans have closed. About 120 of the 233 deformed basins with surviving salt tectonics are shown on maps by Hudec &Jackson (2007, figs. 234 5a & b). 235 236 237 3. Salt diapirs and extrusions Ice and salt can be considered some of the weakest rocks (Table 1) so that substantial bodies 238 of both deform as crystalline elasto-viscous fluids as they flow “down” load gradients induced 239 mainly by gravity. 9 240 The closest salt comes to emulating ice and flowing in the place it first forms is where clastic 241 sediments accumulate and sink into thick, recently formed, salt layers. Although less dense than 242 salt until compacted at depths >~1km, these clastic sediments sink into the weak underlying salt as 243 they accumulate and drive the salt along lateral pressure gradients into adjacent highs. These form 244 passive cylindrical or wall-like diapirs with crests that stay near the depositional surface in a 245 process known as downbuilding (Fig. 3A, Barton, 1933; Jackson et al., 1994; Hudec and Jackson 246 2006). 247 The Zechstein (250 Ma) salt sequence in Germany and the North Sea and is known to have 248 developed downbuilt diapirs and extruded stacked namakiers in Late Triassic (Mohr et al., 2007) 249 and Cretaceous times (Zirngast, 1996). Similarly, the Jurassic Louanne salt sequence began 250 moving at shallow depths in the Gulf coast interior basins (Rosenkrans and Marr, 1967) and was 251 buried to as deep as14 km by younger clastic sediments that built out into the Gulf of Mexico and 252 drove most of the salt basinward in several generations of diapirs that spawned the hundreds of 253 submarine salt sheets now extruding on its floor (Wu et al., 1990, Gemmer et al., 2004). 254 Other salt sequences have been buried by overburdens of constant thickness faster than the 255 salt could rise by its (rather weak) inherent buoyancy. However, some diapirs actively upbuild 256 through overburdens that are only slightly denser but ductile (Fig. 4B, e.g. Jackson et al., 1990). 257 More usually, differential loading of a salt source layer by the mass of the overburden drives 258 reactive salt diapirs upward where faults have thinned and weakened stronger brittle overburdens 259 (Fig. 4C and Jackson et al., 1994, Hudec and Jackson, 2006a). 260 The crests of passive salt diapirs remain near the surface as their source layer deepens- and 261 the crests of many active or reactive diapirs rise to the surface (at 0.1 to 3 mm a-1: Jackson and 262 Talbot, 1986) from their deep source layer(s). What happens then depends on the surface 263 environment into which the salt extrudes. High rainfalls onshore can dissolve salt as fast as it 264 extrudes so that the insoluble residues in the extruding sequence accumulate atop the salt as 265 downward thickening cap rocks. These develop as hard cemented caps in reducing conditions 266 below the water table in the US gulf coast (e.g. Seni, 1987, Posey et al., 1987, Warren, 1997), but 267 as loose soils on salt emerging in oxidizing conditions in Iran (Talbot, 1998; Bruthans et al., 2007). 268 Equivalent diapirs rising into seawater probably dissolve more slowly in less aggressive seawater 269 and can extrude large volumes of salt that flow downslope with or without the burial that can lead to 10 270 further cycles of intrusion and extrusion (Jackson et al., 1994). The first geologists to encounter 271 such subaerial salt extrusions in the south of Iran referred to them as salt glaciers (Lees 1927, Busk 272 1929, DeBockh et al., 1929, Harrison, 1930). However, the term salt glacier involves a bilingual 273 redundancy, and because “river of salt” and “saltier” have other meanings, we use here another 274 bilingual term for sheets of allochthonous salt extruding over the former surface from salt diapirs: 275 namakier (from namak, Farsi for salt; Talbot and Jarvis, 1984). 276 Ice analogues of emergent salt diapirs are pingoes, where overpressured groundwater rises 277 through ground icing on fine-grained flats in permafrost regions and grow large “frost-blisters”. 278 These bulge the surface on scales of hundreds of metres in terrestrial structures and are 279 sometimes referred to as cryo-volcanoes (Washburn, 1979). Larger cryovolcanos exist on the icy 280 moons around the outer planets (Lucchitta, 2001; Pearle, 2003). 281 Salt diapirs are more common and usually larger than pingoes on Earth, being typically a few km 282 across in their shallow levels and 3-12 km from crest to root on wavelengths between 6 and 26 km. 283 However, the subaerial namakiers that simulate bodies of comparatively young glaciers are smaller, 284 fewer and usually extrude long after the salt in them formed. Such disparities in the ages of ice and 285 salt bodies indicate that ice on the surface is more vulnerable to degradation than buried salt. Being 286 denser than water, most salt extrudes over the seafloor offshore. The focus here is on subaerial 287 extrusions of salt over rocks and soils onshore that evolve through similar shapes in various arid 288 regions of Iran despite consisting of salts of two very different ages. 289 290 291 3.1 Shapes of salt extrusions Although all viscous and strain-rate softening fluids (like ice and salt) flow downhill beneath 292 their top free boundaries, all valley glaciers, most outlet glaciers and several of the known ice 293 streams also generally flow downslope over their bottom boundaries (i.e. they are steered by the 294 topographies of their substrates). 295 Subaerial salt extrusions are also shaped by the rate at which distal salt advances over a 296 particular substrate from a vent of a particular shape. Thus axisymmetric extrusions spreading by 297 pure shear over plains slow as they expand circular sheets of allochthonous salt that can be ~ 100 298 m thick with surface velocities calculated at between 4.10-4 and 20 m a-1 (Wenkert, 1979). Faster 299 and thinner (e.g. 5 x 2 x 0.3 km) monoclinic namakiers spreading by simple shear down the dip 11 -1 300 slopes of anticlines in the surrounding limestones (at velocities measured at up to 17.6 m a : Talbot 301 et al., 2000) do not widen downslope, even where they advance over a plain (Fig. 1SB). Current 302 subaerial namakiers in Iran do not coalesce (Fig. 1SB). However, the bulbs of some groups of 303 diapirs have merged to form “salt canopies”, the nearest saline equivalents of coalesced piedmont 304 glaciers. Twelve diapirs contribute to an onshore salt canopy (38 x13 km) in central Iran (e.g. 305 Jackson et al., 1990), perhaps 30 form an incomplete 140 x 40 km canopy off the shore of Yemen 306 (Heaton et al., 1995) and hundreds have merged beneath the Gulf of Mexico (Hudec and Jackson, 307 2006). Most offshore salt extrusions are reactivated into later, smaller diapirs after being buried by 308 younger clastic sediments. 309 Although still-flowing salt is known to bypass now static salt along steep strike-slip faults 310 (Talbot, 1979), few individual salt sheets are sufficiently well known to recognise possible “salt 311 streams” (but see fig 21 in Hudec & Jackson, 2007). The closest equivalents to ice streams 312 currently known in salt sheets are where salt flows faster between partially sunken roof blocks (e.g. 313 fig. 6 in Hudec & Jackson 2006). 2 314 The largest source-fed thrust sheets of allochthonous salt are > 1000 km in area (Hudec & 315 Jackson, 2006) but the subaerial salt extrusions in Iran are much smaller (~ 25 km2) and flow more 316 slowly than most ice streams and glaciers. 317 Domes and namakiers of extruding salt emulate many more than visual aspects of ice sheets 318 and glaciers (Fig. 1 & 2) – with intriguing differences. The mass budgets of many active ice bodies 319 depend mainly on climatic and topographic influences on the supply of snow and the loss of ice. By 320 comparison, the mass budgets of active salt extrusions in Iran can be more complicated (and more 321 like those of ice streams) because their supply depends on the rates that already crystalline 322 material is fed into them (at mm to m a-1) from small vents in their country rocks (Fig. 6) and the 323 loss of extruded salt depends on the rate of its gravity spreading as well as and local river erosion 324 and general surface dissolution (Talbot and Jarvis, 1984) in the local rainfall (e.g. <1500 mm a in 325 Iran: Bruthans et al., 2007). 326 -1 The salt extrusions in Iran are studied as accesiblle versions of submarine salt extrusions that 327 can trap hydrocarbons, and as smaller and faster natural models of steady-state mountains of other 328 metamorphic rocks extruding from near the suture zones where continents converge (e.g. Talbot 329 and Jackson 1987; Talbot, 1998) Piles of nappes of salt and other metamorphic rocks can have 12 330 steady-state topographies and fluxes of mass and heat if they degrade by gravity spreading and 331 climatic processes as fast as they extrude from sub-horizontal channels in the crust (e.g. Pazzaglia 332 and Knueper, 2001). Thus the Himalaya extrude silicate gneisses at mm a on length scales of 10 333 km for 107 years (Beaumont et al., 2001) but the salt mountains of Iran are easier to study as they 334 extrude at 0.1-1 m a on scales of 10 m for 10 years. By comparison, ice characteristically flows 335 1m per 0.1-0.01 yr for <<106 years and a typical cycle from snow to melt is six orders of magnitude 336 faster than the Himalaya orogen (Hambrey and Lawson, 2000). -1 -1 3 3 5 337 Extrusions of different salt sequences in different regions of Iran appear to evolve through 338 similar geometries although on different time and length scales (Fig. 6). Thus perhaps a hundred 339 diapirs of Oligocene to Miocene salt (20-5 Ma) emerge on the central plateau of Iran and something 340 like 150 diapirs of Hormoz (Neoproterozoic to Cambrian in age, ~500 Ma) salt extrude in the Zagros 341 Mountains and their foreland basin (Kent, 1970). Most such extrusions spread from rhombic vents a 342 few km across where salt has been expelled from depths of 3-12 km up the stems of reactive 343 diapirs pulled apart at releasing bends along strike-slip faults (Fig. 7 and fig. 29 in Hudec & 344 Jackson, 2007). Whatever its age, the salt first extrudes a rounded topographic dome that rises a 345 few hundred metres above its vent, probably in millenia, whether this is on a plain near sea level or 346 atop an anticlinal ridge >1 km high (like diapir 1 in Fig. 1SA). These domes of salt cannot support 347 their own weight in the surrounding air and so their lower slopes gravity spread short wide aprons 348 and/or narrow namakiers up to 8 km over the surrounding scenery (diapir 2 in Fig. 1SA and 7A). 349 While supplied by salt expelled from depth, extrusions of salt develop the distinctive smooth profiles 350 of viscous fountains (Lister and Kerr, 1989) that contrast with the angularity of the surrounding 351 mountains of other rock types (Figs. 6 & 7). 352 A high ratio of rate of extrusion to rate of spreading raises a summit dome well above the apron 353 spreading from the flanks of a salt fountain. After something like 60,000 years (Talbot et al., 2000, 354 fig. 7), a salt fountain locally exhausts its deep source layer, its ratio of extrusion- to spreading-rates 355 falls, and the summit dome subsides into the salt apron. No longer fed locally from below, the 356 former viscous fountain then gravity spreads to the profile of a viscous droplet (dashed in Fig. 6B 357 after Huppert, 1982) that characterise most ice caps with their comparatively widespread supplies 358 of snow. Unless salt extrusion is re-invigorated by lateral forces or further burial, salt erosion 359 eventually outpaces extrusion and the droplet profile degrades to conical piles of residual cap soils 13 360 growing atop the dissolving salt as insoluble components in the extruded sequence accumulate. 361 Eventually only a hollow of residual soils above a chimney of mega-breccia marks a former salt 362 extrusion (Fig. 6B). 363 Annual surveys of markers on ten mountains of Hormoz salt in the Zagros Mountains in the 364 1990s found that five were extruding faster that they degraded while five were wasting (Talbot et al, 365 2000, fig.1). Modelling such results indicate that the extruding salt rises out of its vent at over 1 m 366 a with a surface bulge due to budget irregularities travelling faster downslope than the salt (Talbot 367 et al, 2000). Fountains fed by 0lgocene-Miocene salt (~5-20 Ma) 3 km deep in central Iran are 368 smaller than those in the Zagros and an example reaches only 315 m above the surrounding plain 369 because it has been extruding more slowly e.g. 82 mm a-1 in an average annual rainfall near 23 mm 370 a for the last 42,000 years (Talbot and Aftabi, 2004). -1 -1 371 The stratigraphy of a flowing body of ice only accumulates strain as it gravity spreads from 372 beneath a visible upslope accumulation zone to its ablation zone downslope. By contrast, the salt 373 equivalent of the glaciologists’ accumulation zone is out of sight in the extrusive vent (Fig. 1). The 374 salt surfacing on the crest of a salt extrusion is already strongly deformed having flowed many km 375 along the deep source layer and then up the diapir. It has also undergone strong lateral extension 376 where it passed through a flow separation zone just beneath the crest of a dome (starred on 377 profiles on Fig. 7). Once exposed on the surface, the salt can be dissolved by rain falling on any 378 part of it. The resulting surface relief in flowing salt allows study of the initiation and development of 379 internal deformation structures and fabrics more easily than in flowing ice where relief tends to be 380 lower, particularly in accumulation zones. 381 382 383 3.2 Dynamics of salt diapirs and extrusion Hudec & Jackson (2007) detailed the forces driving and hindering salt flow. Differential loading 384 provides the dominant driving stress which is opposed by the strength of the overlying rocks and 385 friction along the boundaries of the sal body. Salt flows as soon as the driving forces overcome the 386 resisting forces, otherwise, it can remain static for hundreds of millions of years. Three types of 387 loading can drive salt flow: gravitational loading, displacement loading, and thermal loading. 388 Because salt behaves as a fluid over geologic time scales, Hudec & Jackson (2007) simplified the 389 effects of gravitational loading by using the fluid statics concept of hydraulic head (Kehle, 1988) and 14 390 distinguished hydraulic head gradients into two components: elevation head and pressure head. An 391 elevation head gradient is between two particles of fluid at different elevations above an arbitrary 392 horizontal datum. A pressure head is the height of a fluid column that could be supported by the 393 pressure exerted by the overlying rock. A pressure head gradient is induced by lateral variations in 394 the load on salt layers induced by erosion, sedimentation or deformation. The geometry of the 395 bottom boundary of the fluid layer does not influence the head gradient although it may influence 396 the geometry of flow once it begins. 397 Hudec and Jackson (2007 p.29) defined displacement loading as the “results from the forced 398 displacement of one boundary of a rock body relative to another” (e.g., Suppe, 1985). Although they 399 went on to state that “in salt tectonics, this type of loading occurs when the flanks of a salt body 400 move toward or away from one another during regional shortening or extension” we find the first 401 sentence obscure as to whether the boundaries of the salt body are moved by gravity or lateral 402 tectonic forces. We will therefore refer to lateral tectonic forces rather than displacement loading. 403 Subhorizontal shears are known to be very important along gently dipping salt layers in thin- and 404 thick-skinned tectonics but the flanks of salt allochthons have such small areas that the normal 405 stresses induced by lateral tectonic forces may only become significant when applied to salt 406 structures that already have significant relief. 407 Theoretically, the thermal expansivity of salt is sufficiently higher than its compressibility that -1 408 geothermal gradients quite normal in sedimentary basins (e.g. 30ºC km ) could drive thermal 409 convection within salt layers with a viscosity of 1016 Pa s and thickness >2.6 km (Jackson et al., 410 1990, fig 3.8). Such internal circulation could be either immediately beneath the surface (e.g. in 411 Afar: Talbot, 1978, 2008) and offer an accessible analogue of thermal convection in the mantle, or 412 beneath insulating layers of other rocks; however, no such movements have yet been 413 demonstrated. The vortex coils in the mushroom shaped diapirs of Iran are difficult to explain 414 without invoking a component of thermal convection (Jackson et al., 1990) but thermal loading 415 probably usually merely plays an additional subsidiary role in most other salt movements. 416 Ice is too compressible to convect on Earth. However, the salt mushrooms in central Iran were 417 quoted as analogues of the cellular pattern of gravity overturns of three layers of ice in the ~30 km 418 thick crust of Triton (Schenk & Jackson, 1993). Similarly, a space filling pattern of pits, spots and 419 domes (5- to 30-km-in diameter) on the sparsely cratered ice-rich surface of the Jovian satellite 15 420 Europa have been attributed to solid-state thermal convection in the ice shell as a result of tidal 421 heating (Pappalardo et al., 1998), a concept since partially tested by numerical models (Showman 422 & Han, 2004). 423 Largely to counteract pre-1988 ideas, Hudec & Jackson (2007) argued that buoyancy is 424 unimportant for initiating salt diapirism and, in effect, only converts reactive diapirs of sufficient relief 425 into active diapirs. They also argue that “Most salt diapirs in the world pierced during extension 426 events” (Hudec & Jackson, 2007, p.26). However, no large scale diapirs feed terrestrial namakiers 427 in modern extension settings (Mohr et al., 2007) which is why Fig. 4B illustrates a reactive diapir in 428 the locally oblique extensional setting that is usual in Iran which is subject to regional N-S 429 shortening. 430 431 3.3 How important are lateral forces? 432 Two forces are usually invoked as drives for the extrusion of emergent salt diapirs: gravity and 433 lateral tectonic forces. However, noting that most salt sheets form in compressed orogens and the 434 compressed toes of passive continental margins, Hudec & Jackson (2006) went further and 435 suggested that salt sheet emplacement requires lateral compression. 436 The effects of gravity on salt sheets can only vary at the rates that sedimentation can thicken 437 the overlying rocks or erosion can thin either them or the salt. Typical sedimentation and erosion 438 rates are a few m Ma , about the rate that subsurface salt diapirs rise. One might expect, therefore, 439 that lateral tectonic forces vary faster than the effects of gravity. However, such effects are not 440 visible in the 3 best-documented salt extrusion rates in the literature. -1 441 Arguing that the heights of salt mountains are potential tectonic strain gauges, Talbot & Alavi 442 (1996, fig. 9) used them to map active faults in part of the Zagros Mountains laterally shortening 443 where Arabia and Asia converge. However, most of the salt extrusions in the Zagros are no higher 444 than the highest peak of country rocks nearby- implying that gravity is their main drive (Talbot 445 1998). The only Zagros salt extrusions that rise higher than their surroundings are the offshore salt 446 islands close to the Zagros deformation front but even these could be supported by dense 447 carbonates loading their deep salt source. 448 The heights of salt fountains as tectonic gauges are complicated by the climatic signal (Talbot 449 and Jarvis, 1984). More direct gauges of the forces extruding them are emergent salt diapirs that 16 450 have had their extrusive salt removed or bevelled by sea-level at known times. This happened to 3 451 salt diapirs emerging close to two of the boundaries of the lithospheric plate capped by the Zagros 452 Mountains. 