Quaternary Geomorphic Processes and Landform Development in the Thar Desert of Rajasthan

Home , Aeolian landform, Jawai River, Tilwara

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/302902544

Chapter · January 2011

CITATIONS READS 6 4,293

1 author:

Amal Kar Central Arid Zone Research Institute (CAZRI)

90 PUBLICATIONS 1,166 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Thar Desert Natural resources and their management View project

Late Quaternary paleoclimate of the Thar Desert View project

All content following this page was uploaded by Amal Kar on 11 May 2016.

The user has requested enhancement of the downloaded file. acb publications Landforms Processes & Environment Management Kolkata, India Editor: S. Bandyopadhyay et al. [email protected] ISBN 81-87500-58-1 2011 (223-254)

Quaternary Geomorphic Processes and Landform Development in the Thar Desert of Rajasthan

Amal Kar1

Abstract: Evolution of landforms in the Thar Desert of Rajasthan is very much influenced by the exogenic and endogenic processes operating in the region during the Quaternary period. Studies have revealed that several fluctuations in climate between drier and wetter phases and periodic earth movements decided the type and intensity of geomorphic processes. The paper describes the broad sedimentation pattern in the desert, known facets of Quaternary climate and landform characteristics. It also discusses the influence of Quaternary climate change, neotectonism and human activities on landform evolution.

Introduction The Thar, or the Great Indian Sand Desert, is situated in the arid western part of Rajasthan state in India and the adjoining sandy terrain of Pakistan. It forms a distinctive, but integral part of the arid lands of western India that runs through the states of Punjab, Haryana, Rajasthan and Gujarat. The eastern limit of the desert can be marked along the calculated moisture availability index (also called the aridity index) of –66.6, which roughly passes through the foothill zone of the degraded, NNE to SSW-trending Aravalli mountain ranges. In the west, the desert extends up to the fertile alluvial plains of the Indus in Pakistan. The Aravalli hill ranges, which partially control the spatial pattern of present-day rainfall in the region, and through it the efficiency of different geomorphic processes, were formed more than 2500 million years ago. It underwent at least three cycles of orogenesis and planation since the Proterozoic, and is now one of the oldest hill ranges in the world. The Aravalli orogenic cycles and the attendant widespread igneous activities were responsible for the construction of much of the basement for subsequent sediment accumulation. The basement in much of the desert area today is made up of granite and rhyolite and the gneissic complex. From the upper Proterozoic period onward sedimentation here took place in several basins, under continental and marine conditions. The identified basins are: (i) Marwar basin, (ii) Lathi basin, (iii) Jaisalmer basin, (iv) Barmer basin, (v) Palana-Ganganagar Shelf, and (vi) Sanchor basin. These

1Central Arid Zone Research Institute, Jodhpur 342 003 224 Landforms Processes and Environment Management basins, separated from each other by major tectonic features, including faults, formed together a shelf region that merged with the Indus geosyncline further west. Excellent reviews on the geological and structural frameworks of the Aravalli and the desert area to the west of it are available in Heron (1953), Narayanan (1964), Sen (1970), Chatterji (1977), Dasgupta and Chandra (1978), Pareek (1984), Roy (1988) and Sinha Roy et al. (1998). Broadly, the landforms in the Thar are fluvial, aeolian and lacustrine in nature (Fig. 1; Table 1). Fluctuating climate since the beginning of the Quaternary period about 2 million years ago played a major role in their evolution. There is also growing evidence of periodic tectonic activities during the Quaternary, which accelerated the subaerial processes and left their imprints on the landforms. The basement for

Figure 1. Landforms of Thar desert, Rajasthan

Quaternary sedimentation in Rajasthan part of the Thar was essentially a vast pediplaned surface which was composed largely of the Pre-Cambrian metasediments and igneous in the east and a gradually thickening sedimentary deposits of Mesozoic and Tertiary periods in the west. We propose to discuss here the present understanding of landform development in the region, the related geomorphic processes, as well as the possible forcing mechanisms. Earlier reviews on the geomorphology of the desert

226 Landforms Processes and Environment Management are available in Ghose et al. (1977a), Allchin et al. (1978), Kar (1992, 1995), Singhvi and Kar (1992), and Singh et al. (1997).