453 Mount Sedom, a diapiric salt wall extruding from the Dead Sea oblique-extension fault zone, was 454 turned into such a simple tectonic gauge when the precursor to the adjoining Dead Sea removed 455 any earlier extruded sat and trimmed its crest to a smooth flat surface at ~14 ka. Using geological 456 markers, precise ground leveling and Synthetic Aperture Radar interferometry, Weinberger et al., 457 (2006) were able to constrain the rates of rise of different parts of Mount Sedom to between 5 and 9 458 mm a over the last 14 ka. After carefully modeling the Sedom architecture, these workers 459 attributed the steady salt extrusion solely to gravity without any need for additional lateral forces. 460 Similarly, a small group of diapirs of a much older salt in and beside the Hormoz straits on the -1 461 southern contracting boundary of the same lithospheric plate were also converted into simple 462 tectonic gauges when they were partially truncated by the Persian Gulf at 9 ka BP. Using the 463 current altitudes of dated oysters and bivalves that lived close to sea level in the sediments 464 deposited on these now-tilted Holocene marine terrace, Bruthans et al., (2006) were able to 465 constrain the rates of rise of the centres of two such diapirs to 7 ±1 mm a-1 over the last 9 ka BP. 466 Despite involving salt of two different ages, all the three above diapirs are reported to have 467 extruded at the same steady rates (near 7 mm a-1) for several millennia. This is particularly 468 remarkable for Mount Sedom that extrudes from the Dead Sea oblique-extension fault zone where 469 major earthquakes associated with surface faulting were common in the two millennia before 1894 470 but have been absent since (Ambraseys and Jackson, 1998, cf. figs 4 & 5). This is one of the few 471 documented cases where the historical record indicates a possible change in either the direction or 472 values of lateral tectonic forces. The steady extrusion of Mount Sedom from a fault along which 473 major earthquakes stopped ~200 years ago implies that gravity is its main driver. Hereafter we will 474 assume that the influence of lateral tectonic forces and their changes on salt extrusions are 475 sufficiently small that they can be neglected. We also consider it unlikely that lateral forces in their 476 bedrocks are likely to be transmitted into ice bodies. 477 478 479 17 480 4. Compositional layering, inclusions and moraines 481 Apart from falls of volcanic tephra that sometimes add spectacular competent marker layers to 482 ice masses, most primary stratification in ice is picked out by differences in grain-size and bubble- 483 content surviving from layering in the initial snow pack and such diagenetic processes as refreezing 484 of ice due to raditative and advective heat events (Hudleston and Hooke, 1980). The resulting 485 layering is comparatively subtle compared to stratification in salt, largely because fewer impurities 486 accumulate with snow than commonly accumulate with salt. 487 While many salt sequences are interlayered with other sediments or igneous rocks that can 488 range in thickness from mm to km, beds of relatively pure, clear and transparent salt often reach 489 thicknesses of hundreds of m. Stratification, layering or bedding in such pure salt is defined by 490 halite of different grain size and colours due to various concentrations of different impurities (e.g., 491 magenta clay, buff-coloured anhydrite, black iron or green volcanic dust). The colours in extrusive 492 salt sheets generally intensify down their exposed lengths as dissolution concentrates the insoluble 493 components. Even in salt that may have flowed 10 km along the source layer, perhaps 5 km up the 494 diapir, and 5 km downslope over the surface, compositional layering in distal salt is generally 495 accepted as inherited from the initial bedding and can look like simple planar bedding until colour 496 repetitions and rare isoclinal fold hinges attest to the huge strains the salt has undergone (Talbot, 497 2004). 498 Like ice, exposed salt bodies develop nearly every known category of karst feature although 499 subsurface stream channels erode down to base drainage levels rather than the base of salt (e.g. 500 see Bruthans et al., 2007) 501 Unlike ice, salt extrusions rarely pluck, quarry or incorporate their surrounding strata so most 502 inclusions (that can range in siz from km to microscopic) are part of the initial salt sequence 503 disrupted and dispersed by salt flow in the diapir and are equivalent to en-glacial moraines 504 (Weinberg, 1993). Denser than salt, even the largest solid inclusions (e.g. 3 x 2 x 03 km- Kent, 505 1958) carried >3 km up active diapirs in Iran can be stranded on the modern surface or choke the 506 vent when the salt dissolves (as in the Flinders Range of, Australia: Dyson, 2004). Large dense 507 inclusions lifted high in an active diapir can sink back through otherwise inactive salt diapirs (Koyi, 508 1991; Chemia et al., 2008). 18 509 By contrast, submarine salt extrusions are often buried by beds of younger rocks that can be 510 disrupted by salt flow to simulate supra-glacial moraines that are often subsequently shed as 511 complex slumps equivalent to side or terminal ice moraines (see Fig. 1SA). Inclusions concentrated 512 at the sutures of submarine salt canopies help to delineate where salt sheets have coelesced 513 (Rowan, 2003). 514 Push moraine-type structures with simple thrusts and/or imbricate thrust complexes are 515 common in the soft sediments in or beneath the fronts of salt sheets in the Gulf of Mexico (Jackson 516 and Hudec, 2004). However, subaerial salt sheets do not erode U-shaped channels or quarry and 517 abrade such distinct bed-forms as roches moutonée or striations like ice because they attach to 518 their substrates along frictional contacts. Both ice and salt are too weak to abrade hard substrates 519 significantly but, more importantly, because most extrusive salt is dissolved by rain falling from 520 above, potential abrasive tools concentrate on the tops of namakiers rather than their bases (Fig. 521 1SA). This means that the only saline equivalents of sub-glacial moraines are shed and overridden 522 en-and supra-glacial moraines. Saline equivalents of eskers are likely to exist locally but have not 523 yet been recognised as such. 524 There also appear to be very few saline equivalents of the glaciotectonic structures in weak sub- 525 glacial substrates or moraines (Boulton, 1996; Maltman et al., 2000, p.4). What appear to be 526 ground moraines (Fig. 1SA) are equivalent to “melt-out” tills that remain where supra-salt moraines 527 that cloak large areas of most salt extrusions are dropped after eventual salt dissolution (Fig. 1SA, 528 Talbot, 1979, 1998). Conical mounds of loose residual cap soils-and-blocks accumulating on top of 529 dissolving salt commonly become badlands with steep slopes scarred by landslides. 530 Former moraines and ice-sculpted features are the main records of the many past ice ages. The 531 equivalent of erratic blocks and melt-out tills remaining after the complete dissolution of past 532 extrusions of salt in Iran are recognisable as black erratic blocks ± red soils spread over the plains 533 (Fig. 1SA) or interbedded among the surrounding buff coloured carbonates Such insoluble residues 534 in the stratigraphy of subsequently downbuilt sediments show that many of the salt diapirs currently 535 active in the Zagros Mountains first began extruding as long ago as the Triassic, 220 Ma ago 536 (Player, 1969; Kent, 1970; 1979). 537 538 There is evidence that evaporites formed in Archaean times (Warren, 1997). However, because salt is so weak and soluble, it is usually recycled back to the world’s oceans as the surrounding 19 539 rocks pass through diagenetic into metamorphic facies. The salt is dissolved while it is still in place 540 or after it has been squeezed to the surface by the early stages of orogeny. Just as there were 541 many former ice ages, many former salt sequences have already been recognised from their 542 metamorphosed remnants or their mega-breccias after dissolution (see Warren, 1997; Jackson et 543 al., 2003). 544 545 5. Internal fabrics, structures and deformation mechanisms 546 547 Deformation structures and fabrics on all scales in glaciers and namakiers resemble those in 548 more usual metamorphic rocks and naturally model them at rates that can be monitored over a few 549 years. 550 5.1 Grain-scale fabrics 551 The hexagonal crystal lattices in a grain of ice are mechanically anisotropic, while the cubic 552 lattice of a halite grain is comparatively isotopic. Initially isotropic materials (like snow) develop 553 anisotropic (mainly subhorizontal) fabrics (as in new glacier ice) by homogeneous compaction and 554 planar markers (like stratification) can develop structures (like folds) by inhomogeneous strains. 555 Markers are strain active where they involve materials that deform at different rates (across 556 contacts that refract foliations) or strain passive where the strain rates of any distinguishable 557 components are negligible (e.g., colour bands). 558 Grain boundaries in flowing ice and salt can readily dynamically recrystallise and anneal (Fig.8). 559 As a result, grain shapes in deformed ice and salt (Table 1) are said to have short strain memories 560 because they record only the latest stages in their strain history (Talbot and Jackson, 1987). The 561 strain memories of grain fabrics in ice are even shorter than those in salt. This means that both ice 562 and salt that can look much the same however far it has flowed. Thus drill cores of salt (~170 Ma) 563 that has probably flowed distances approaching 100 km near the floor of the Gulf of Mexico can still 564 display horizontal beds of uniform thickness made up of isotropic grains cm across (Talbot, 2004) 565 just as distal and proximal ice in an Antarctic ice sheet can be indistinguishable. 566 Fabrics in salt deformed in different environments show different dominant deformation 567 mechanisms (Fig. 8) so their rheological behaviour should be described by different flow laws 568 (Schleder & Urai, 2008). Thus for still-allochthonous (Zechstein) salt, the subgrain sizes indicate 20 569 stresses between 0.6 and 1 MPa at depths implying deformation temperatures <50ºC (Schleder & 570 Urai, 2008). In thin sections, patches of primary crystallisation structures full of dispersed primary 571 brine inclusions are surrounded by areas of recrystallised grains with evidence of grain boundaries 572 that migrated to collect fluid inclusions and become very mobile by solution-precipitation creep. 573 Solution-precipitation creep and dislocation creep contribute almost equally to the total strain rate 574 and the equation describing the salt rheology should reflect this. 575 Diapiric salt tends to consist of grains of uniform size (e.g. 0.5-5 cm) but irregular isotropic 576 shapes implying that, like ice, it can flow large distances with a steady-state dynamic recrystal- 577 lisation fabric involving mainly dislocation creep and fluid assisted grain boundary migration. Every 578 grain is competent with respect to its neighbouring grain boundaries and so undergoes the 579 oscillating super-shear strains of stiffer inclusions rotating with vorticity higher than the bulk flow 580 (Ramberg, 1975; Ramsay and Huber, 1983, Talbot and Jackson, 1987; Talbot, 1992, 2004). The 581 application of experimentally derived flow laws relating the steady-state size of sub-grains with flow 582 stress indicates that diapiric salt deforms by climb-controlled dislocation processes at rates 583 between 10 584 40-150ºC (Schleder and Urai 2006, 2008). 585 -11 to 16 s –1 at differential stresses 1-5 MPa at depths implying deformation temperatures Diapiric salt with its typical course uniform grain size is too strong to flow in subaerial surface 586 conditions. Extrusive salt in Iran exhibits the necessary weakening process in fabrics that have 587 bimodal grain sizes with an average grain size that decreases downslope (Fig. 8SD-SE). Such 588 bimodal grain size fabrics are rare in ice and render shape and orientation grain-fabrics in salt much 589 more obvious than most secondary foliations picked out by discontinuous layers of different grain 590 size and bubble content in ice (Hambrey and Lawson, 2000). The bimodal grain sizes of extrusive 591 salt means that foliations can appear with new orientations (e.g. axial planar to new folds) over 592 distances as short as a few dm (Talbot, 1979). As in all rocks, foliations develop parallel to the 593 longest and intermediate axes of the cumulative strain ellipse that fomed them (Hambrey & 594 Lawson, 2000). 595 Salt extruding from a diapiric vent surfaces with both a strong a horizontal stratification and 596 gneissose foliation as a result of strong lateral extension and vertical flattening in the crest of the 597 summit dome (Fig. 7). Halite megacrysts (cm across, e.g. Figs. 8SA-SE) surviving from the diapir 598 exhibit very small subgrains indicating dislocation climb at stresses 3-5 MPa in the diapir. These 21 599 porphyroclasts decrease in size and number downslope as they disperse within increasing volumes 600 of smaller (< a few mm) daughter grains shed from their cores. The lack of subgrains in the small 601 daughter grains indicates flow by pressure solution at differential stresses of <<1 MPa by pressure- 602 solution, at temps from 20-40ºC implying effective viscosity values as low as 1014 Pas (Schleder & 603 Urai, 2006, 2008). Dynamic recrystalisation decreases the grain size of both salt (Fig. 8 SE) and ice 604 in shear zones (Figs. ID & IE, Hudleston and Hooke, 1980; Wilson and Zhang, 1996). 605 Deformation fabrics in extrusive salt can thus be defined by the shapes of both the large 606 porphyroclasts and the orientations of the surrounding smaller grains of halite ± spicules of 607 anhydrite (CaS04) or specularite (Fe2SiO3). Shape and orientation fabrics in the porphyroclasts are 608 more obvious (Fig. 8SD) but the fabrics in the finer grained groundmass are often stronger. 609 Nevertheless, grain shape fabrics are close to isotropic over most bodies of flowing salt (Fig. 8SC) 610 and become noticeably anisotropic only locally. Halite grains with length/width ratios of <1.5:1 611 characterise zones of salt gneisses tens to hundreds of metres wide (Fig. 8SB-E), and smaller 612 halite grains (<~0.2 mm) with length/width ratios up to 5:1 characterise salt mylonites that may be 613 hundreds of metres long but are seldom >1 dm wide (Fig. 8SE, Talbot 1979, 1981). Lacking 614 crystallographic preferred orientations, the micro-fabrics in halitic mylonites indicate that they 615 deform by solution precipitation creep involving non-conservative grain boundary migration and 616 grain boundary sliding (Schleder and Urai, 2006). Sub-grain sizes indicate that such mylonites are 617 capable of the rapid strain rates of 10 618 temperatures of ~10 to 40ºC found by field measurements of markers on namakiers after recent 619 rain (e.g. Talbot and Rogers, 1980; Talbot et al., 2000). 620 -10 -1 s at differential stresses of 0.4 MPa and surface The simple listing of physical properties (Table 1) obscures some of the complexities that must 621 be borne in mind when considering the deformation of natural bodies of ice or salt. Thus, both ice 622 (Hambrey & Lawson 2000) and salt (Critescu & Hunsche, 1998) deform with bulk properties that 623 depend, on their purity, grain size, water content, and fabric as well as the PT conditions; scale also 624 becomes important if strain inhomogeneities such as shear or mylonite zones are included within 625 the volume of interest. 626 On scales >1 km, the smooth outlines of both intrusive and extrusive salt structures (Figs. 5-6), 627 together with the parabolic profiles of Holocene terraces on diapirs truncated by marine erosion 628 (Bruthans et al, 2006), indicate that confined by loads >~1 Mpa salt deforms as a linear viscous 22 629 crystalline fluid (n=1) consistent with model formulae indicating effective viscosities in the range 630 1016-18 Pa s (Critescu & Hunsche, 1998). Future formulae for the mechanics of rock salt flowing 631 from the conditions of alllocthponous source depths through diapirs to namakiers will have to 632 extend the range of effective viscosities to at least 1014-19 Pa s (Schleder & Urai, 2008). 633 Because glaciologists usually treat ice as a strain-rate-softening power-law crystalline fluid with 634 a stress exponent near 3 since Nye (1952) they are wary of quoting even effective viscosities. 635 Shumskii (1964) wrote of 10 10-11 14 Pa s for “warm” ice and 10 Pa s for “cold” ice. 636 The not-quite parabolic arcs of insoluble moraines on the tops of namakiers (Fig. 2A) can be 637 compared to plug flows along channels. The best-fit between such arcs and the non-dimensional 638 velocity curves for channel flow (Turcotte and Schubert, 1982) implies that gneissose salt on scales 639 of hundreds of m flows as a strain-rate-softening power-law fluid with a strain exponent near 3 640 (Talbot and Jarvis, 1984), as commonly taken for ice (Nye, 1965; Walker and Waddington, 1988; 641 Hambrey & Lawson, 2000 but see Harper et al., 1998). On still smaller length scales (<dm), the 642 power-law exponent of dilated salt increases to account for the n= 6-8 measured in mine, borehole 643 and laboratory tests (Cristescu and Hunsche, 1998). We will return to strain rates when considering 644 surging flows in section 9. 645 646 5.2 Macroscopic structures 647 From now on we will attempt to distinguish the deformation structures and fabrics that develop 648 within visco-elastic fluids flowing down gravity-driven pressure head gradients into three categories 649 on the basis of their origins: 650 Kinematic deformation structures and fabrics are those that develop where passive markers 651 (like stratification and foliation) develop new structures where they adapt to changes in the 652 geometries of their external boundaries they flow past. Examples are where passive markers and 653 flow lines in ice or salt remain parallel but develop forced or drape folds over irregularities in their 654 basal boundaries or follow bends in the rigid boundaries of a channel. 655 Dynamic deformation structures (± fabrics) are the result of inhomogeneous strains imposed on 656 strain-active markers by differential stresses. Examples are buckles in competent markers 657 shortened along their length, pinches or boudins developed in competent markers extended along 23 658 their length, mullions in incompetent markers shortening along their length and “inverse folds” in 659 incompetent markers extened along their length. 660 Lateral tectonic forces are more likely to strain salt bodies (e.g. by changing channel gradient) 661 than shorter-lived ice bodies. Nevertheles, regional stress fields (dominated by gravity) generally 662 impose longitudinal extension in the upper reaches and longitudinal compression in the lower 663 reaches of both glaciers (Hambray & Lawson, 2000) and sheets of allochthonous salt. 664 An important sub-category of dynamic structures (± fabrics) can develop where local changes 665 in velocity or volume induce flow lines to cross passive markers. The most obvious way to change 666 local volumes or velocities is to invoke local changes in the budget of supply relative to loss of the 667 flowing fluid. As a result, such structures will be called here budget change induced structures (± 668 fabrics). Examples are recumbent flow folds induced in ice or salt by budget changes (Hudleston, 669 1976). 670 Dynamic structures (± fabrics) in another sub-category develop where flowing fluids encounter 671 (or are subject to) changes in frictional resistance along parts of their external boundaries. Such 672 change in the forces retarding flow can be local iin time or space. Examples of such changes in 673 space are where flowing ice detaches from a rock substrate and thins as it accelerates over a free- 674 slip base with water; another is where salt in high frictional contact with the surrounding rocks 675 extrudes into air. Examples of changes of boundary conditions in time are where batches of surface 676 melt-water hydraulically fracture their way to the base of a glacier or ice sheet (Sohn et al., 1998) or 677 where rain weakens the top boundary of a namakier. We refer to these as boundary induced 678 budget change structures. 679 In contrast to the short strain memories of grain shape fabrics in deformed ice and salt, 680 macroscopic markers like stratification and veins, have indefinite strain memories and can record 681 large cumulative strains that can be lost by ablation, melting or dissolution, but are unlikely to 682 anneal (e.g. for salt see: Talbot and Jackson 1987). Most macroscopic strain markers in ice are 683 more or less strain active because compaction and diagenetic layering increasingly orientates the 684 bulk mechanical anisotropy of their constituent ice crystals. Thus stratifications with conformable ice 685 lenses are slightly strain active, crevasses, crevasse traces or infilled-crevasses are moderately 686 strain active and dirt bands, silt beds and tuff layers are strongly strain active. 687 24 688 689 5.3 Macroscopic kinematic structures Stratification in both ice and salt form horizontal and usually first deform by flowing down nearly 690 horizontal load pressure gradients. Clear 3D grain shape and orientation fabrics with short strain 691 memories map the flow lines. They generally define a planar foliation ± a lineation although the 692 lineation can dominate in parts of some salt structures (e.g. constricted in the stem of a diapir or 693 rolled around the hinges of folds in extruded salt: Talbot and Jackson 1987). In steady penetrative 694 flows, flow lines and the foliation parallel stratification and their nearest external boundaries and 695 remain parallel as they sweep together around changes they encounter in their external 696 boundaries, even right angle corners (e.g. in planform for glaciers; from source-to-diapir and diapir- 697 to-extrusion in profiles of salt). The top boundary is generally a smoother version of the often- 698 irregular bottom boundary in both namakiers and glaciers. The infills of moulins and crevasses in 699 ice can develop vertical foliations (Hooke and Hudleston, 1978). 700 The stratification in ice converging on a narrow channel from several accumulation basins, or 701 downstream from nunataks (bedrock highs exposed through ice), tends to develop upright 702 symmetrical kinematic folds with subhorizontal axes and a vertical axial planar foliation (Fig. 2ID-E, 703 Hambrey et al., 1999; Iizuka et al., 2001). Equivalent kinematic folds and foliations develop in the 704 deep source layer where flowing salt converges radially on the base of the diapir stem (Ramberg, 705 1981; Talbot and Aftabi, 2004). The radial and subhorizontal axes of these folds turn through a 706 kinematic bend to become vertical up the diapir where they are known as curtain folds (Talbot and 707 Jackson, 1987). These curtain folds emerge built into the salt that first surfaces on the crest of a 708 salt extrusion. Rather than open in the divergent radial flow within the summit dome, the steep axial 709 surfaces of curtain folds turn through another kinematic bend (related to the divergent flow points 710 starred in Fig. 7) to parallel the top free surface of the extruding salt (Talbot and Aftabi, 2004). As a 711 result, unlike ice that usually only accumulates strain in its compaction and downslope flow, most 712 salt is extruded from depth reaches the surface with mature folds already incorporated into the 713 stratification. Indications of these folds can persist into the most distal salt as local layers with steep 714 along-flow strikes. 715 716 717 25 718 719 5.4 The kinematic tank-track fold Analogue experiments (Ramberg, 1981, pp. 219-226) studying how gravity spreads rectangular 720 blocks of elasto-viscous silicone putty with different length/ height ratios over frictional and no-slip 721 basal boundaries are relevant to the effect of gravity spreading bodies of both salt and ice. Where 722 blocks of silicone putty spread over a horizontal no-slip basal boundary, their initially vertical sides 723 bulged and then advanced by rolling their tops over the front and attaching to the substrate (Fig. 724 9A1-3). As a result, initially-horizontal passive marker layers near the distal edges of experimental 725 viscous flows advance like lavas and caterpillar tank-tracks. The top surface overtakes the toe, rolls 726 down a steep flow front (and over any debris spilled from it) and attaches to its substrate along a 727 frictional basal boundary that lengthens with further advance. Passive stratification in the flow rolls 728 over the hinge of a kinematic tank track fold with an axial surface that climbs slightly as it elongates 729 behind an axis parallel to the the advancing flow front (Figs. 7 and 9 and Talbot, 1998). 730 Stratification that faces upward in the upper limb (thinned to invisibility in Figs. 9B-F) is inverted to 731 face downward in their strongly sheared lower limbs. 732 In extrusion experiments (Talbot and Aftabi, 2004), the upper limbs of tank-track folds are 733 simple if they advance steadily over planar horizontal substrates (Fig. 9B) but develop flow folds 734 with geometries that relate to changes in the relative rates of extrusion and gravity spreading (Fig. 735 9C-F). Despite significant dissolution from their top surfaces, every subaerial salt extrusion studied 736 in Iran involves a fundamental recumbent sheath-like tank-track fold. In marked contrast, major 737 recumbent folds have been reported from very few distal profiles of glaciers and ice caps and those 738 have been attributed merely “to ice flowing fastest at mid-levels” (Shumskii, 1962). 739 Hudleston (1976) described discontinuous examples of recumbent folds near the base of the 740 Barnes ice cap, listed a few other examples in the literature, and expected such folds to be more 741 common than then generally recognised- but very few additional examples have been added since. 742 Chinn (1989) attributed single folds in short dry-based glaciers to their advance- but invoked basal 743 buckling and thrusting of a sub-glacial till rather than the upper limb of a supra-glacial till rolling over 744 the advancing front and then being truncated by ablation. Recently several authors have 745 investigated the potential for folding near the base of ice caps in order to understand the deposition 746 and deformation history in ice cores taken from ice caps (e.g. Jacobson and Waddington, 747 2005). 26 748 A clue for this absence comes from blocks of silicone putty spreading over free slip basal 749 contacts in the laboratory. These spread rapidly over mercury and thinned uniformly by almost 750 uniform pure shear so that passive horizontal markers curl upward in distally as a result of their toes 751 spreading faster than their tops (Figs. 9A4-6). Whereas stratifications curling upward distally are so 752 far unkown in subaerial namakiers that have advanced over frictonal substrates of rocks and soils, 753 they appear to be the norm for profiles of glaciers and ice sheets (Hambrey and Lawson, 2000) 754 suggesting that they have low basal friction (upstream of a periphery with high basal friction in cold 755 examples). Seismic profiles rarely image stratification within submarine salt sheets but the most 756 distal few km of their bases curl upward above simple thrusts and/or imbricate thrust complexes in 757 some of the frontal sediments beneath the Gulf of Mexico (Jackson and Hudec, 2004). Submarine 758 salt sheets probably advance faster and further over saturated sediments than subaerial namakiers 759 usually flow over dry rocks and soils. In summary, we infer that stratifications curling downward 760 distally signal ductile flow over high-frictional basal contacts, whereas stratifications curling upward 761 distally signal flows over low-frictional basal contacts. 762 763 5.5 Kinematic structures associated with irregular bedforms 764 In addition to their fundametal tank-track folds, namakiers with irregular beds have fairly well 765 developed basal zones of folds ± thrusts like many glaciers (Shumskii, 1964 p. 444). Most of these 766 structures are dynamic and treated in the following section (5.6); here we mention kinematic 767 undulations where the fluid adapts to irregularities in its bottom boundaries. Both ice, salt and 768 viscous analogues develop internal upright kinematic similar-type folds (class 2 folds of Ramsay 769 1967) wherever gravity drapes their internal structures over irregularities in their beds (Figs. 7C & 770 9F). Such kinematic “undulations” are characterised in vertical profiles by upright folds with 771 amplitudes that diminish upward from the base but can reach the top free surface as ogives. Even 772 the stratification of near-basal ice beneath the current divides of ice sheets develops such upright 773 folds over bed relief (Souchez et al., 2000; Waddington et al., 2001; Thorsteinsson and 774 Waddington. 2002; Jacobson and Waddington. 2005). Equivalent structures in other rocks are 775 known as compaction, drape or forced folds (Cosgrove and Ameen, 2000). The dynamic folds of 776 the next section can appear upright in profiles perpendicular to the direction of flow but are 777 asymmetric in other sections. 27 778 5.6 Budget change induced flow folds and their axial planar foliations 779 Axial surfaces to drape folds that are vertical in near-basal ice can curve upward to become 780 axial planar to asymmetric reclined, even recumbent folds, where carried downstream. Anatomising 781 stripes, planes of ice grains with c-axes approximately in the dip direction of the planes, are 782 commonly axial planar to both deep symmetric upright folds and less deep asymmetric folds (Alley 783 et al., 1997). Such asymmetric folds in ice are dynamic in the sense that they record variations in 784 flow velocity in either space or time. Hudleston’s (1976) explanation for how such folds develop in 785 glaciers when the flow is unsteady on a wide variety of time scales is equally applicable to 786 namakiers. Folds develop wherever and whenever flow trajectories cross passive markers. Minor 787 advances and retreats change the thickness and/or surface slope of glaciers in time so that 788 markers already draped over an irregular base deform into increasingly asymmetric folds as they 789 are carried downstream. Complex budget changes over complex basal irregularities can lead to 790 multiple recumbent folds in distal ice and salt (Fig. 10). 791 It is not clear whether all folds verging downstream in ice are deformed upright folds or if any 792 can initiate as such (Alley et al., 1997). However, many folds within namakiers certainly initiate with 793 asymmetries, perhaps because the bimodal grain size of extrusive salt allows the generation of new 794 foliations to record a new set of stream lines over only a few dm (Talbot, 1979). 795 Our understanding of how dynamic flow folds generate near the base of flowing fluids benefited 796 considerably from two linked papers by Gudmundsson (1997a & b). In this paragraph (only) we use 797 the term “extrusive flow” in the usual glaciological sense of referring to increases in horizontal flow- 798 velocities with depth. Gudmundsson (1997a) introduced a form of Local Extrusion Flow (LEF) 799 where the vertical expansion and contraction of the ice changes laterally close to an undulating 800 free-slip bed so that LEF above the trough dominates that above the ridge. Rather than adding to 801 ice flux as considered normal for the general extrusive flow of the preceding glaciological literature, 802 Gudmundsson's (1997a) LEF instead decreases ice flux by the general flow separating from, and 803 flowing beyond, ice stagnant in troughs. The effect of sinusoidal bed relief is largest a distance 804 above the bed that relates to the depth/width ratio of its relief. In appropriate conditions, a high- 805 pressure zone can develop above the bed and superpose a Poiselle flow on the gravity driven 806 plane flow. Gudmundsson (1997b) used numerical models to show that LEF increases with the 807 power law exponent of stress. (When and where the depth of a trough is >1.5 times its along-flow 28 808 width, the main downslope flow to the right will separate from, and flow over ice that remains in the 809 trough and is either stagnant or slowly circulating in a clockwise corner eddy). Multiple flow folds 810 can develop where flow perturbations have higher harmonics than the bedrock topography inducing 811 them, as found empirically by Hooke et al., (1987) and experimentally by Reeh (1987). 812 Similar phenomena have been described from a different viewpoint when relating trains of 813 asymmetric folds and slides in a namakier to step-like (rather than sinusoidal) irregularities in its no- 814 slip bed (Fig. 10 from Talbot, 1981). Rather than eroding obstructions, namakiers instead smooth 815 their channels by infilling irregularities with static salt and then detaching from and flowing beyond 816 the infills alnong near-basal mylonite zones. 817 To surmount a steep upstream-facing step in its bottom boundary, a namakier slows and 818 thickens. The gneissose grain shape fabric mapping the flow surfaces diverge in the decelerating 819 salt and cross the passive colour layers to become axial planar to a suite of new flow folds. These 820 similar-type folds amplify and tighten as the flowing salt detaches along mylonite shear zones and 821 flow up and over static veined salt immediately upstream of the stationary step (Fig. 10B). The 822 stratification curls upward distally as though the near-basal mylonite zones act as low frictuional 823 detachments between the ice above and below them. Downstream of the obstruction, the foliation 824 converges in the accelerating salt, the folds tighten to isoclinal, and the mylonite zones that acted 825 as thrusts upstream, turn to slides (extending detachments) downstream (Fig. 10B). A steep strike- 826 slip shear zones marks where an active salt stream by-passes salt now damed by bedrock ridge in 827 response to budget changes (Talbot 1979). 828 Although folds of most categories in salt structures have been simulated in laboratory 829 experiments with viscous ductile fluids, trains of dynamic folds relating to irregularities in the bottom 830 boundary of namakiers (Talbot, 1979) have not (Talbot & Aftabi, 2004 p. 331). 831 832 833 5.7 Budget induced dynamic folds near top boundary Mapping the vergence of minor folds and how axial planar foliations relate to stratification implies 834 that major similar-type folds face outward in the upper slopes of salt fountains. All subaerial salt 835 extrusions studied in Iran are characterised by these major flow folds in the upper limbs of their 836 tank-track folds (Fig. 7A-D). Absent from the crest of the dome, these folds first appear as gentle 837 and open reclined asymmetric flow folds in layering on the steep (>15°) upper slopes of the dome. 29 838 Axial surfaces dip steeply upslope where they first develop but flatten in successive folds 839 downstream as they amplify to isoclinal recumbent folds in distal salt (Fig. 7C-D). Their axes 840 parallel the flow front and therefore sub-parallel smoothed topographic contours on the salt. These 841 similar-type flow folds are salt equivalents to the ductile fold-thrust nappes that characterise the 842 metamorphosed cores of orogenic belts of all ages and, like Helvetic-type nappes, their lower limbs 843 often detach along mylonitic shear zones or slides (e.g. Pfiffner, 1993). No equivalent folds appear 844 to have been reported from ice. 845 It is easy to simulate viscous extrusions in the laboratory by loading a horizontal wooden board 846 (representing stiff overlying rocks) floating in a pan of viscous fluid (e.g. polydimethylsiloxane GM- 847 26, a transparent silicone polymer with a viscosity of ~ 5.105 Pa s in Talbot and Aftabi, 2004). The 848 sinking board displaces the viscous fluid (representing the salt) up around its edges (from where it 849 is removed) and up a central hole that simulates the diapiric vent of a salt extrusion. Thin horizontal 850 colour bands built into the source layer represent passive stratification in the salt. 851 If suitably small loads are applied to the board, the fluid rises slowly up the rhomboidal vent and 852 extrudes over the board with the profile of a droplet (Huppert, 1982) or a miniature ice cap (dashed 853 in Fig. 6B). Simulating realistic loads on salt layers buried beneath overburdens a few km thick 854 lead to the extrusion of a dome that spreads as an axisymmetric viscous fountain with a diameter of 855 about a dm after 6 hours and 2 dm after 24 hrs. 856 In the laboratory, it is easier to change the rate of viscous extrusion than to change the rate at 857 which it gravity spreads or is removed by erosion. Thus the viscous fluid can be extruded in pulses 858 separated by brief intervals without a load when the spreading rate of the extrusion changes 859 comparatively little (Talbot and Aftabi, 2004). The number of outward facing folds in the upper limb 860 of the basic tank-track fold is a direct measure of the number of loading phases (Fig. 9B-F). This is 861 because each renewal of the load significantly increases the extrusion rate so that resurgent flow 862 trajectories cross the passive markers re-orientated by the gravity spreading that continued during 863 the pause in loading. The resultant flow folds therefore have budget-induced dynamic origins. The 864 relative lengths of their lower limbs reflect intervals of increasing extrusion rate (or slowing gravity 865 spreading in nature) and the lengths of their upper limbs reflect intervals of falling extrusion rate (or 866 accelerating gravity spreading in nature). Early flow folds rolled over the axis of the tank-track fold 867 are effectively unfolded by strong shear along the bottom boundary. 30 868 In essence, the laboratory loading history of an experimental viscous extrusion can be read 869 from the shapes of folds in the upper limbs of its tank-track fold (Talbot and Aftabi, 2004). 870 Generating experimental dynamic folds by varying the extrusion rate relative to the spreading rate 871 does not rule out the effects other mechanisms in nature. Indeed, the spreading rates of natural salt 872 extrusions depend on the climate and are thus likely to change much faster than changes in the 873 extruding forces. If the extrusion rates of emergent salt diapirs are essentially steady because they 874 are driven mainly by gravity (section 4.3), then the major dynamic flow folds within subaerial salt 875 extrusions in Iran are more likely to record significant (millennia-scale) changes in their rates of 876 gravity spreading than any slower changes in their extrusion rate. Because warm salt flows faster 877 than cold salt and, even more significant, wet salt flows faster than dry salt (Urai et al., 1986), 878 changes in the rate of gravity spreading of extruded salt depend mainly on the climate. Each of the 879 major folds within Iranian salt extrusions therefore presumably record wet intervals when the salt 880 gravity spread faster relative to its almost steady extrusion rate. The structures within salt 881 extrusions therefore offer proxy records of climate changes, particularly rainfall. The simplest test of 882 this hypothesis is that salt extruded from different vents in the same region over the same time 883 interval should have developed similar sequences of flow folds in the upper limbs of their tank-track 884 folds. 885 It is generally accepted that the rise of mountain chains can result in orographic rainfall and that 886 the resultant increase in erosion rate can feed back to influence the orogenic structures (e.g. 887 Beaumont et al., 2001). However, it is still debateable whether the dynamic flow folds represented 888 by piles of orogenic nappes record fluctuations in climate and/or tectonic forces. 