Climate change during the Quaternary and sedimentation pattern The discovery of the wide, dry bed of the Ghaggar in the northern fringe of the Thar, its suggested link with the legendary Saraswati river from the Himalayas and establishment of the contribution of the Sutlej in its survival (Oldham, 1893), provoked the earth scientists to search for clues on climate change and tectonism. Ghose (1964) reconstructed from aerial photographs the early integrated drainage systems in the south-central part of the desert between Pali and Jalor, and concluded that some time in the past the climate in the desert was wetter. Several multi-disciplinary studies since then have established that the Quaternary climate fluctuated many time between wetter and drier phases (Bryson and Baerries, 1967; Singh et al. 1974; Ghose et al., 1977b; Allchin et al., 1978; Wasson et al., 1983; Singhvi and Kar, 1992, 2004; Kar et al., 2001, 2004). Observation of several shallow and deep Quaternary sequences confirms that the Quaternary lithofacies in the desert comprise essentially of the alternate fluvial and aeolian sedimentary deposits. Lacustrine and fluvio-lacustrine deposits are noticed in some favoured locations. Correct assessment of land forming processes and chronology of events during the early part of the Quaternary is difficult, especially because of a sharp erosional contact between the Quaternary and the underlying Tertiary/pre-Tertiary deposits, shallowness of the Quaternary sediment thickness at many places, paucity of good exposures in the areas of deep sequences, and problems of dating the old sediments through currently available techniques. Reliable chronology of events, on the basis of radiocarbon dating of organic materials and luminescence dating of quartz and feldspar in the sediments (Singhvi and Krbetschek, 1996), is available for the late Quaternary period only. The number of sites explored systematically for such purposes is, however, limited. Therefore, the spatio-temporal resolution of the derived chronology of events is coarse, and needs improvement. So far the majority of older surviving Quaternary deposits have been found to be a mixture of sand and gravel. Signatures like cross-bedded gravel-and-pebble-rich sediments over many of the Tertiary/ pre-Tertiary sequences in the western part of the desert suggest a wetter climate during the transition from Tertiary to Quaternary period, with high intensity rainfall that was responsible for such high energy fluvial deposition. Wadhawan (1988), Raghav (1992), Dhir et al. (1994), Wadhawan and Kumar (1996), and Rakshit and Sundaram (1998) have described typical Quaternary litho-stratigraphy from different parts of the desert and its eastern fringe. A summary is provided in Singhvi and Kar (1992). At many places the lowermost conglomerates consist of boulders and gravels of quartzite and other metamorphics of Aravalli provenance, as also granites, gneisses and sedimentaries from within the desert. However, considerable discordance was found in the sedimentation pattern, which provoked Wadhawan and Sural (1992) to suggest that the sedimentation took place Quaternary Geomorphic Processes and Landform Development in the Thar Desert of Rajasthan 227 in four different neotectonically disturbed linear sub-basins, named as ‘Shahgarh- Kishangarh’, ‘Sanchor-Shergarh-Dechu’, ‘Merta-Degana-Jayal-Didwana’ and ‘Bikaner-Churu-Ganganagar’. They opined that a series of NE-SW trending horst and graben structures was responsible for controlling the origin, configuration and development of these sub-basins of Quaternary deposition. These supposed sub-basins are, however, different from those which controlled the pre-Quaternary depositional pattern in the region, and which were described by Dasgupta and Chandra (1978). Although the early Quaternary deposits at many places were found to contain fine to medium sand of possible aeolian provenance, the paucity of purely aeolian strata within or below the conglomerate deposits, and a sharp contact with the underlying Pre-Cambrian or Tertiary beds still intrigue researchers about the subaerial processes which preceded the early Quaternary fluvial activity. Only on rare occasions the contact with basement is in the form of aeolian sand deposits, as for example near Osian (60 km N of Jodhpur) where the Pre-Cambrian Jodhpur Sandstone of Marwar Supergroup is followed upwards by a slightly lithified, fine aeolian sand, and then by the calcreted sand and gravel. It is possible that aeolian deposits of earlier dry phases in the desert were mostly reworked and mixed with fluvial sand and gravel during a prolonged and intense wet phase sometime during the early part of the Quaternary, which resulted in obliteration of older sand units, except at few favoured niches where one might come across aeolian deposits beneath the ‘basal conglomerate’. A