889 Hudleston (1976) envisaged flow of ice developing folds where and when the flow lines cross 890 old stratification perturbed by relief in the bottom boundary. However, the clearer folds in salt flows 891 suggest that dynamic flow folds near the top boundary of flowing ice may be even clearer records of 892 budget changes due to fluctuations in climate. The potential for folds in ice to record climate change 893 has been recognised in ice streams (see e.g. Bindschadler, 1993; Jacobel et al., 1993; Merry and 894 Whillans, 1993; and Fahnestock, et al., 2000) but woefully neglected in ice sheets (Alley et al., 895 1997) and glaciers (Hambrey and Lawson, 2000). 896 897 31 898 899 6. Fractured carapaces in ice and salt Every exposed rock mass develops a dilated and fractured outer layer. This elastic carapace 900 is ~10-70 m thick in ice and ~5-10 m thick in salt. Most fractures in both materials initiate with 901 apertures less than a few mm: however, whereas many fractures in extending or shearing ice 902 develop significant gapes as crevasses, very few open in exposed salt (Talbot 1998). 903 Complex crevasse patterns that record strain histories in flowing ice (Nye, 1952) can be 904 emphasised by seracs or ice clefts sculpted by rapid ice sublimation. Dissolution by rain, snow melt 905 and streams (Bruthans et al, 2007) widens a few joints in salt (see photos in Talbot, 1998) but 906 generally frets a jagged salt relief of pinnacles that is largely independent of fractures (Fig. 1SB-E). 907 The topographic relief weathered into extruded salt in Iran increases down-slope from a minimum 908 (e.g., 5 m) near the summit to a maximum (e.g. 40 m) near the terminus (inserts on Fig.7B). 909 Bergschrunds are frequent forms in glacial settings, where they develop between the ice 910 body and the head wall in a combination of thermal and ice dynamics (e.g. Mair, and Kuhn. 1994). 911 Bergschrunds also gape behind a few local salt cliffs <10 m high that shed falls of m-scale blocks 912 every few hours or days (Talbot, 1979) but, in general, the only fractures that gape in salt define 913 top-surface-parallel slabs in about the top 10 m (Fig. 11). These define top-surface-parallel slabs in 914 the outermost ~10 m of all subaerial salt extrusions. When dry, these salt carapaces strain 915 elastically in immediate response to temeperature changes and fracture noisily during rapid 916 temperature drops but accumulate permanent ductile strains silently within minutes of light rainfall 917 (Talbot and Rogers 1980). Talbot (1998) assumed that the carapaces of brittle dilated dry salt 918 encasing Iranian salt extrusions are only veneers as thick as the local salt relief. However, master 919 fractures over a 100 m long suggest that regional stress fields have broken dry salt deep within at 920 least one salt fountain at unknown times in the past (e.g. Talbot and Aftabi 2004). 921 922 6.1 Fracture dynamics 923 Factures develop in elastic solids when and where stresses in them exceed their tensile or shear 924 strength. In ice, such stresses are usually generated by differential flow rates. In salt they, these 925 fractures are attributed to salt extruded from depths of km de-stressing to the free surface on 926 exposure with the aid of weathering processes (Talbot and Aftabi, 2004). 32 927 Monitoring the movements of markers on salt extrusions in Iran (e.g. Talbot and Rogers,1980; 928 Talbot et al, 2000; Talbot and Aftabi, 2004) finds them to move back and forth on time scales from 929 minutes to months or even (locally) years. The (unusually high) standard value of the linear thermal 930 expansivity of salt (Table 1) has been confirmed to an accuracy of a few % by translating to strains 931 the changes in the distance between two wooden stakes 2 m apart and relating them to the 932 temperature changes on the intervening dry ground over a few hours (Aftabi, unpublished MSc 933 thesis, 2000). These elastic strains indicate that the salt expands on heating and contracts on 934 cooling as expected; however salt masses also expand on wetting and shrink on drying. Marker 935 movements are therefore complicated because salt temperatures fall when it rains and rise when 936 sunshine dries the salt. However, in general, thermal expansion and contraction dominates the 937 swelling and shrinkage with wetting and drying. Large as they are, elastic strains are eventually 938 exceeded by ductile strains that accumulate when the salt is damp. 939 The ambient conditions of salt are usually warmer than ice at all depths but surficial salt is 940 commonly subject to larger diurnal and annual temperature changes than ice (30-50ºC). The 941 thermal expansivity of salt is an order of magnitude higher than most other rocks (including ice: 942 Table 1). Calculating the thermal skin depth (τκ/π)1/2 (equation 4.90, Turcotte and Schubert, 1982) 943 for ice and salt with a thermal diffusivity of κ, subject to regular temperature fluctuations with cycles, 944 τ indicate that the daily and annual temperature cycles reach depths of 19cm, and 3.66 m in ice and 945 32 cm, and 6 m in salt respectively. 946 Thermal strains and shocks induced in surficial salt by temperature cycles of about a year 947 therefore account for the outer 10 m or so of subaerial salt bodies in Iran being broken into brittle 948 elastic carapaces. When they are dry during winter sunsets, such salt carapaces sound like a 949 small-arms battles when new fractures develop as a result of rapid temperature falls. Surficial 950 glacier ice also snaps and cracks under thermal stresses but more impressive is the singing of lake 951 ice as it expands near noon in the warm Spring-time sun. Whereas the 10-70 m thick rigid surficial 952 crusts carried by flowing ice are there because they are cold, the 10 m thick crusts usual on salt 953 flows are fractured by falling temperatures but their yield strengths depend on water content. On the 954 way to considering the effects of water on ice and salt we consider their fluid inclusions. 955 956 33 957 7 Fluid inclusions 958 Most if not all mineral grains contain fluid inclusions that trap samples of any volatiles surrounding 959 them when they crystallised. In ice, these chemical and physical impurities are usually ejected from 960 the ice matrix by crystal growth, although some chemical species do dissolve with water in the 961 molecular structure (e.g. Legrand and Mayewski, 1997). Since the first ice coring experiments (e.g. 962 Langway, 1967), impurities in the ice matrix have been used as proxies of climatic and 963 environmental changes, Such studies have grown exponentially in the last two decades (e.g., Petit 964 et al. 1999, Alley, 2002) and a recent ice core from Antarctica documented the record of more than 965 eight glacial cycles over 900 ky (EPICA, 2004). 966 Primary and diagenetic textures in undeformed crystalline salt are distinctive of the environment 967 in which they formed and also trap fluid inclusions that sample the surrounding atmosphere or the 968 brines they evaporated from (Lowenstein and Hardie, 1985). Halite crystals that accumulate in 969 episodically desiccating playas typically grow vertically from horizontal substrates. Cubic inclusions 970 of the brines in which they crystallised are commonly incorporated in distinctive face-parallel 971 chevron-shaped growth zones that point upward where they are not interrupted by wavy dissolution 972 zones (stylolites) recording episodic additions of fresher water. In permanent brines, chevron 973 crystals of halite plus inclusions radiate from reef-like substrates and can grow more or less 974 continuously and very rapidly (Talbot et al., 1996). 975 Growth and cementation fabrics inherited from the initial crystallisation of salt can survive 976 indefinitely in salt confined by burial without flow. Studies of paleoclimatic records in salt lag behind 977 such studies in ice. Nevertheless, a 186 m core of salt interbedded with muds from Death Valley in 978 California documented a 200 ky paleoclimate record (Lowenstein et al. 1999). Similarly, a 106 ky 979 paleoclimate record has been recovered from a 100 m salt core from northern Chile and 980 demonstrated that wet conditions in tropical South America were coeval with cold sea surface 981 temperatures in the high-latitude North Atlantic on both millennial and orbital (20,000-year) 982 timescales (Bobst et al., 2001; Baker et al., 2001). 983 Paleoclimatic records from salt can be much longer than ice because salt can often survive 984 dissolution longer than most ice survives ablation or melting. Lowenstein and Demicco (2006) used 985 the co-precipitation of nahcolite (NaHCO3) and halite (NaCl) preserved in evaporites of the western 986 United States to infer that atmospheric CO2 was >1125 ppm (four times pre-industrial 34 987 concentrations) between ~56 and 49 Ma and to confirm that the highest prolonged global 988 temperatures of the past 65 million years were in the early Eocene. 989 In addition to paleoclimate studies, primary fluid inclusions in marine halites allow studies of 990 systematic changes in the chemistry of evaporated seawater going back to the late Precambrian 991 (~500 Ma; Lowenstein et al., 2001). These fluctuations are in phase with oscillations in seafloor 992 spreading rates, volcanism, global sea level, and the primary mineralogies of marine limestones and 993 evaporites. 994 995 996 7.1 Biological materials It has slowly become clear in the last decade or so that micro-organisms are present in virtually 997 all rock samples examined for biological properties. Anomalies in the concentration and isotope 998 ratio of biogenic trace gases (e.g., CH4, N2O) in ice (Abyzov, 1993, Priscu et al., 2005) indicate that 999 some of the micoorganisms (that include fungi, bacteria, and viruses: Castello and Rogers, 2005) in 1000 solid ice and subglacial water to depths >3.6 km are living and reproducing (Sowers, 2001, Campen 1001 et al., 2003). The difficulties in sampling such extreme life-forms in ice and salt provide useful 1002 practise for sampling potential equivalents in the hundreds of lakes beneath the Antarctic ice and 1003 on extraterrestrial bodies. The subaerial biota on ice usually resides in small melt cups on bare ice 1004 called cryocontite (Säwström et al., 2004). The dark bacterial communities in these cryoconite 1005 holes can reach of 6 area % of glaciers in Svalbard and, by lowering their albedo, increase their 1006 rate of melting and so become a negative component of their mass balance. 1007 Rather than increase with the age of the surrounding ice iin a core, the numbers of recoverable 1008 bacteria correlate with the climate during deposition of the initial snow. Cores from low-latitude 1009 glaciers contain more bacteria than polar ice core but the same genera and species occur in both 1010 (Priscu and Christner, 2004). The fact that particular strains can be revived and sub-cultured from 1011 ice everywhere points to them having adapted to the extreme conditions of deep ice. Priscu and 1012 Christner (2004) estimated that the polar ice sheets contain a total organic carbon pool almost 1013 equivalent to the bacterial carbon contained in all the freshwaters on the Earth s surface. The trace 1014 gasses that these microbes consume or liberate have to be taken into account when changes in the 1015 paleo-climate are read from paleo-gas records in ice cores (Campen et al. 2003, Sowers 2001). 35 1016 Conversely, layers within ice repositories dated accurately for paleoclimate studies provide 1017 important stratigraphic ages for studies of how these microbial populations have evolved. 1018 Rogers et al. (2004) speculated that the annual release of 10 17 21 to 10 viable microbes by 1019 melting ice up to 105 to 106 years old mixes the genotypes of ancient and modern microbial 1020 populations and therefore affect their mutation rates, fitness, survival, and pathogenecity. Smith et 1021 al. (2004) even speculated that releases of such ancient microbiota may explain cyclic calicivirus 1022 events and decades-long disappearances and reappearances of influenza-A subtypes. 1023 Organic components can also be trapped in brine inclusions within crystallising halite and their 1024 degradation can generate methane so that many salt mines face an explosion risk. Microbes 1025 trapped in crystallising salt go much further backs in time than the ~1 Ma record in ice. This was 1026 dramatically demonstrated by encysted halophytic bacteria carefully recovered from an undisturbed 1027 fluid inclusion in Permian salt (~ 250 Ma) preserving primary and diagenetic textures at the WIPP 1028 repository for isolating military nuclear waste in New Mexico (Vreeland et al, 2000). These bacteria 1029 divided after being released and “revived” with primitive sugar 1 that no longer exists in nature. 1030 Nowadays, before publication, these authors should have had their results replicated by an 1031 independent laboratory (where possible contaminants would be unlikely to be the same). Since 1032 then, Sattefield et al., (2005) confirmed that the brine inclusions from the same layer of salt that 1033 housed the revived bacillus are indeed ~250 Ma old. Gragg et al., (2006) reported that they are 1034 studying a 220 m long salt core from the Salar de Uyuni in Bolivia with its >276 ka record to 1035 acertain whether the preservation of viable microorganisms trapped in fluid inclusions relates to the 1036 age of the halite, or (as in ice) the environment of organism entrapment, or both. 1037 Fluid inclusions in salt can also lead to unexpected physical effects. Experiments to study the 1038 effects of thermal gradients imposed by isolating nuclear waste in engineered salt cavities found 1039 that, because the solubility of halite in water increases with temperature, 3-phase fluid inclusions 1040 can migrate up thermal gradients through the salt (by salt dissolving on the warm wall, diffusing 1041 through the brine and precipitating on the cool wall), vent the gas bubble at the engineered surface, 1042 and then migrate back into the rock mass (Roedder and Belkin, 1980). Similar phenomena are 1043 involved in the migration of solid particles and fluid inclusions and films to wam interfaces during 1044 thermal regelation in ice (Worster & Wettlaufer, 1999). 1045 36 1046 8 Ice, Salt and Water 1047 Water and its ice are different phases of the same chemistry so that changes in P and T can 1048 result in each spontaneously appearing in the other by melting or crystallisation. By contrast, salt 1049 and water have different chemistries so that, although increasing pressure increases the density of 1050 halite, there is no equivalent in salt of pressure melting regelation or basal lubrication by water, as 1051 in ice. 1052 1053 8.1 Solubility 1054 Apart from ice which is infinitely soluble, salt is one of the most soluble rocks and 100 g of pure 1055 water can dissolve 36 g of NaCl at 25°C (Langer & O ffermann, 1982) so that 1 cm of rain can 1056 dissolve 0.1667 cm of salt (Lockner, 1995) implying that the potential dissolution rate of salt can 1057 reach 17% of the annual precipitation. 1058 Talbot (1998) and Talbot and Aftabi (2004) anticipated that much less salt is usually dissolved 1059 than predicted by these values. This is because most salt extrusions are topographic highs so that 1060 most precipitation drains off them without carrying its potential load of dissolved salt, and as springs 1061 issuing from the base of Iranian namakiers are very rare. Although secondary salt deposits on and 1062 around them demonstrate that subaerial salt extrusions do lose salt via dissolution, most 1063 precipitation probably evaporates on their surfaces (Talbot, 1998). The first direct measurements of 1064 denudation rates of salt exposed in an arid climate (Frumkin, 1996) reinforced this impression. Thus 1065 99% of current denudation of Mount Sedom in Israel is by dissolution estimated at 0.5 to 0.75 mm 1066 a (which is 1 to 1.5 % of the annual rainfall of 50 mm/y). However most of that dissolution is 1067 restricted to a system of subsurface conduits only metres wide that entrench at rates between 4 1068 and 25 mm a-1. 1069 -1 In strong contrast, salt erosion rates measured on Cardona diapir in Spain were 20 mm per 100 1070 mm rainfall per year (Mottershead et al., 2005), greater than the theoretical potential. Similarly, 1071 Bruthans et al., (2007) used 26 plastic pegs implanted on a variety of slopes with different aspects 1072 on 3 salt extrusions in Iran from 2000-2005 and standardised their direct measurements of 1073 denudation rates of salt surfaces and their cap-soils to horizontal surfaces subject to long term 1074 mean precipitation rates. Long term denudation rates for exposed salt were 30-40 mm a-1 for two 1075 coastal diapirs (Namakdan and Hormoz island) where the rainfall averaged 103 mm a-1 for the 37 -1 1076 relevant 5 years (60% of the 30 year average) and 120 mm a for an inland diapir (Jahani) where 1077 the average rainfall was 295 mm a-1 for the same 5 years. Thin residual cap soils cut the 1078 denudation rate on coastal diapirs to 3.5 mm a , ~1/10 that for salt surfaces exposed nearby. 1079 Erosion rates for exposed salt were found to relate to rainfall directly and to the slope inversely, 1080 being lowest on vertical faces and highest on horizontal surfaces. Salt slopes steeper than 70º were 1081 estimated to retreat ~10 m/1000 years on coastal diapirs and ~30 m/1000 years on inland Jahani. 1082 Bruthans et al (2007) attributed the mean calculated values of denudation rate of exposed salt 1083 surfaces being between 29 and 39% higher than the maximum dissolution capacity of the 1084 corresponding rainfall to the cap soils retaining water like a sponge. However, halophytic 1085 microorganisms might also be involved (Castanier, et al., 1999). -1 th 1086 1087 1088 8.2 Strength The rheologies of both ice and salt depend strongly on their liquid water content. A liquid water 1089 content of ~1 volume % in ice (Murray et al., 2005) and of ~0.01 volume % in salt (Cristescu and 1090 Hunsche, 1998) increases the strain rate by ~400%. In glacier ice, this weakening is mainly due to 1091 free water at triple-grain contacts (e.g. Duval, 1977). Experiments on thin sections of natural salt 1092 filmed under the microscope have demonstrated that thumb pressure can distort initially isolated 1093 cubic brine inclusions through amoeboid shapes that connect into continuous films of brine along 1094 new, very mobile, grain boundaries (Urai et al., 1986). Such distortion of dispersed brine inclusions 1095 into continuous brine-filled grain boundaries affectively changes the stress exponent, n, of salt 1096 deforming as a power law fluid from the usual n=3 to 8 in the laboratory (Cristescu and Hunsche, 1097 1998) to an n=1 viscous fluid (Urai et al., 1986). 1098 Water is denser than ice but less dense than salt. As a result, water tends to migrate to different 1099 boundaries in flowing ice and salt. Water forms at, or migrates to, the bottom boundary of ice 1100 masses, either as gravitational flow through porous firn, or by drainage via structural weakness 1101 zones forming arcuate en- and subglacial networks (Fountain and Walder, 1998). Conversely, 1102 extruded salt is wet by rain that falls on the top surface before draining off or evaporating (Talbot, 1103 1998). Ice with surface temperatures well below 0ºC and without crevasses (as in the snouts of 1104 smaller and less dynamic polar or sub-polar glaciers) is usually regarded as impermeable, and thus 1105 an analogue of salt). 38 1106 Salt diapirs were being considered as repositories for radioactive waste in the 1960s and 70s 1107 (Gera, 1972) despite the salt mechanics of the time indicating that salt could not flow over the 1108 surface as in Iran (Heard, 1972). Gussow (1968) suggested that the Iranian salt extrusions were 1109 emplaced rapidly by lava-like, hot and temporary flow; a claim contested by Kent (1966) who 1110 argued that the local inhabitants often heard major rock falls on salt mountains that would not 1111 survive for long in the local rainfall if they were not actively supplied from depth. 1112 Assuming steady state flow and dissolution, Wenkert (1979) calculated the flow rates for 5 of 1113 the salt extrusions in the Zagros to be ~108 faster than allowed by contemporary salt mechanical 1114 measurements in the laboratory (Heard, 1972) and suggested that rainfall softened salt by allowing 1115 intragranular liquid diffusion. Fifty years after they were first suggested by Bailey (1931), the first 1116 direct measurements of salt movement were made in late 70s (Talbot and Rogers 1980) and 1117 confirmed that salt flow was still active. These first measurements found that markers on a 1118 namakier oscillated back and forth when the salt was dry but moved downslope almost 1 m in two 1119 days after a rain storm. Fibres of halite <1 mm thick and a few cm long are brittle when dry but 1120 some flex like a human hair when help under running water (Odé, 1968, p. 543). This, the Joffée 1121 effect, is attributed to water dissolving gasses poisoning the grain boundaries. A similar but 1122 macroscopic effect explains how a namakier that had been deforming noisily as an elastic solid for 1123 most of the preceding year was turned into a silent ductile fluid by only 20 minutes of light drizzle 1124 that precipitated 7 mm in the next 24 hrs (Talbot and Rogers, 1980). 1125 Inspired by these field studies (Urai et al .,1986 and Spiers et al .,1990) found that damp 1126 evaporites, halite, bischofite, MgCl2·6(H2O), and) carnallite KMgCl3·6(H2O), do indeed flow much 1127 faster at lower stresses by solution-transfer creep than the dislocation creep of dry evaporites. Later 1128 laboratory tests suggest that fine-grained rock salt flows about an order of magnitude faster than 1129 coarse-grained salt and that damp dilated salt flows about 40 times faster than dry confined salt 1130 (Cristescu and Hunsche, 1998). Even an increase in humidity >67% accelerates the creep of 1131 dilated salt by a factor of 12 (Cristescu and Hunsche, 1998). 1132 1133 9 Surging flows in ice and salt 1134 Both glaciers and namakiers can temporaily surge and flow hundreds of times faster than usual. 1135 Such surges can be local in time and space and we will treat ice streams as longer-lived spatial 39 1136 surges in ice sheets (e.g. Engelhardt, 2004)- like ocean currents flowing rapidly through the slow- 1137 flowing surrounding oceans or as rivers, fed by sluggish Darcian flow, in the surrounding ground 1138 water zones). 1139 Maximum values for strain rates in both compressed and extended ice are typically in the range 1140 of 0.1-0.2 a (=3 to 6.10 s ) for normal ice flow and >100 a (=3.10 s ) in surging flows (table 2 1141 in Hambrey & Larson 2000). By comparison, subsurface salt with coarse uniform grain-size flows at 1142 strain rates between ~3.10 to 3.10 a (=10 1143 sizes flows at velocities between cm to m a-1 at strain rates of 3.10-6 to 3.10-4 a-1 (10-13 to 10-11 s-1) 1144 over years before being recycled by dissolution. Namakiers surge for a few days at ~dm day at 1145 3.10-3 to 0.3 a-1 (=10-10 to 10-8 s-1) (Jackson and Talbot, 1986). -1 -9 -1 -8 -1 -6 -1 -15 to 10 -13 -6 -1 -1 s ) and extruded salt with bi-modal grain -1 1146 1147 9.1 surging flows in ice 1148 The ice activated in a surging glacier may result in the terminus advancing by km in a few 1149 months (Clarke, 1991; Kotlyakov et al., 1997). Most glacial surges last only 2-3 years but some on 1150 Svalbard last longer (12 years) and run-out further (4 km) more slowly, at ~ 2 m day (Lørnne, 1151 2006). Surging glaciers are more likely on weak sedimentary substrates which may explain why 1152 they cluster in particular areas (Clarke, 1991; Jiskoot et al., 1998). In longitudinal profiles, bulges of 1153 surging ice undergoing plug flow (± marginal and internal anastomosing shear zones) tend to fold 1154 against or override slower ice still frozen to the bed downstream along one or more thrusts (e.g. 1155 Lawson et al., 1994). This process results in stratifications above basal contacts with low frictional 1156 contact curling upward distally where friction increases along the base (cf. Fig. 9A4-6). In plan views, 1157 surges result in distinctive tear-drop or bulbous folds in moraines with axes parallel to valley sides 1158 that are clearest where glaciers converge (Hambrey and Lawson, 2000 p.72). Of particular note is 1159 that the plug-like flows of surging glaciers superpose cross-cutting longitudinal and transverse 1160 crevasses on the distinctive crevasse patterns associated with steady-state flows (Lawson, 1996). 1161 Furthermore, crevasse patterns in normal and surging glaciers are very different from each other 1162 and from crevasse patterns in both ice streams and ice sheets (Mayer and Herzfield, 2000). -1 1163 The generally accepted explanation of surges in glaciers (Meier and Post, 1969) is that the 1164 accumulation of the ice mass upstream reaches a critical normal pressure where above-normal ice 1165 pressures trigger a collapse of an arcuate-tunnel-based subglacial drainage system into a cavity- 40 1166 type drainage system. This smears the subglacial water along a larger area of the ice bed and 1167 decreases basal friction so that the surge occurs by enhanced basal sliding (Kamb, 1987; Raymond 1168 and Harrison, 1988; Eisen et al., 2001). Glaciologists also tend to contrast fast and slow ice flow in 1169 terms of water-lubricated basal ice and ice that is “still frozen” to the bed. The salt of namakiers is of 1170 course “frozen” but temperature does not play a significant role in whether it flows slowly or rapidly. 1171 Even if the transition from normal to surging ice flow does involve basal melting due to increased 1172 friction and increased buoyancy, Ramberg’s experiments on basal traction (Fig. 9A), suggest that 1173 the most relevant aspect of melting or adding water to the basal ice contact is that it decreases the 1174 basal friction. We therefore follow Fowler et al., (2001) and equate the basal sliding of surging ice 1175 (± the often associated processes of shearing or plastic yielding of lubricated underlying soft 1176 sediments ± shearing above the basal contact) as equivalent to what mathematician refer to as a 1177 free slip basal boundary. At its simplest, focussing on the bottom boundary effect suggests that 1178 glaciers surge when water reduces friction along their soles and ice streams appears where water 1179 reduces friction along their soles. However, Joughin’s (2008) study of the lower reaches of ice 1180 streams and outlet glaciers point to friction at the margins of a glacier as a control area accelerating 1181 ice flow. These findings agree with the implications of namakiers surging when water dampens their 1182 bulk mass and their upper levels rather than merely their basal contacts. 1183 More than 150 subglacial lakes beneath the ice of Antarctica have been found by ground 1184 penetrating, flown or satellite radar since the 1960s (Kohler, 2007; Siegert, 2007). Study of these 1185 has helped understanding of the link between smaller surging glacier systems and ice streams. The 1186 subglacial lakes are associated with smooth -floored depressions in the surface topography of the 1187 ice sheet above the lake due to acceleration of the ice associated with longitudinal stretching where 1188 ice detaches from its solid substrate. Surface melt is negligible in inland Antarctica so subglacial 1189 lakes there cannot be supplied by surface melt waters fracturing or draining their way through the 1190 ice as in Greenland (Sohn, 1998; van de Wal et al., 2008). Instead geothermal and frictional heat 1191 can bring large areas of basal ice to its melting point. 1192 Subglacial lakes were assumed to be stagnant until satellite radar from East Antarctica showed 1193 that the ice surface dropped suddenly while rising simultaneously hundreds of km away (Clarke, 1194 2006; Wingham et al., 2006). The surrounding ice appears to buttress most subglacial lakes 1195 (Kohler, 2007) but Belll et al., (2007) provided evidence for an all-important link between subglacial 41 1196 water and the dynamics of the overlying ice. Discharges comparable to the river Thames have been 1197 documented between subglacial lakes >300 km apart over 16 months (Bell et al., 2007). Repeat- 1198 track satellite altimetry by laser also demonstrated several lake drainage events as water 1199 pressurised to near the ice load was squeezed between different parts of the lower reaches of the 1200 West Antarctic ice streams (Fricker et al., 2007). One large lake near the grounding line drained ~2 1201 cubic km water over ~3 years. 1202 Bell et al., (2007) suggested that subglacial hydrology could be responsible for the initiation of 1203 an ice stream by changing the basal thermal regime of the ice sheet and/or by introducing water 1204 and reducing friction at the ice/bed interface. Thus lake water freezing onto the underside of the 1205 floating ice adds thermal energy and softens a zone of basal ice that has a lower frictional contact 1206 with downstream bedrock. Water spilling downstream of the lake also reduces basal friction and 1207 speeds ice flow. Processes like these likely account for sudden accelerations of ice streams in 1208 Greenland (Joughin et al., 2004, 2008; Luckman et al., 2006) while the rerouting of former 1209 subglacial water flows could account for the drastic slowing (“shutdown”) of the Kamb ice stream in 1210 Antarctica (Zwally et al., 2006). 1211 One problem with blaming the initiation of ice streams on subglacial lakes is that this approach 1212 neglects the thermodynamics behind the onset of streaming ice. However it develops, a pool of 1213 water beneath cold ice is insufficient on its own to initiate an ice stream. If the base of the ice is 1214 cold, the water from the subglacial lake will freeze onto the base, and prevent sliding, so there is no 1215 ice stream downstream of Lake Vostok (Bell et al., 2002; Pattyn et al., 2004). The possibility that 1216 frictional heating due to ice motion and high local geothermal heat flux could initiate ice streams has 1217 been discussed and modelled extensively (e.g. Clarke et al., 1977). Although the suggestion that a 1218 thermal run-away could lead to a surge of the Antarctic ice sheet (Yuen, et al., 1986) was later 1219 discounted, numerical modelling has demonstrated that strain heating could initiate streaming flow 1220 (Payne and Dongelmans, 1997) and satisfactorily explain modern (Payne, 1999) and past ice 1221 streams (Payne and Baldwin, 1999) as well as find the zone of initiation of fast glacial flow (Pohjola 1222 and Hedfors, 2003). The recent recognition of hydraulic connections between subglacial lakes 1223 revitalised this debate. In order for the lakes to be where they are, the geothermal heat flow must 1224 be higher than previously estimated, basal strain heat is better conserved than earlier believed, or 1225 heat conducts through the ice more slowly than expected. 42 1226 Bell (2008) distinguished how subglacial water affects the mass balances of the 3 largest 1227 current ice sheets. In the warmest, Greenland, (with a volume equivalent to 7.3 m sea-level rise) 1228 basal meltwater produced in regions of elevated heat flow is joined by surface meltwater that drains 1229 through the ice sheet on elevated crust and reaches the ocean through fjords that pierce the 1230 mountain rim. The West Antarctica ice sheet, similar in area to Greenland, has a volume equivalent 1231 to 5.5 m sea-level rise. Basal melt develops here beneath the interior ice with a base largely 1232 beneath sea level and the ice-stream tributaries (particularly in regions of elevated geothermal 1233 heatflow); water stored in subglacial lakes episodically drain into the ocean beneath the ice streams 1234 that are still largely buttressed by two large ice shelves. In East Antarctica, the largest current ice 1235 sheet, with a volume equivalent to ~65 m sea-level rise, large deep subglacial lakes in the interior 1236 are connected by intermittent subglacial flooding events that transmit water along the axes of ice 1237 streams and outlet glaciers. 1238 Monitoring large areas of the Greenlandic ice sheet found that surface meltwater reaching the 1239 basal contact almost doubled the velocity of the local ice sheet (Joughin et al., 2008). However, 1240 very little of this acceleration reached the relevant outlet glaciers, probably because of friction along 1241 their walls (Joughin et al., 2008). Water at the ice bed has been known to accelerate the overlying 1242 ice since the 1980s (e.g. Iken and Bindschadler, 1986). Pressurised water backing up behind 1243 overwhelmed subglacial plumbing systems lubricates the bottom boundary locally but can soon be 1244 dissipated by hydrodynamic changes in the subglacial sheets and channels. Recent work by 1245 Bartholomaus et al, 2008 adds to the knowledge of water decreasing the strength of the ice mass 1246 and friction along boundaries other than the bases as playing roles in the surging of flowing ice. 1247 1248 9.2 Surging flows in salt 1249 1250 Comparisons of Synthetic Aperture Radar interferograms with ground measurements of marker 1251 arrays dispersed over subaerial extrusions of salt of different ages and at different stages of 1252 evolution in Iran (Fielding and Talbot, unpublished) find that steady extrusion from the unseen vent 1253 continuously inflates the upper reaches of salt extrusions during dry weather. Such swellings of 1254 extruded salt are stored at high structural levels by elastic carapaces that are strong during dry 43 1255 years but are dissipated by accelerated gravity spreading when the carapaces are weakened by 1256 rain in wet years. 1257 An obvious working model for the surges of subaerial salt extrusions assumes that salt 1258 extrudes steadily from its nearly constant loading at depth and then gravity spreads downslope at 1259 rates that depend on the surface temperature but mainly on the local rainfall. Salt extruded steadily 1260 through the subsurface vent can be stored high in the mountain by its outer carapace or corset of 1261 broken elastic salt that case-hardens every salt mountain during the usual dry weather (Fig. 12A). 1262 When it is dry, the extruded salt flows at a few cm to m a-1 and generates a tank track fold as it 1263 advances and attaches to its high-friction basal contact. Flow folds deforming the upper limb of this 1264 fold probably record changes in the rate of gravity-spreading controlled by variations in rainfall over 1265 millennia. Constraint by the elastic corset is lost when the salt is saturated and weakened by 1266 rainstorms a few days per decade. This allowis the potential energy stored in the salt to be released 1267 by surges at dm day for a few days (Fig. 12B). 1268 -1 The closest water comes to migrating along the basal contact of a namakier is probably when 1269 unsaturated brines fed by storm-water dampen near-basal mylonitic shear zones <1 dm thick in the 1270 generally gneissose salt of bimodal grain size. These detach the overlying salt and curl upward 1271 distally allowing the overlying salt to shear over slower or static salt (Fig. 9, Talbot, 1998). Such 1272 halitic mylonites have fine grain sizes of <2 mm (Fig. 8SE) that indicate strain rates near 0.3 a-1 and 1273 micro-fabrics that imply that they deform by solution-precipitation creep and grain-boundary sliding 1274 (Schleder and Urai, 2006). Their sub-grain sizes indicate differential stresses of <0.4 MPa at 1275 temperatures <40ºC consistent with field observations (Talbot and Rogers, 1980). Storm-water 1276 draining through near-basal mylonitic detachment zones in namakiers that only surge for a day or 1277 two a decade take the place of the wet basal contacts that allow much faster and longer surges of 1278 glaciers and ice streams. However, the main control of surges of namakiers over their substrates of 1279 rocks and sediments appear to be weakening of the salt mass and its top boundary (Fig. 12). This 1280 working model for surging salt converges on recent suggestions that surges in the lower reaches of 1281 ice streams and outlet glaciers are controlled more by changes in strength of the ice mass and its 1282 outer boundaries. This argument suggests that salt extrusions in the damper mountains of Iran 1283 surge more than those along the drier coast. 44 1284 Individual salt sheets beneath the Gulf of Mexico have been calculated to have advanced at 1285 rates between 1.5 and 275 mm a-1 during Miocene and Pleistocene times (Wu et al., 1990). 1286 Submarine salt sheets therefore advance further and faster over saturated sediments than 1287 subaerial equivalents advance over dry substrates of rocks and sediments. Although it is not yet 1288 known whether submarine salt sheets surge, we consider this unlikley because their outer zones 1289 may undergo minor changes in strength during relatively small and slow temperature changes but, 1290 being surrounded by a perpetual bath of seawater, are unlikely to undergo the sudden drops in 1291 strength that allow surges in subaerial in desert climates. 1292 1293 10 CONCLUSIONS 1294 Despite differences in origin, chemistry, mineralogy, physical properties and supply processes 1295 etc, subaerial salt extrusions have visual and mechanical similarities to ice caps and glaciers (see 1296 Figs.1, 2, 8 and Table 2). Both ice and extruded salt can gravity-spread to the profiles of viscous 1297 fountains or droplets with profiles that may approximate steady-states. 1298 In bodies consisting of polygonal grains of ice with its hexagonal crystallography, primary 1299 stratification, diagenetic ice lenses and foliations picked out by discontinuous layers of different 1300 grain size and bubble-content are slightly strain active with a subhorizontal planar anisotropy. As a 1301 result, kinematic drape or forced folds (± axial planar stripes) form upright in ice above irregular 1302 bottom boundaries and probably have to be carried increasing distances downstream to be sheared 1303 to inclined or recumbent geometries. 1304 By contrast, in bodies of polygonal grains of cubic halite, primary stratification and secondary 1305 foliations are much more obvious in extrusive salt with its bimodal grain-size but are essentially 1306 strain passive. The grain shape fabrics in salt are clearer and more informative than grain-size 1307 foliations in ice and have longer strain memories. Nevertheless the strain memories of deformed 1308 salt fabrics are sufficiently short for foliations to map new, locally developed flow lines, most of 1309 which sub-parallel both the stratification and external boundaries. If this were true in flowing ice it 1310 would be difficult to distinguish secondary foliation from primary stratification (Hambrey and Muller, 1311 1978; Hooke and Hudleston, 1978). Passive markers like stratification and foliation in salt can 1312 remain parallel where they flow steadily round right-angled corners from source to diapir, and from 45 1313 diapir to extrusion; they can also drape over bedrock obstructions (Fig. 9G) in kinematic equivalents 1314 of upright folds in near-basal ice. 1315 The top free surface of a gravity-spreading fluid moves faster under air than where it is in 1316 frictional contact with a solid substrate (Talbot, 1998). As a result, the stratification rolls over the 1317 hinge of a kinematic tank track fold that grows behind every active namakier and inverts in its lower 1318 limb. A kinematic tank-track fold is inevitable in and behind the fronts of all viscous flows that have 1319 advanced over fictional substrates unless it is truncated by loss of material from their top surface. 