number of recent studies on the simulation of palaeomonsoon rainfall distribution and wind fields have enriched our knowledge on the possible process efficiency in the past, as well as on the relationship of Indian summer monsoon with earth’s orbital changes and other forcing mechanisms. The simulation of monsoon variability for the last 150 thousand years (or kyr) (Prell and Kutzbach, 1987), and other contemporary studies encompassing the last major glacial and interglacial periods (Bryson, 1989; Sirocko, 1991; Sirocko et al., 1993; Overpeck et al., 1996; de Noblet et al., 1996; Zonneveld et al., 1997) are notable in this regard. Simulation studies of Prell and Kutzbach (1987) suggest that the monsoon rainfall was perhaps less than the present-day average during the following periods (in kyr before present): 131- 150, 108-118, 87-98, 60-75, 14-50. The intervening periods possibly witnessed higher than the present-day average rainfall. Studies of oceanic cores from near Oman by Zonneveld et al. (1997) reveal extremely weak SW monsoon between 17.2 and 14.7 calendar kyr. Strong SW monsoon occurred between 14.7 and 11.8 calendar kyr, with several pulses of low monsoon (e.g., between 12.6 and 12.4 calendar kyr, perhaps responding to the Younger Dryas cooling). Between the last glacial maximum (LGM) and the early Holocene, two abrupt peaks of increased monsoon activity were noticed by Overpeck et al. (1996). One was between 15.4 and 13.9 calendar kyr BP and another was between 13.4 and 10.4 calendar kyr. They also noticed that during the Holocene period maximum monsoon intensity lagged peak insolation forcing by about 3 calendar kyr. Sirocko et al. (1993) noticed a similar lagged response of the SW monsoon to the insolation forcing. Provided these phases are identifiable in the 228 Landforms Processes and Environment Management sedimentary sequences in the desert, the results will help much in linking the regional climate dynamics with the past geomorphic processes, their spatial domains and contribution to sedimentation pattern. For example, during the Holocene Climatic Optimum (6 kyr) and the last major Pleistocene interglacial (126 kyr) the simulated rainfall intensity and rainy days in south Asia was found to be significantly higher than at present (de Noblet et al., 1996). By implication, the periods had a very effective fluvial activity, and the imprints should be widely noticed in the stratigraphy. During the periods when rainfall was significantly lower than at present, aridity was widespread and the desert expanded beyond its present limit. The more severe aridity took place at about 20 kyr (LGM) and ~137 kyr, when the rainfall amounts were at least 20 per cent less than the present average (Prell and Kutzbach, 1987). The dry phases were also the periods of much subdued fluvial activities when the aeolian processes were dominant. Conventionally, the efficiency of aeolian processes is thought to be synonymous with the intensity of aridity. This follows from the argument that in a desert, the decreasing rainfall generally creates conditions favourable for sand reactivation. The strength and duration of wind are usually underplayed; often these are thought to be common in the desert. New data on the timing of late Quaternary sand reactivation and dune formation in the Thar (Chawla et al., 1992; Kar et al., 1998a, 2004; Thomas et al., 1999), as well as interpretation of data from ocean cores on the strength of SW monsoon wind in the past (Sirocko et al., 1993), now suggest that the conventional views need serious reconsideration. Palynological evidence from a number of saline lakes in the Thar suggests that a hyperarid condition prevailed in the region from about the LGM. It continued up to ~13 kyr when the summer monsoon picked up. During the Holocene the mean annual precipitation was about twice the present value in 7.5–6.0 kyr, but it showed decreasing tendency from about 5 kyr (Singh et al., 1990). From ~3 kyr the region experienced a fluctuating climate between wet and dry phases, the notable dry phase being between 0.6 and 0.3 kyr, corresponding to the Little Ice Age. Analysis of rainfall data for the last 100 years at meteorological stations within the desert and its eastern fringe suggests a trend towards slight increase in the margin areas (Pant and Hingane, 1988). At the same time, rainfall in the humid eastern part of the country shows a declining trend. Such a scenario was also simulated for the past by de Noblet et al. (1996). Broadly the stratigraphic records from the desert provide numerous evidence of climatic events sketched above, especially as fluvial, aeolian and lacustrine deposits. We shall discuss in the following sections how the geomorphic process variation and landform development in the desert are related to climate change and other forcing like tectonic activities.