1320 Their lack of recognition in glaciology may be because their upper limbs are truncated by ablation 1321 or because the stratification is so rarely exposed in cold ice flows. 1322 By comparison with laboratory experiments (Fig. 9) we have attributed the tank track folds that 1323 characterise the profiles of subaerial namakiers of Iran to their advance over high-frictional basal 1324 contacts. This would be confirmed if future studies find that tank track folds are absent from 1325 submarine salt extrusions that have advanced further and faster but essentially steadily over low 1326 friction basal contacts with weak saturated sediments. 1327 Both Ice and salt are supplied to significant surface bodies at rates that are close to steady over 1328 decades. By contrast, the rates at which ice (Shumskii 1964, p.. 326) and extruded salt gravity 1329 spread downslope are noticeably irregular on a large range of time scales. Accepting that the 1330 extrusion rates of individual salt diapirs are close to steady, flow lines marked by a foliation cross 1331 the (mainly) passive bedding and impose asymmetric dynamic flow folds on the upper limb of the 1332 kinematic tank-track fold when the ratio of their rates of extrusion and spreading changes. As 1333 rainfall is likely to change faster than the tectonic forces affecting subaerial salt extrusions, such 1334 budget-induced dynamic flow folds within subaerial salt extrusions offer a potential record of climate 1335 change over tens of millennia. Any flow folds in submarine salt sheets might record long term 1336 temperature changes in the surrounding seawater. Indivivual flow folds in the upper limbs of tank 1337 track folds also offer natural analogue models of individual nappes within piles of nappes of 1338 extruded infrastructures in orogens like the Himalaya. They suggest that individual nappes within 1339 orogenic nappe-piles might reflect changes in the ratios of their rates of tectonic extrusion relative 1340 to their rates of gravity spreading. Whether orogenic nappes spread ar rates affected by the 1341 climates is debateable. A great advantage of studying rock deformation structures and fabrics 1342 within bodies of ice or salt is that internal structures, such as folds and shear zones of different 46 1343 categories, could be instrumented and monitored and treated monominerallic analogues of other 1344 metamorphic rocks. 1345 Hudleston (1976) envisaged flowing ice bodies developing folds where and when the flow lines 1346 crosses old stratification perturbed by relief in the bottom boundary. However, the clearer folds in 1347 namakiers suggest that dynamic flow folds near the top boundary of flowing ice may be even 1348 clearer records of budget changes due to fluctuations in climate. The potential for folds in ice to 1349 record climate change has been recognised in ice streams (see e.g. Bindschadler, 1993; Jacobel et 1350 al., 1993; Merry and Whillans, 1993; and Fahnestock, et al., 2000). 1351 Fluctuations in supply and loss (in ice) and gravity spreading and loss (in salt) result in budget- 1352 induced dynamic flow folds in both glaciers and namakiers. However, while both glaciers and 1353 namakier undergo slow penetrative creeping flow that depend mainly on their thickness, slope, 1354 grain size, purity and temperature, both may surge at velocities hundreds of times faster than usual 1355 when water weakens these rock masses and/or sufficient areas of their external boundaries. The 1356 factors controlling surges in namakiers are more obvious than those controlling surges in ice. The 1357 possibility that surging ice is influenced by water weakening the bulk ice and its outer boundaries 1358 (as well as its base) has been recognised only recently. 1359 As for all exposed rock masses, the outer zones of bodies of ice and salt salt readily dilate to 1360 brittle elastic carapaces aided by destressing and weathering or ablation. The resulting fractures 1361 and micro-fractures render the outer zones of both ice and salt masses sensitive to weather- 1362 induced changes. That this applies to ice tends to be overlooked because variations in wet and dry 1363 basal conditions are more obvious. Ice has long been recognised as capable of detaching and 1364 sliding over wet substrates because most water associated with ice forms or migrates to the bottom 1365 boundary. Although salt can detach along damp near-basal mylonite zones and slide over the 1366 underlying salt, the water that has most effect on the downslope flow rate of namakiers falls as rain 1367 on their top surfaces and seldom reaches the bottom boundary of denser salt. 1368 Detailing the records of changes in past climates from fluid inclusions in ice go back ~1 Ma whereas 1369 salt deposits potentially offer a discontinuous record of changes in climate and oceanic chemistry 1370 back to at least ~500 Ma; on Mars and some of the icy moons of the outer planets, equivalent 1371 records may extend to billions of years in both ice and/or salt. Microbiota up to ~1 Ma may be 1372 released by old ice ablating or melting, and some of the ~500 Ma microbiota released by salt 47 1373 dissolution may yet turn out to be viable although the foodstocks they originally fed on are no longer 1374 freely available on the modern Earth. The implications of mixing such ancient and modern genome 1375 pools of cryophile and halophiltic micro-biota bears continued examination. 1376 1377 Acknowledgments 1378 Various versions of this manuscript benefited considerably from constructive reviews by Ian 1379 Davison, J.K. Warren, Peter Hudleston, Michael Hudec, and Martin Jackson; any mistakes are ours 1380 rather than theirs. Both authors thank innumerable colleagues who have helped them in the field 1381 throughout their careers. Most of their travels to and from the field have been funded by The 1382 Swedish Natural Science Foundation (now VR). CT would also takes this opportunity to thank Dr 1383 Korehei, former General Director of the Geological Survey of Iran and his colleagues, for their 1384 enthusiastic encouragement and considerable logistic support for the 1993-2006 project on the 1385 kinematics of Iranian salt. 1386 1387 Talbot and Pohjola References 1388 Abyzov, S. S., 1993. Microorganisms in the Antarctic ice, In: Friedmann, E. I. (Ed.), Antarctic 1389 Microbiology John Wiley & Sons, Inc., New York., pp. 265-295. 1390 Alkattan, M., Oelkers, E.H., Dandruran J-L., Schott, J., 1997. Experimental studies of saltrock 1391 dissolution kinematics, 1: the effects of saturation state and the presence of trace metals. 1392 Chemical Geology 137. 201-219. 1393 1394 1395 Alley, R.B., 2002. The Two-Mile Time Machine: Ice Cores, Abrupt Climate Change, and Our Future. Princeton University Press, Princeton, New Jersey, USA, 240 pp. Alley, R.B., Gow, A.J., Meese, D.A., Fitzpatrick, J. J., Wadddington, E. D., Blozan, J. F., 1997. 1396 Grain scale processes, folding and stratigraphic disturbances in the GISP2 ice core. Journal of 1397 Geophysical Research 102, C12, 26,819-26,830. 1398 1399 1400 Ambraseys, N. N., Jackson, J. A., 1998. Faulting associated with historical and recent earthquakes in the eastern Mediterranean region. Geophysical Journal International 133, 390-406. Bailey, E.B., 1931.Salt plugs. Geological Magazine 68, 335-339. 48 1401 Baker, P.A., Rigsby, C.A., Seltzer, G.O., Fritz, S.C., Lowenstein, T.K., Bacher, N.P., Veliz, C., 2001. 1402 Tropical climate changes at millennial and orbital timescales on the Bolivian Altiplano. Nature 1403 409, 698-701. 1404 1405 1406 1407 Barton, D. C., 1933. Mechanics of formation of salt domes with special reference to Gulf Coast salt domes of Texas and Louisiana. AAPG Bulletin 17, 1025-1083. Bartholomaus, T. C., Anderson, R.S., Anderson, S.P., 2008. Response of glacier basal motion to transient water storage. Nature geoscience 1, 33-37. 1408 Beaumont, C, Jamieson, R. A., Nguyen, M. H., Lee, B., 2001. Himalayan tectonics explained by 1409 extrusion of a low-viscosity crustal channel coupled to focussed surface denudation. Nature 1410 414, 738-742. 1411 1412 Bell, R. E., 2008. The role of subglacial water in ice-sheet mass balance. Nature geoscience 1, 297– 304. 1413 Bell, R.E., Studinger, M., Tikku, A.A., Clarke G.K.C., Gutner, M.M., Meertens. C., 2002. Origin and 1414 fate of Lake Vostok water frozen to the base of the East Antarctic ice sheet. Nature 416, 307- 1415 310. 1416 1417 1418 Bell, R. E., Studinger, M., Suman, C. A., Fahnestock, M. A., Joughin, I., 2007. Large subglacial lakes in East Antarctica at the onset of fast flowing streams. Nature, 445, 904-907. Bindschadler, R., 1993. Siple Coast Project research of Crary Ice Rise and the mouths of Ice 1419 Streams B and C, West Antarctica: review and new perspectives. J. Glaciol., 39, 538-552. 1420 Bindschadler, R. 1998. Monitoring ice sheet behavior from space. Reviews of Geophysics 36, 79- 1421 1422 104. Bobst , A.L., Lowenstein T.K., Jordan T.E., Godfrey, L.V., Ku T.L., Luo, S., 2001. A 106ka 1423 paleoclimate record from drill core of the Salar de Atacama, northern Chile .Palaeogeography, 1424 Palaeoclimatology, Palaeoecology 173, 21-42. 1425 1426 Boulton, G. S., 1996. Theory of glacial erosion, transport and deposition as a consequence of subglacial sediment deformation. J. Glaciol., 42, 43-62. 1427 Bruthans, J., Filippi, M., Gersl, M., Zare, M., Melkova, J., Pazdur, A., Bosak, P., 2006. Holocene 1428 marine terraces on two salt diapirs in Persian Gulf (Iran): age, depositional history and uplift 1429 rates. Journal of Quaternary Science 2, 843-857. 49 1430 1431 1432 Bruthans, J., Asadi, N., Filippi, M., Wilhelm, Z., Zare, M., 2007. A study of erosion rates on salt diapir surfaces in the Zagros Mountains, SE Iran. Environmental Geology 53, 1079-1089. Busk, H. G., 1929, Earth flexures; their geometry and their representation and analysis in geological 1433 section with special reference to the problem of oil finding: Cambridge University Press, 1434 Cambridge, U.K., 106 pp. 1435 1436 Campen, K.R., Sowers, T., Alley, R.B., 2003. Evidence for microbial consortia metabolizing within a low latitude mountain glacier. Geology 31, 231–234. 1437 Castanier, S., Perthuisot, J-P., Matrat. M., Morvan, J-Y., 1999. The salt ooids of Berre salt works 1438 (Bouches du Rhône, France): the role of bacteria in salt crystallisation, Sedimentary Geology 1439 125, 9-21. 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 Castello, J.D., Rogers, S.O., 2005. Life in Ancient Ice. Princeton University Press. Princeton, New Jersey, USA. 328 pp. Chemia Z., Koyi H., Schmeling H., 2008. Numerical modelling of rise and fall of a dense layer in salt diapirs. Geophysical Journal International, 172, 798-816. Chinn, T. J. H., 1989. Single folds at the margins of dry-based glaciers as indicators of glacial advance. Annals Glaciology 12, 23-30. Clarke, G.K.C., 1987. A short history of scientific investigations on glaciers. J. Glaciol., Special Issue, 4-24. Clarke, G.K.C., 1991. Length, width and slope influences on glacier surging. J. Glaciol., 37, 236246. 1450 Clarke, G.K.C., 2006. Glaciology: Ice-sheet plumbing in Antarctica. Nature 440, 1000-1001. 1451 Clarke, G.K.C., Nitsan, U., Paterson, W.S.B., 1977. Strain heating and creep instability in glaciers 1452 1453 and ice sheets. Reviews of Geophysics and Space Physics 15, 235-247. Cobbold, P.R., Szatmari, P., Demercian, L.S., Celho, D., Rossello, E.A.., 1995. Seismic and 1454 experimental evidence for thin-skinned horizontal shortening by convergent gliding on 1455 evaporites, deep-water Santos Basin, Brazil. In: Jackson, M.P.A., Roberts, D.G., Snelson, S., 1456 (Eds.) Salt tectonics; a global perspective; AAPG Memoir 65, 305-321. 1457 Cosgrove, J.W., Ameen, M.S., 2000. A comparison of the geometry, spatial organisation and 1458 fracture patterns associated with forced folds and buckle folds. In: Cosgrove, J. W., Ameen, M. 1459 S., (Eds.) Forced folds & fractures: Geological Society London Special Publications 169, 7-21. 50 1460 Coward, M., Stewart, S., 1995. Salt-influenced structures in the Mesozoic–Tertiary cover of the 1461 southern North Sea, U.K. In: Jackson, M.P.A., Roberts, D.G., Snelson, S. (Eds.) Salt Tectonics: 1462 a Global Perspective. AAPG Memoir 65, 229–250. 1463 1464 1465 1466 1467 1468 1469 Cristescu, N., Hunsche, U. 1998. Time effects in Rock mechanics- Series: Materials, modelling and Computation. John Wiley and Sons, Chichester. 342 pp. De Böckh, H., Lees, G.M., Richardson, F.D.S., 1929. Contribution to the stratigraphy and tectonics to the Iranian ranges. In Gregory, J.W. (Ed) The structure of Asia. Methuen, London, 58-176. Demorest, M., 1942. Glacier regimens and ice movements within glaciers. American Journal Science 240, 31-66. Diegel, F.A., Karlo, J.F., Scuster, D.C., Shoup, R.C., Tauvers, P.R., 1995. Cenozoic structural 1470 evolution and tectonostratigraphic framework of the Northern Gulf Coast continental margin. In 1471 Jackson, M.P:A., Roberts, D.G., Snelson, S., (Eds.) Salt Tectonics: a global perspective, AAPG 1472 Memoir 65, 109-151. 1473 Drewry, D., 1986. Glacial geological processes. Edward Arnold, London, 276 pp. 1474 Durham, W. B., Abey, A.E., Trimmer, D.A., 1981. Thermal Conductivity and diffusivity and 1475 expansion of Avery Island salt at pressure and temperature. Technical Report University 1476 California, Lawrence Berkeley Laboratory, Livermore, and CA. USA. 1477 1478 1479 1480 1481 Duval, P., 1977. The role of water content on the creep rate of polycrystalline ice. IUGG General Assembly of Grenoble, 1975. IAHS Publication 118, 29-33. Duval, B., Cramez, C., Jackson M.P.A., 1992. Raft tectonics in the Kwanza Basin, Angola. Marine and Petroleum Geology 9, 389-404. Dyson, D.A., 2004, Christmas tree diapirs and the development of hydrocarbon reservoirs; a model 1482 from the Adelaide Geosyncline, South Australia. In Post, P. et al., (Eds.) Salt sediment 1483 interactions and hydrocarbon prospectivity: concepts, applications and case studies for the 21st 1484 century: 24th Annual Gulf Coast Section SEPM Foundation Bob F. Perkins Research 1485 Conference, Houston Texas pp. 79-96. 1486 Eisen, O., Harrison, W.D.M., Harrison, C.F., 2001. The surges of Variegated Glacier, Alaska, 1487 U.S.A., and their connection to climate and mass balance. J. Glaciol., 47, 351-358. 1488 Engelhardt, H. 2004. Thermal regime and dynamics of the West Antarctic ice sheet. Annals 1489 Glaciology 39, 85-92. 51 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 EPICA community members (56 authors). 2004. Eight glacial cycles from an Antarctic ice core. Nature 429, 623-629. Fahnestock, M.A., Scambos, T.A., Bindschadler R.A., Kvaran. G., 2000. A millennium of variable ice flow recorded by the Ross Ice Shelf, Antarctica. J. Glaciol., 46, 652-664. Fricker, H. A., Scambos, T., Bindschadler, R., Padman, L., 2007. An active subglacial water system in West Antarctica mapped from space. Science 315, 1544-1548. Fountain, A.G., Walder, J.S., 1998. Water flow through temperate glaciers. Reviews in Geophysics 36, 299–328. Fowler, A.C., Murray, T., Ng, F.S.L. 2001. Thermally controlled glacier surging. J. Glaciol., 47, 527538. Frumkin, A., 1996. Determining the exposure age of a karst landscape. Quaternary Research 46, 99-106. Gemmer, L., Ings, S. J., Medvedev, S., Beaumont, C., 2004. Salt tectonics driven by differential 1503 sediment loading: Stability analysis and finite-element experiments. Basin Research 16, 199- 1504 218. 1505 1506 1507 1508 1509 Gera, F. 1972. Review of salt tectonics in relation to the disposal of radioactive wastes in salt formations. Bulletin of the Geological Society of America 83, 3551-3574. Gevantman, L. H. (Ed.), 1981, Physical Properties Data for Rock Salt; US Department of Commerce NBS Monograph 167, p. 214. Gragg, K, E.. Lowenstein, T. K., Timofeeff, M. N., Schubert, B. A., Parker, M., 2006, Long-term 1510 survival of microorganisms preserved in halites, Salar De Uyuni salt core, Bolivia. Abstracts of 1511 Philadelphia Annual Meeting (22–25 October 2006). 1512 1513 1514 1515 1516 1517 1518 1519 Gudmundsson, G.H., 1997a. Basal flow characteristics of a linear medium sliding frictionless over small bedrock undulations. J. Glaciol .43, 71-79. Gudmundsson, G.H., 1997b. Basal flow characteristics of a non-linear flow sliding frictionless over strongly undulating bedrock. J. Glaciol 43, 80-89. Gussow, W.C. 1966. Salt temperature: a fundamental factor in salt dome intrusion. Nature 210, 518-519. Halbouty, M.T., 1979. Salt domes, Gulf region, United States and Mexico (2nd edition). Gulf Publishing Company, Houston, Texas, 561 pp. 52 1520 1521 1522 1523 1524 Hambrey, M.J., Muller, F., 1978. Structures and ice deformation in the White Glacier. J. Glaciol. 20, 41-66. Hambrey, M.J., Bennett, M.R., Dowdeswell, J.A., Glasser, N.F., Huddart, D., 1999. Debris entrainment and transfer in polythermal valley glaciers. J. Glaciol., 45, 69-86. Hambrey M. J., Lawson, W., 2000. Structural styles and deformation fields in glaciers: a review. In: 1525 Maltman A. J, Hubbard, B., Hambrey, M. (Eds.) Deformation of Glacial materials. Geological 1526 Society of London, Special publication 176, 59-83. 1527 1528 1529 1530 1531 1532 1533 1534 1535 Hardie, L. A., 1991. On the significance of evaporites, Annual Reviews of Earth and Planetary Science 19, 131-168. Harper, J.T, Humphrey, N.F., Pfeffer, W.T., 1998. Three-dimensional deformation measured in an Alaskan glacier. Science 281, 1340-1342. Harrison, J.V., 1930. The geology of some salt-plugs in Laristan, southern Persia; Geological Society of London Quarterly Journal 86, 463-522. Heard, H.C, 1972. Steady-state flow of polycrystalline halite at pressure of 2 kilobars, American Geophysical Union Geophysical Monograph 16, 191-209. Heaton, R.C., Jackson, M.P.A., Bamahmoud, M., Nani, A.S.O., 1995. Superposed Neogene 1536 extension, contraction and salt canopy emplacement in the Yemini Red Sea. In: Jackson, 1537 M.P.A., AAPG Memoir 65, 333-367. 1538 Hooke, R. LeB., Hudleston, P., 1978. Origin of foliation in glaciers. J. Glaciol., 20, 285-299. 1539 Hooke, R. LeB., Hanson. B., 1986. Borehole deformation experiments, Barnes Ice Cap, Canada. 1540 1541 1542 1543 Cold Regions Science and Technology 12, 261-276. Hooke, R.LeB., Holmlund, P., Iverson. N. R., 1987. Extrusion flow demonstrated by borehole deformation measurements over a riegel, Storglaciären, Sweden. J. Glaciol. 33, 72-78. Hovland, M., Fichler, C., Rueslåtten, H., Johnsen, H.K., 2006. Deep rooted piercement structures in 1544 deep sedimentary basins- manifestations of supercritical water generation at depth. Journal of 1545 Geochemical Exploration 89, 157–160. 1546 Hovland, M., Kuznetsova, T., Rueslåtten, H., Kvamme, B., Johnsen, H.K., Fladmark, G.E., Hebach, 1547 A., 2006. Subsurface precipitation of salts in supercritical seawater. Basin Research 18, 221– 1548 230. 53 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 Hudec, M.R., Jackson, M.P.A. 2006, Advance of allochthonous salt sheets in passive margins and orogens. AAPG Bulletin 90, 1535-1564. Hudec, M.R., Jackson, M.P.A. 2007. Terra Infirma: understanding salt tectonics. Earth Science Reviews 82, 1-28. Hudleston, P., 1976. Recumbent folding in the base of the Barnes Ice Cap, Baffin Island, Northwest Territories, Canada. Geological Society of America Bulletin 87, 1684-1692. Hudlestone, P., Hooke, R.LeB., 1980. Cumulative deformation in the Barnes Ice Cap and implications for the development of foliation. Tectonophysics 66, 127-146. Huppert, H. E., 1982. The propagation of two-dimensional and axisymmetric viscous gravity currents over a rigid horizontal surface. Journal of Fluid Mechanics 121, 43-58. 1559 Iken, A., Bindschadler, R.A., 1986. Combined measurements of subglacial water pressure and 1560 surface velocity of Findelengletscher, Switzerland: conclusions about drainage system and 1561 sliding mechanism. J. Glaciol., 32, 101-119. 1562 Iizuka, Y., Satake, H., Shiraiwa, T., Narus R., 2001. Formation processes of basal ice at Hamna 1563 Glacier, Sôya Coast, E. Antarctica, inferred by detailed co-isotopic analyses. J. Glaciol., 47, 1564 223-231. 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 Jackson, M.P.A., 1995. Retrospective salt tectonics, In: Jackson, M.P.A., Roberts, D.G., Snelson, S., (Eds.) Salt tectonics; a global perspective; AAPG Memoir 65, 1-28. Jackson, M.P.A., Talbot, C. J., 1986. External shapes, strain rates and dynamics of salt structures: Geological Society of America Bulletin 97, 305-328. Jackson, M.P.A., Cornelius, R.R., Craig, C.R., Gansser, A., Stöcklin, J., Talbot C.J., 1990. Salt diapirs of the Great Kavir, central Iran. Geological Society of America Memoir 177. Jackson, M.P.A., Talbot, C.J., 1994. Advances in salt tectonics. In: Hancock, P.L, (Ed) Continental deformation: Pergamon Press and IUGS, Oxford, UK, 159-179 pp. Jackson, M.P.A., Vendeville, B.C, Schultz-Ela, D.D., 1994. Structural dynamics of salt systems. Annual Reviews of Earth and Planetary Sciences 22, 93-117. 1575 Jackson, M.P.A., Warin, O.N., Woad, G.M., Hudec, M.R., 2003. Neoproterozoic allochthonous salt 1576 tectonics during the Lufilian orogeny in the Katangan Copperbelt, central Africa: Geological 1577 Society of America Bulletin 115, 314–330. 54 1578 Jackson, M.P.A., Hudec, M.R., 2004. A new mechanism for advance of allochthonous salt sheets, 1579 in Post, P.J., Olson, D.L., Lyons, K.T., Palmes, S.L., Harrison, P.F., Rosen, N.C., (Eds.) Salt- 1580 sediment interactions and hydrocarbon prospectivity: concepts, applications, and case studies 1581 for the 21st century: 24th Annual GCSSEPM Foundation Bob F. Perkins Research Conference, 1582 pp. 220-242. 1583 Jacobel, R.W., Gades, A. M., Gottschling, D.L., Hodge, S.M., Wright, D. L., 1993. Interpretation of 1584 radar-detected internal layer folding in West Antarctica ice steams. J. Glaciol. 39, 528-537. 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 Jacobson, H.P. Waddington, E.D., 2005. Recumbent folding of divide arches in response to unsteady ice-divide migration. J. Glaciol., 51, 201-209. Jiskoot, H., Boyle, P., Murray, T., 1998. The incidence of glacier surging in Svalbard: evidence from multivariate statistics. Computers and Geosciences 24, 387-399. Joughin, I. Abdalati, W., Fahnestock, M., 2004, Large fluctuations in speed on Greenland's Jakobshavn Isbræ glacier. Nature 432, 608-610. Joughin, I., Das, S. B., King, M. A., Smith, B. E., Howat, I. M., Moon, T., 2008. Seasonal Speedup along the Western Flank of the Greenland Ice Sheet. Science 320, 781 – 783. Kamb W. B., 1987.Glacier surge mechanism based on linked cavity configuration of the basal water conduit system. Journal Geophysical Research 92 (B9): pp. 9083-9100. Kawamoto, E., Shimamoto, T., 1997. Mechanical behavior of halite and calcite shear zones from 1596 brittle to fully plastic deformation and a revised fault model. In Yadong, Z., Davis, G.A., Yin, A 1597 (Eds) Structural Geology and Geomechanics, Proceedings of the 30th International Geological 1598 Congress, Beijing, China, 4-14 August 1996, Volume 14, pp 89-105. 1599 1600 Kehle, R.O., 1988. The origin of salt structures. In: Schreiber, B.C. (Ed.), Evaporites and Hydrocarbons. Columbia University Press, New York, pp. 345–404. 1601 Kent, P.E., 1958. Recent studies of South Persian salt plugs. AAPG Bulletin 42, 2951-2979. 1602 Kent, P.E., 1966. Salt temperature: a fundamental factor in salt dome intrusion: discussion. Nature 1603 1604 1605 1606 1607 211, 1387-1388. Kent, P.E., 1970. The salt plugs of the Persian Gulf region. Leicester Literary and Philosophical Society Transactions 64, 56-88. Kent, P.E., 1979. The emergent Hormuz salt plugs of southern Iran. Journal of Petroleum Geology 2, 117-144. 55 1608 Kockel, F., 1998. Salt problems in Northwest Germany and the German North Sea sector. In 1609 Kockel, F., Marschall, R. (Eds.) Geology and Geophysics of Salt Structures. Journal of Seismic 1610 Exploration 7, 219–235. 1611 Kohler, J., 2007. Glaciology: lubricating lakes. Nature 445, 830-831. 1612 Kotlyakov, V.M., Osipova., G.B., Tsvetkov D.G., 1997. Fluctuations of unstable mountain glaciers: 1613 scale and character. Annals Glaciology 24, 338-343. 1614 Koyi, H., 1991. Mushroom diapirs penetrating into high viscous overburden. Geology 19, 1229-123. 1615 Langer, H., Offermann, H., 1982. On the solubility of sodium chloride in water. Journal of Crystal 1616 1617 1618 1619 1620 1621 1622 Growth 60, 389-392. Langway, C.C., 1967. Stratigraphic analysis of a deep ice core from Greenland. Research Report Cold Regions Research and Engineering Laboratory 77, 130 pp. Lawson, W., 1996. The relative strengths of debris-laden basal ice and clean glacier ice: some evidence from Taylor Glacier, Antarctica. Annals Glaciology 23, 270-276. Lawson, W.J., Sharp, M.J., Hambrey, M.J., 1994. The structural geology of a surge-type glacier. Journal of Structural Geology 16, 1447-1462. 1623 Lees, G.M., 1927, Salzgletcher in Persien: Mitt der deutschen Gesellschaft, Wien 20, 29– 34. 1624 Legrand, M., Mayewski, P., 1997. Glaciochemistry of polar ice cores: a review. Reviews of 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 Geophysics 35, 219-243. Lister, J.R., Kerr, R.C, 1989. The propagation of two-dimensional and axisymmetric viscous gravity currents at a fluid interface. Journal of Fluid Mechanics 203, 215-249. Lockner, D.A., 1995. Rock failure. In Ahrens, T.J. (Ed). Rock Physics and Phase Relations. A Handbook of physical constants, AGU Reference Shelf 3, AGU, 127-147. Lowenstein, T.K., Hardie, L.A., 1985. Criteria for the recognition of salt-pan evaporites. Sedimentology 32, 627-644. Lowenstein, T.K., Demicco, R.V. 2006. Elevated Eocene atmospheric CO2 and its subsequent decline. Science 313. 1928. Lowenstein, T.K., Li, J., Brown, C., Roberts, S.M., Ku, T-L., Luo, S., Yang, W., 1999. 200 ky paleoclimate record from Death Valley salt core. Geology 27, 3-6. Lowenstein, T.K., Timofeeff, M. N. Brennan, S.T., Hardie, L.A., Demicco, R.V., 2001. Oscillations in Phanerozoic Seawater Chemistry: Evidence from Fluid Inclusions. Science 294, 1086 – 1088. 56 1638 1639 1640 1641 1642 Lørnne, I., 2006. Low-velocity glacial surges- processes unlocked by modern surge on Svalbard. Geology 34, 663-556. Lucchitta, B.K., 2001. Antarctic ice streams and outflow deposits on Mars. Geophysical Research Letters 28, 403-406. Luckman, A., Murray, T., de Lange, R., Hanna, E., 2006. Rapid and synchronous ice-dynamic 1643 changes in East Greenland, Geophysical Research Letters 33, L03503, 1644 doi:10.1029/2005GL025428. 1645 Lunine, J.I. Atreya, S.K., 2008. The methane cycle on Titan. Nature Geoscience 1, 159-164 1646 Mair, R., Kuhn M., 1994. Temperature and movement measurements at a bergschrund. J. Glaciol., 1647 1648 1649 1650 40, 561-565. Maltman A.J, Hubbard, B., Hambrey, M. 2000. Deformation of glacial materials. Geological Society of London Special Publication 176, 1-9. Mayer, H., Herzfeld, U.C., 2000. Structural glaciology of the fast-moving Jakobshavn Isbræ, 1651 Greenland, compared to the surging Bering Glacier, Alaska, U.S.A. Annals Glaciology 30, 243- 1652 249. 1653 1654 1655 1656 1657 Meier, M.E., Post, A., 1969. What are glacial surges? Canadian Journal of Earth Sciences 6, 807817. Merry, C. J., Whillans. I.M., 1993. Ice-flow features on Ice Stream B, Antarctica, revealed by SPOT HRV imagery. J. Glaciol., 39, 515-527. Mohr, M., Warren, J.K., Kukla, P.A., Urai, J.L., Irmen, A., 2007.Subsurface record of salt glaciers in 1658 an extensional intracontinental setting (Late Triassic of northwestern Germany). Geology 35, 1659 963-966. 1660 Mottershead, D.N., Duane, W., Inkpen, R.J. Wright, J.S., 2005. Subaerial karstic erosionof small- 1661 scale saltrock terranes. In Gutierrez, F., Guitiierrez, M., Desir, G. Guerro, J. Lucha, P. Marin, 1662 C., GuarciaRuiz, J. M., (Eds.) 6th International conference on Geomorphology Zaragosa, 1663 Cardona, Spain. Abstract volume, p. 453. 1664 Murray, T., Rippin. D, Barrett, B., Kilner, M., 2005. Thermal regime and ice-water content of a 1665 glacier in the European Alps using borehole and surface radar. Eos, Trans. AGU, 86(52), Fall 1666 Meet. Suppl., Abstract C51B-0280. 57 1667 Nelson, T.H., Fairchild, L., 1989. Emplacement and evolution of salt sills in the northern Gulf of 1668 Mexico (abs.). Houston Geological Society Bulletin 32, 6–7. 1669 Nye, J. F., 1952. The mechanics of glacial flow. J. Glaciol., 2, 82-93. 1670 Nye, J.F. 1965. The flow of a glacier in a channel of rectangular, elliptic or parabolic cross-section. 1671 1672 1673 1674 J. Glaciol., 5, 661-690. Odé, H., 1963, Mechanical properties of salt. Geological Society of America Special Paper 88, 120. Pappalardo, R.T., Barr, A.C., 2004. The origin of domes on Europa: the role of thermally induced 1675 compositional diapirism. Geophysical Research Letters 31, L01701, 1676 doi:10.1029/2003GL019202. 1677 Pappalardo, R.T., Head, J.W., Greeley, R., Sullivan, R.J., Pilcher, C., Schubert, G., Moore, W.B., 1678 Carr M.H., Moore, J.M., Belton, M.J., Goldsby, D.L., 1998. Geological evidence for solid-state 1679 convection in Europa's ice shell. Nature 391, 365-368. 1680 Paterson, W.S.B., 1994. Physics of glaciers. Third edition, Pergamon Press, Oxford, 390 pp. 1681 Pattyn, F., De Smedt, B., Souchez. R., 2004. Influence of subglacial Vostok lake on the regional 1682 1683 1684 1685 1686 1687 1688 ice dynamics of the Antarctic ice sheet: a model study. J. Glaciol., 50, 583-589. Payne, A.J. 1999. A thermomechanical model of ice flow in West Antarctica. Clim.Dyn., 15, 115125. Payne, A.J., Dongelmans, P.W., 1997. Self-organization in the thermomechanical flow of ice sheets. Journal of Geophysical Research 102(B6), 12219-12233. Payne, A.J. Baldwin, D.J., 1999. Thermomechanical modelling of the Scandinavian ice sheet: implications for ice-stream formation. Annals Glaciology 28, 83-89. 1689 Pazzaglia, F.J., Knueper, P.L.K., (Eds.), 2001. The steady-state orogen: concepts, field 1690 observations, and models; American Journal of Science 302, Issues Nos. 4 & 5. 1691 1692 1693 Peale, S.J., 2003. Tidally induced volcanism. Celestial Mechanics and Dynamical Astronomy 87, 129–155. Peel, F.J., Travis, C.J., Hossack, J.R., 1995. Genetic structural provinces and salt tectonics of the 1694 Cenozoic offshore U.S. Gulf or Mexico: a preliminary analysis. In: Jackson, M.P.A., Roberts, 1695 D.G., Snelson, S., (Eds.) Salt tectonics; a global perspective; AAPG Memoir 65, 153-175. 58 1696 Perutz, M. F., 1948. A Description of the Iceberg Aircraft Carrier and the Bearing of the Mechanical 1697 Properties of Frozen Wood Pulp upon Some Problems of Glacier Flow. J. Glaciol., 1, 95–104. 1698 Petit, J-R., Jouzel, J. Raynaud, D. Barkov, N.I. Barnola, J.M. Basile, I. Benders, M. Chappellaz, J. 1699 Davis, M. Delaygue, G. Delmotte, M. Dotlyakov, V.M. Legrand, M. Lipendoc, V.Y. Lorius, C., 1700 Pepin, L. Ritz, C., Saltzman, E., Stievenard, M., 1999. Climate and atmospheric history of the 1701 past 420,000 years from the Vostok ice core, Antarctica. Nature 399, 429-436. 1702 1703 1704 1705 1706 Pfiffner, O.A., 1993. The structure of the Helvetic nappes and its relation to the mechanical stratigraphy. Journal of Structural Geology 15, 511-521. Player, R.A., 1969, Salt plugs study. Iranian Oil Operating Companies, Geological and Exploration Division, Tehran, Report No. 1146. Pohjola, V.A., Hedfors. J. 2003. Studying the effects of strain heating on glacial flow within outlet 1707 glaciers from the Heimefrontfjella Range, Dronning Maud Land, Antarctica. Annals Glaciology 1708 37, 134-142. 1709 Posey, H.R., Price, P.E., Kyle, J.R., 1987. Mixed carbon sources for calcite cap rocks of gulf coast 1710 salt domes. In Lerche, I., O’Brien, J.J., (Eds.) Dynamical Geology of salt and related structures. 1711 Academic Press, Orlando, Florida, 593-630. 1712 1713 1714 1715 1716 1717 1718 1719 Priscu, J.C., Christner, B., 2004. Earth’s icy biosphere. In: Bull A. T. (Ed), Microbial Diversity and Bioprospecting. American Society for Microbiology, Washington, D.C. p. 130-145. Priscu, J.C, Christner, B.C, Foreman, C.M., Royston-Bishop, G., 2005. Biological Material in Ice Cores. Encyclopedia of Quaternary Sciences Submitted 6 October 2005. Ramberg, H., 1975. Particle paths, displacement, and progressive strain applicable to rocks. Tectonophysics 28, 1-37. Ramberg, H., 1981. Gravity, Deformation and the Earth’s crust (second edition), Academic Press, London. 452 pp. 1720 Ramsay, J.G., 1967. Folding and fracturing of rocks, McGraw Hill, New York. 568pp. 1721 Ramsay, J.G., Huber, M.I., 1983. The techniques of modern structural Geology, Vol 1: Strain 1722 1723 1724 analysis, Academic Press, London, 307 pp Raymond, C. F., Harrison, W. D. 1988, Evolution of Variegated Glacier, Alaska, U.S.A., prior to its surge. J. Glaciol., 34, 154-169. 59 1725 1726 1727 Reeh, N., 1987. Steady-state three-dimensional ice flow over an undulating base: first-order theory with linear ice rheology. J. Glaciol. 33, 177-185. Reyss, J. L., Pirazzoli, P. A., Haghipor, A., Hatté, C., Fontugne, M., 1998. Quaternary marine 1728 terraces and tectonic uplift rates on the south coast of Iran. Geological Society London Special 1729 Publications 146; p. 225-237 1730 1731 Robin, G.De Q., Swithinbank, C., 1988. Fifty years of progress in understanding ice sheets. J. Glaciol., Special Issue, 33-48. 1732 Roedder, E., Belkin, H.E., 1980. Thermal gradient migration of fluid inclusions in single crystals of 1733 salt from the Waste Isolation Pilot Plant site (WIPP), In Northrup, C.J.M., (Ed), Scientific Basis 1734 for Nuclear Waste Management: Volume 2, Plenum Press, New York City, New York, p. 453- 1735 464. 1736 1737 1738 1739 1740 Rogers, S.O, Starmer, W.T., Castello, J.D. 2004. Recycling of pathogenic microbes through survival in ice. Medical Hypotheses 63, 773-777. Rosenkrans, R.R., Marr, D.J., 1967. Modern seismic exploration of the Gulf coast Smackover trend. Geophysics 32, 184-206. Rouby, D., Raillard, S., Guillocheau, F., Bouroullec, R., Nalpas, T., 2002. Kinematics of a 1741 growth/raft system on the W African margin using 3-D restoration. Journal of Structural Geology 1742 24, 783–796. 1743 Rowan, M.G., 2003, Stratigraphy and biostratigraphy in and immediately adjacent to salt bodies, an 1744 Overview AAPG Annual Convention Abstracts May 11-14, 2003, Salt Lake City, Utah (no page 1745 #). 1746 1747 1748 Rowan, M.G., Jackson, M.P.A., Trudgill, B.D., 1999. Salt-related fault families and fault welds in the northern Gulf of Mexico. AAPG Bulletin 83, 1454–1484. Sage, S.F., Gudela Von Gronefeld, G.V., Déverchère, J., Gaullier, V., Maillard, V., Gorini, C., 2005. 1749 Seismic evidence for Messinian detrital deposits at the western Sardinia margin, northwestern 1750 Mediterranean Marine and Petroleum Geology 22, 757-773. 1751 Satterfield, C.L., Lowenstein, T.K., Vreeland, R.H., Rosenzweig, W.D., Powers, D.W., 2005. New 1752 evidence for 250 Ma age of halotolerant bacterium from Permian salt crystal: Geology 33, 265– 1753 269. 60 1754 Säwström, C., Mumford, P., Marshall, W., Hodson, A., Laybourn-Parry. J., 2004. The microbial 1755 communities and primary productivity of cryoconite holes in an Arctic glacier (Svalbard 79°N). 1756 Polar Biology 25, 591-596. 1757 1758 Schenk, P., Jackson, M. P. A., 1993. Diapirism on Triton: a record of crustal layering and instability. Geology 21, 299–302. 1759 Schenk, P.M., Pappalardo, R.T., 2004. Topographic variations in chaos on Europa: implication for 1760 diapiric formation. Geophysical Research Letters 31, L16703, doi:10.1029/2004GL019978. 1761 Schleder, Z., Urai, J.L. 2006. Deformation and recrystallization mechanisms in mylonitic shear 1762 zones in naturally deformed extrusive Eocene-Oligocene rocksalt from Eyvanekey plateau and 1763 Garmsar hills (central Iran), Journal of Structural Geology 29, 2241-255. 1764 Schleder, Z., Urai, J.L. 2008. 1765 http://www.ged.rwthaachen.de/Ww/people/alumni/zsolt/SchlederUrai_SPP04_v08/ 1766 Halite_microstructures_SPP.htm 1767 Schuster, D.C., 1995. Deformation of allochthonous salt and evolution of related salt-structural 1768 systems, Eastern Louisiana Gulf Coast. In: Jackson, M.P.A., Roberts, D.G., Snelson, S., (Eds.), 1769 Salt tectonics; a global perspective; AAPG Memoir 65, 177-198. 1770 Seni, S.J., 1987. Evolution of Boling dome cap rock with emphasis on terrgigenous clastics, Fort 1771 Bend and Wharton Counties, Texas. In Lerche, I., O´Brien J.J., (Eds.), Dynamical Geology of 1772 salt and related structures. Academic Press, Orlando, Florida, 543-591. 1773 1774 Sherkati, S, Letouzey, J., 2004. Variation of structural style and basin evolution in the central Zagros (Izeh zone and Dezful Embayment), Iran. Marine and Petroleum Geology 21, 5535-554. 1775 Showman, A.P., Han, L., 2004. Numerical simulations of convection in Europa's ice shell: 1776 implications for surface features Journal of Geophysical Research 109, E01010, 1777 doi:10.1029/2003JE002103. 1778 Shumskii, P.A., 1964. Principles of structural glaciology. Dover Publications, New York, 497 pp. 1779 Siegert, M.J. 2007. Sunless Seas. Geoscientist 17, 20-23. 1780 Smith, A.W., Skilling, D.E. Castello, J.D. Rogers. S.O., 2004. Ice as a reservoir for pathogenic 1781 human viruses: Specifically calciviruses, influenza viruses, and enteroviruses. Medical 1782 Hypotheses 63, 560–566. 61 1783 1784 Sohn, H.G., Jezek, K., van der Veen C., 1998. Jakobshavn Glacier, West Greenland: 30 years of spaceborne observations, Geophysical Research Letters 25, 2699-2702. 1785 Souchez, R., Vandenschrick, G., Lorrain, R., Tison, J.L., 2000 Basal ice formation and deformation 1786 in central Greenland; a review of existing and new core data. In: Maltman A.J, Hubbard, B., 1787 Hambrey, M., (Eds.) Deformation of Glacial materials. Geological Society London Special 1788 Publications 176. 13-22. 1789 1790 1791 Sowers, T., 2001. The N2O record spanning the penultimate deglaciation from the Vostok ice core. Journal Geographical Research 106, 31903-31914. Spiers, C.J, Schutjens, P.M.T.M, Brzesowsky, R.H, Peach, C.J., Liezenberg, J.L, Zwart, H.J., 1990. 1792 Experimental determination of constitutive parameters governing creep of rocksalt by pressure 1793 solution. In Knipe, R.J., Rutter, E.H., (Eds.), Deformation mechanisms, rheology and tectonics. 1794 Geological Society of London Special Publications 54, 509-522. 1795 1796 Suppe, J., 1985. Principles of Structural Geology. Prentice-Hall, Englewood Cliffs, New Jersey. 537 pp. 1797 Talbot, C. J., 1978. Halokinesis and thermal convection. Nature, 173, 739-741. 1798 Talbot, C.J., 1979. Fold trains in a glacier of salt in southern Iran. Journal of Structural Geology 1, 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 5-18. Talbot, C.J., 1981. Sliding and other deformation mechanisms in a salt glacier, Iran. Geological Society of London Special Publication 9, 173-183. Talbot, C.J., 1992. Quo vadis tectonophysics? With a pinch of salt! Journal of Geodynamics 16, 120. Talbot, C.J., 1998, Extrusions of Hormuz salt in Iran, In: Blundell, D.J., Scott, A. C., (Eds.), Geological Society of London Special Publication 143, 315-334. Talbot, C.J., 2004, Extensional evolution of the Gulf of Mexico basin and deposition of Tertiary evaporite: Discussion. Journal of Petroleum Geology 27, 95-104. Talbot, C.J., 2008. Hydrothermal salt- but how much? Discussion. Marine & Petroleum Geology 25, 191-202. Talbot, C.J., Rogers, E., 1980. Seasonal movements in an Iranian salt glacier. Science, Wash. 208, 395-397. 62 1812 1813 1814 1815 1816 1817 1818 1819 Talbot, C.J., Jarvis, R. J., 1984: Age, budget and dynamics of an active salt extrusion in Iran. Journal of Structural Geology 6, 521-533. Talbot, C.J., Jackson, M. P. A.., 1987. Internal kinematics of salt diapirs. AAPG Bulletin 71, 10681098. Talbot, C.J., Alavi, M., 1996. The past of a future syntaxis across the Zagros. Geological Society London Special Publications, 100; 89-109. Talbot, C.J., Stanley, W., Soub, R., A-Sadoun, N., 1996. Epitaxial salt reefs and mushrooms in the southern Dead Sea. Sedimentology 43, 1025-1047. 1820 Talbot, C.J., Medvedev, S., Alavi, M., Shahrivar, H., Heidari, E., 2000. Salt extrusion rates at Kuh-e- 1821 Jahani, Iran: June 1994 to November 1997, Geological. Society London Special Publications 1822 174, 93-110. 1823 1824 1825 1826 Talbot, C.J., Aftabi, P., 2004, Geology and models of salt extrusion at Qum Kuh, central Iran. Journal of the Geological Society of London 161, 321-334. Thorsteinsson, T, Waddington, E.D., 2002. Folding in strongly anisotropic layers near ice-sheet centers. Annals Glaciology 35, 480-486. 1827 Tison J. L., Hubbard, B., 2000. Ice crystallographic evolution at a temperate glacier: Glacier de 1828 Tsanfleuron, Switzerland. Geological Society London Special Publications 176, 23-38. 1829 1830 1831 1832 1833 1834 1835 Turcotte, D. L., Schubert, G., 1982. Geodynamics: applications of continuum physics to geological problems. John Wiley & Sons. New York. 450 pp. Unwin, B., Wingham, D., 1997. Topography and dynamics of Austfonna, Nordaustlandet, Svalbard, from SAR interferometry. Annals Glaciology 24, 403-408. Urai J,L., Spiers, C. J., Zwart, H. J., Lister, G.S., 1986. Weakening of rock salt by water during long term creep. Nature 324, 554-557. van de Wal, R.S.W, Boot, W. Van de Broeke, M.R., Smeets, C.J.P.P., Reijmer, C.H., Donker, 1836 J.J.A., Oerlmans, J. 2008. Large and rapid melt-induced velocity changes in the ablation zone 1837 of the Greenland ice sheet. Science 321, 111-113. 1838 1839 1840 1841 Ville, L., 1856. Notice geologique sur les salinesde Sahrez et les gites de sel gemme de Rang el Melah et d’Ain Hadjera (Algerie). Annales des Mines 15, 351-410. Volozh, Y., Talbot, C., Ismail-Zadeh, A., 2003. Salt structures and hydrocarbons in the Pricaspian basin. AAPG. Bulletin 87, 313-334. 63 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 Vreeland, R. H., Rosenzweig, W.D., Powers, D.W., 2000. Isolation of a 250 million-year-old halotolerant bacterium from a primary salt crystal. Nature 207, 897-900. Waddington, E.D., Bolzan, J.F., Alley, R.B., 2001. Potential for stratigraphic folding near ice-sheet centers. J. Glaciol., 47, 639-648. Walker, J.C.F., Waddington. E.D., 1988. Early discoverers XXXV: Descent of glaciers: some early speculations of glacier flow and ice physics. J. Glaciol., 34, 342-348. Wang, Y., Thorsteinssson, T., Kipfstuhl, J., Miller, H., Dahl-Jensse, D., Shoji, H., 2002. A vertical girdle fabric in the NorthGRIP deep ice core, North Greenland. Annals Glaciology 35, 515-520. Warren, J K. 1997, Evaporites, brines and base metals: Fluids, flow and 'the evaporite that was'', Australian Journal of Earth Sciences 44, 149 -183. Warren, J., 1999. Evaporites, their evolution and economics. Blackwell Science, Oxford, UK. 438 pp. 1854 Washburn, A.L. 1979. Geocryology. Edward Arnold, London, 406 pp. 1855 Weinberg, R.F., 1993. The upward transport of inclusions in Netwonian and power-law salt diapirs. 1856 1857 Tectonophysics 228, 141-150. Weinberger, R., Begin, Z.B., Waldman, N., Gardosh, M., Baer, G., Frumkin, A., Wdowinski, S., 1858 2006. Quaternary rise of the Sedom diapir, Dead Sea basin, In Enzel, Y., Agnon, A., Stein, M., 1859 (Eds.) New frontiers in Dead Sea paleoenvironmental research: Geological Society of America 1860 Special Paper 401, 33-51. 1861 Wenkert, D.D., 1979. The flow of salt glaciers, Geophysical Research Letters 6, 523-526. 1862 Wilson, C.J.L. Zhang, Y., 1996. Development of microstructure in the high-temperature deformation 1863 of ice. Annals Glaciology 23, 293-302. 1864 Wilson, C.J.L., Marmo, B., 2000, www.virtualexplorer.com.au/ VEjournal/Volume2. 1865 Wingham, D.J., Siegert, M J., Shepherd, A., Muir, A.S., 2006. Rapid discharge connects Antarctic 1866 1867 subglacial lakes. Nature 440, 1033-1036. Worrall, D.M., Snelson, S., 1989. Evolution of the northern Gulf of Mexico, with emphasis on 1868 Cenozoic growth faulting and the role of salt. In: Bally, A., Palmer, A., (Eds.), The geology of 1869 North America-an overview. Geological Society of America, Boulder, Colorado, pp. 97–138 (v. 1870 A). 64 1871 1872 1873 1874 Worster, M.G., Wettlaufer J.S., 1999. In Shyy, W. (Ed) Fluid Dynamics at Interfaces, Cambridge University Press, Cambridge, p. 339. Wu, S., Bally, A.W., Cramez, C, 1990. Allochthonous salt, structure and stratigraphy of the northeastern Gulf of Mexico, Part 11; Structure. Marine and Petroleum Geology 7, 334-370. 1875 Yatsu, E., 1988. The nature of weathering. Sozosha,Tokyo,624 pp. 1876 Yuen, D.A., Saari, M.R., Schubert, G., 1986. Explosive growth of the heating instabilities in the 1877 1878 down-slope creep of ice sheets. J. Glaciol., 32, 314-320. Zirngast. M., 1996. The development of the Gorleben salt dome (northwest Germany) based on 1879 qualitative analysis of peripheral sinks. Geological Society of London Special Publication 100, 1880 203-226. 1881 Zwally, H.J., et al., 2006. Eos Trans. AGU 87 Fall meet suppl. Abstract C14B-00. 1882 1883 Talbot and Pohjola Figure captions 1884 1885 Fig. 1. Photographs illustrating similarities between bodies of ice (In) and salt (Sn): IA: Outlet 1886 glaciers draining from Vatnajökull ice cap, southern Iceland (Photos of ice by VP; of salt by CT 1887 unless otherwise indicated). SA: Perspective satellite view looking east at 4 numbered 1888 extrusions of Hormoz salt at different stages of development in the Zagros Mountains, SE of the 1889 city of Lar, Iran. The dark tones are due to clays in soils of insoluble Hormoz materials left on 1890 the surface after salt dissolution. 1: Vigorous young salt dome rises to 2.2 km above sea level 1891 in the western end of a short Zagros anticline. 2: A namakier without visible salt on its surface 1892 flows 7 km south from diapir breaching Kuh-e-Gach anticline. 3: Another namakier flows north 1893 from a more subdued salt dome in Kuh-e-Burkh anticline. 4: Only gravels of Hormoz materials 1894 remain on the plain north of the remains of a salt dome in an amphitheatre of Cretaceous- 1895 Eocene limestones at the west end of Kuh-e-Gach. Image by 1896 NASA/GSF/METI/ERSDAC/JAROS and the U.S/Japan ASTER Science team and is available 1897 at: http://asterweb.jpl.nasa.gov/content/02_gallery/images/salt-iran-view 1898 IB: Vatnajökull ice cap, southern Iceland, from the south. SB: Salt dome and southern namakier of 1899 Kuh-e-namak (Dashti), Iran from south (28°17'S, 51° 43'E). IC: Franz Josef glacier, New 1900 Zealand from www. SC: Looking up at NE corner of salt dome at Kuh-e-namak (Dashti), Iran. 65 1901 ID: Ice pinnacles of ice towering over terminal face of Franz Josef glacier, New Zealand from 1902 picasaweb.google.com/ SD: Looking NW over northern namakier of mid-reaches of Kuh-e- 1903 namak (Dashti), Iran 1904 1905 Fig. 2. More photographs comparing ice (In) and salt (Sn) on different scales. IA: The advancing 1906 front of Franz Josef Glacier, New Zealand. Debris-rich basal ice facies seen below upwarped 1907 foliation in layered ice. SA: Dirty folded and jointed salt on NE corner of northern namakier of 1908 Kuh-e-namak (Dashti), Iran. IB: Ice pinnacle on Franz-Josef glacier, New Zealand, from 1909 www.virtualtourist.com. SB: salt pinnacle of Ku-e-namak (Dashti) (photo by M.P.A. Jackson). 1910 IC: Icicles and regelated debris bands at the base of Nordenskiöldbreen, Svalbard. SC: 1911 Stalactites and stalagmites of secondary halite fronting river cave well above base of northern 1912 namakier of Kuh-e-namak (Dashti). ID: Looking up-glacier at open upright folding of 1913 stratification with axial plane foliation, Austre Lovénbreen, Svalbard (from Hambrey and 1914 Lawson, 2000, fig 8).SD: Recumbent similar-type fold in mylonitised distal salt of the 1915 axisymmetric “fried egg” salt fountain Zagros Mountains (at 28º00’N, 54º55’E). IE: 'Similar' fold 1916 in fine-grained and coarse bubbly ice foliation with vertical axial plane, Vadret del Forno, 1917 Switzerland (from Hambrey and Lawson 2000, fig 6).SE: Similar folds with axial plane foliation 1918 in Tertiary salt of Qum Kuh, central Iran (see Talbot and Aftabi, 2004). 1919 1920 Fig. 3. Comparison of ice and salt cycles. A: Ice usually flows from where and soon after it forms. 1921 B: Salt sequences have to be buried and squeezed back to the surface for namakiers to 1922 become analogues of glaciers. 1923 1924 Fig. 4. Categories of salt diapirs. A: The crest of a “passive” diapir stays near the depositional 1925 surface as it is downbuilt by other rocks accumulating on the surrounding source layer. B: 1926 Differential loading drives reactive diapirs to pierce overburdens weakened and thinned by faulting, 1927 here as is common in Iran, at a releasing bend along a strike-slip fault. If reactive diapirs reach the 1928 surface they may rise further by active growth through pre-growth overburden and/or downbuilding 1929 by syn-growth overburden. C: Rare active diapirs upbuild through pre-growth ductile then syn- 1930 growth overburden. 66 1931 1932 Fig. 5. Types of salt structures (after Jackson and Talbot, 1994) with extrusive salt analogues of bodies of crystalline ice shaded. 1933 1934 Fig. 6. Topographic profiles of extrusions of salt in the Zagros Mountains of Iran (from fig. 3 in 1935 Talbot, 1998) superposed to illustrate: A: how they grow with time and B, when no longer 1936 supplied by salt from depth, collapse to the theoretical profile of a viscous droplet (dashed) and 1937 then degrade, mainly by dissolution. 1938 1939 Fig. 7. Natural scale centred profiles of extrusions of salt of different ages in Iran and their internal 1940 structures. A. N-S profile of Hormoz salt (~500 Ma) that rose up an opening pulled-apart along 1941 a NS transfer fault in the Zagros Mountains to extrude Kuh-e-Jahani, probably the largest 1942 current salt fountain in Iran, at 28°37’N, 52°25’E (from Talbot et al., 2000). Star indicates flow 1943 separation point beneath summit dome. B. N-S profile of Tertiary salt extruding at Qum Kuh at 1944 34°48’N, 50°41’E on the central plateau of Iran (fr om Talbot and Aftabi, 2004). Notice the cap 1945 soil of insoluble components on the surface (black) thickening downstream as the salt dissolves 1946 and parts of the outer carapace of brittle dilated salt (grey in inserts) in divides between rivers 1947 that erode down toward confined salt. Folds in colour bands within the salt are also indicated. 1948 C: Folds near the southern (feather-edged) retreating terminus of Kuh-e-namak (Dashti). Salt 1949 relief shown is about 45 m. D: Tank-track fold ~100 m short of terminus in northern namakier 1950 off Kuh-e-namak (Dashti). Figure for scale right foreground. 1951 1952 Fig. 8. Photographs of grain shape fabrics in glacier ice and namakiers of salts of different ages in 1953 Iran. Pictures of Ice were taken between crossed polars. IA: Thin slice of coarse grain ice 1954 specimen taken from the front of Engabreen, Norway. SA: a halite “pegmatite” in Kuh-e- 1955 Bachoon, Zagros Mountains (29º00’N, 52º15’E) from Talbot, 1998. IB: Thin slice of 1956 heterogeneous and recrystallised fine grained ice extracted from a tunnel blasted into the base 1957 of Engabreen, Norway. SB: Part of an automated map by electron backscatter diffraction of 1958 polished and chemically etched sample of bimodal salt from the Eyvanekey plateau, ~50 km 1959 east of Tehran. Different orientations of crystal lattice appear in different shades (from 1960 Schleder and Urai 2006 fig. 7A.) IC: Thin slice of homogenous medium grained ice from a core 67 1961 drilled into Storglaciären, Sweden. SC: Bi-modal grain shape fabric in Hormoz salt on southern 1962 namakier Kuh-e-namak (Dashti). ID: Basal ice facies from the ice tunnel beneath Engabreen, 1963 Norway, showing pinched ice layers interbedded with debris rich strata The vertical side is ca 1964 70 cm high SD: Aligned ellipsoidal megacrysts of halite with length/width ratios of 1.5:1 in a 1965 halite gneiss; pipe is 15 cm long, Kuh-e-namak (Dashti). IE Basal ice stratigraphy from the ice 1966 tunnel beneath Engabreen, Norway, showing vertical compression of ice close to larger stones 1967 frozen into the ice (dark objects). The vertical side is ca 70 cm high. SE: Salt mylonites in which 1968 halite grains can reach length/width ratios 4:1 along ductile shear zones typically only dm thick, 1969 here thicker in Tertiary salt in the Eyvanekey plateau, 50 km east of Tehran, Iran. 1970 1971 Fig. 9. Vertical profiles of viscous fluids with passive marker layers gravity spreading over 1972 horizontal substrates with different bottom boundary conditions. A1-3: Three stages in spreading 1973 of a block of silicone putty over a horizontal high-friction substrate (fig. 9-10, Ramberg, 1981). 1974 Distal stratification curls downward distally in tank track fold signaling high-frictional base. A4-6: 1975 Three stages in similar blocks of silicone putty gravity spreading over mercury (figs. 9-7 & 9-12, 1976 rearranged from Ramberg 1981). Stratification curls upward distally over free-slip base. B-F: 1977 Thin vertical slices of experimental extrusions of transparent viscous silicone polymer (from fig. 1978 12, Talbot and Aftabi, 2004). Flow folds in passive markers in the upper limbs of tank-track fold 1979 (numbered) record the number of pauses in their extrusion history. B: Fundamental recumbent 1980 tank-track fold developed with a simple upper limb records steady extrusion. C-E Increasing 1981 number of flow folds in the upper limb of tank-track fold record number of pauses in loading 1982 history. Scale bars in mm. F: Tank track fold undulating over a no-slip irregular slope 1983 (experiment by Rosemary Talbot). 1984 1985 Fig. 10. Passive stratification within salt develops trains of internal similar-type flow folds with an 1986 axial planar grain shape fabric in association with every significant obstruction in their channel floor 1987 they flow beyond (from Talbot, 1981). A: Different regimes of structures shown in B are developed 1988 to different degrees down the namakier. B: Instead of eroding obstructions, the salt infills troughs in 1989 its bed and flows beyond the obstruction by detaching along mylonite zones ≤1 dm thick. C: Only 1990 some of the distal salt flows beyond some of the last obstructions. 68 1991 Fig.11. Stress relief fractures newly exposed beneath an irregular top of the salt in 1992 Drakhsharnamak quarry, NW Eyvanekey Plateau, 60 km east of Tehran (photo taken January 1993 2002). Most of these fractures, a few decimetres apart, are horizontal and slightly oblique to the 1994 salt layering (that dips gently left); many are infilled by relatively insoluble residual soils washed 1995 in from above. The lowest fracture here is estimated to be about 6 m beneath the top of salt 1996 and is unusually long. In the upper slopes of most salt extrusions, the planar fracture 1997 anisotropy may parallel and emphasise a strong gneissose foliation and dies out downward, 1998 passing into sound salt at depths between 4 and 10 m. 1999 2000 Fig. 12. Working model for slow normal flow and short rapid surges in namakiers supplied from a 2001 country-rock vent at an essentially steady rate. A. Most of the time gravity spreads dry salt at 2002 cm/dm/year and develops a tank track fold as extruded salt is stored at high levels by strong 2003 elastic carapace. B. Salt surges at dm/day for a few days over near basal mylonite zones after 2004 3 or 4 rain storms per decade have weakened outer carapace. 2005 Figure 1 Click here to download high resolution image Fig1-color Click here to download high resolution image Figure 2 Click here to download high resolution image Fig2-color Click here to download high resolution image Figure 3 Click here to download high resolution image Figure 4 Click here to download high resolution image Figure 5 Click here to download high resolution image Figure 6 Click here to download high resolution image Figure 7 Click here to download high resolution image Fig7-colour Click here to download high resolution image Figure 8 Click here to download high resolution image Fig-8 colour Click here to download high resolution image Figure 9 Click here to download high resolution image Figure 9-colour Click here to download high resolution image Figure 10 Click here to download high resolution image Figure 11 Click here to download high resolution image Figure 12 Click here to download high resolution image Table 1 Click here to download Table: Table 1.doc Table 1: comparison of properties of ice and salt Chemistry Name Crystallography Typical crystallisation pressure Moh’s hardness Density at 1 bar Ice H20 Water ice Hexagonal/trigonal Up to 253 MPa at -22°C (Shumskii, 1964, p.180) 1.5 at 0°C to 6 at –78.5°C 3 916.8 Kg m- Melting point Solubility in water ~ 0°C Brittle shear strength 6 kg cm at 3°C (anisotropic) Typical flow stresses 0.001% of Young’s modulus Typical viscosity en masse 10 Pa “warm” 14 10 Pa s “cold” (Shumskii, 1964) -1 1-10 m a when dry 4 -1 10-10 m ar with basal slip -6 -1 17-29x10 K (Shumskii1964) -6 2 -1 1.33x10 m s (Drewry 1986) mm to dm 1-5 (usually near 3) Pressure solution creep dislocation creep, dynamic recrystalisation, basal slip Typical flow rates in nature Linear thermal expansivity Thermal diffusivity Typical grain size Flow law exponent Flow mechanisms infinite -2 10-11 Salt NaCl halite Face centred cubic 66-379 MPa depending on brine concentration (Yatsu 1988 p. 86). 3 3 Pure halite 2170 Kg m- , 3 Rock salt often taken as 2200 Kg m~ 700°C (~973 K) At 25°C, 1 cm of rain can potentially dissolve 0.1667 cm of salt (Lockner, 1995). In experiment, water with 300 gr/L of NaCl 2 dissolved 0.2 gr/s/ m (Alkattan et al., 1997). 5 MPa at 100ºC (Kawamoto & Shimamoto, 1997) ~7 MPat ~250ºC (Kawamoto & Shimamoto, 1997) 13-16 Pa s (damp) 10 18 10 Pa s (dry) depending on grain size & temperature (Warren 1999 p. 154) -1 -1 1-10 m a (0.5 m dayr when wet) -5 -1 4.2x10 K (Gevantman, 1981) -6 2 -1 3.6x10 m s (Durham et al., 1981) 5-20 mm (Warren 1999 p. 159) 1 (damp) 3—8 (dry) Pressure solution creep, dislocation creep, dynamic recrystalisation. Table 2 Click here to download Table: Table 2.doc Table 2. Comparison of formation and deformation Material Formation lithification downslope flow Loss mchanisms Ice salt Snow largely sedimented from above(random c axes) but also by freezing of infiltrated melt water resulting in more oriented fabrics Compaction and crystal growth due to vapour transport, and strain induced recrystalisation Compaction to shear in decades Ablation, sublimation, calving Crystallisation from brines evaporating to dryness or epitaxial growth in permanent evaporating brines. By cementation Power law (n=3) penetrative creep, basal slip, dynamic recrystalisation Subdued on accumulation zone, ≥10 m in serac zones Active strain markers Stratigraphy: grain size, bubble & impurity content, Foliation, crevasses, crevasse traces & infills, grain boundaries, bubble elongation Stratigraphy, foliation (difficult to see) Passive strain markers Gravity spreading mechanisms Surface relief 6 Burial, diapir, extrusion over 10 a Erosion (mainly dissolution) over whole area of extrusion, local river erosion Power law (n=3) penetrative creep, basal slip, dynamic recrystalisation ≤5 m on summit dome, increasing to ≥20 m in lower reaches Non-salt layers, veins Bedding, obvious grain shape and orientation foliation, Folds boudins, stripes, thrusts, ogives, lags Easy to see folds, boudins, shears, structures Difficult to see thrusts, lags. Supra-, en-, sub-, lateral, terminal En- and supra moraines Upright, surface waves, time transients Reclined, recumbent Dynamic folds Flow parallel to boundaries so convergent Where flow is convergent, Tank-Track Fold Easy Kinematic folds bed topography related, subtle in low relief to see on surface relief Significant by basal tools where basal friction None, no basal tools instead salt infills Bedrock erosion hollows and flows beyond detaching in-salt mylonites Impurities and water content, fossil in Flow regime control Temperature, water content, strong/weak beds intrusions, mainly rain in extrusions Clean salt attached no slip. subglacial impurities common. Cold ice frozen to bed, warm Types of basal ice subject to basal slip & regelation contact Entrainment, plucking, quarrying and ploughing common Very minor equivalents where salt flows Basal erosion over soil substrates