Major fluvial landforms Most landforms in the Thar are polygenetic in nature, but it is possible to classify them according to the processes which dominated for a fairly long period. The major fluvial landform sequence is hills and uplands - rocky/gravelly pediments (and pavements) - buried pediments (colluvial plains) - flat older alluvial plains - younger Quaternary Geomorphic Processes and Landform Development in the Thar Desert of Rajasthan 229 alluvial plains - river beds. The hill slopes in the arid areas have usually a general paucity of debris. Concave and straight segments dominate over the convex segment. Lithological and structural variations are faithfully replicated in the slope configurations, as noticed in the cuestas and mesas, as well as in the shapes of the summits in hills formed of rhyolite, granite and other rocks. The pediments at the base of hills usually have a slope of less than 4º, where the debris character is highly influenced by local lithology. Usually the pediment slope is concave with less than 2( slope, but the joint-controlled weathering and erosion along the granite pediments produce a multi-convex profile. The piedmont angle is more pronounced on sandstone and granite, but the least on rhyolite. The long history of pediplanation in the southeastern Thar, dating back to the pre-Tertiary era, has left the remnants of many dissected gravelly pediments without any trace of an adjacent hill. The desert pavements occur in the very dry western part of the Thar, mostly in the Bap-Phalodi and Pokaran-Chandan-Devikot-Rajmathai tract, where the surface is characterised by closely packed gravels and pebbles. Another notable area of occurrence is near Jayal, which yielded a rich hoard of palaeolithic tools (Misra et al., 1980). These pavements have a broadly convex outline, with rills and gullies along their margins. Down the slope from pediments and pavements, the buried pediments/ colluvial plains are composed of heterogeneous sediments. The thickness is more than a few meters near the Aravallis, but it decreases gradually westward where it varies from 30 to 60 cm. The flat older alluvial plains occur further downslope and are usually characterised by zones of illuviated soft nodular kankar, or gypsum at 20 to 200 cm depth within the sand and alluvium. The younger alluvial plains occur downslope of the older alluvial plains, especially along the major ephemeral channels, including the Luni and its tributaries in the southern Thar and the Ghaggar in northern Thar. In the non-sandy areas of the very dry western part the above landform assemblage is rarely found because of the meager rainfall (<200 mm). This is especially true for the Pokaran-Jaisalmer-Ramgarh tract where the Tertiary and pre-Tertiary beds of sandstone, limestone and shale with their typical near-horizontal or gently northwestward-dipping disposition, have been sculpted into high level structural plains (or hamadas) and cuestas, bounded by steep southeast-facing escarpments. Usually, behind the escarpment a cuesta or a hogback is followed by a narrow serir plain where ephemeral channels from the cuesta converge. Then a broadly convex and northwestward dipping rocky surface follows, the continuation of which is broken by another escarpment. The hamadas on the silicious Khuiyala Limestone and the Bandah Limestone have a number of palaeokarstic features, including shallow dolines and sink holes (Kar, 1989). Arora et al. (1992) suggested that the ferricreted surfaces in the latter area were formed due to sedimentation in a shallow lagoon or a similar low-lying terrain during the recession of the sea in Tertiary.

Past fluvial processes: Climatic and tectonic controls Construction of sandy alluvial plains was the result of sedimentation by integrated 230 Landforms Processes and Environment Management stream networks, many of which have become extinct now. Satellite images and aerial photographs show buried courses of several former streams which used to flow through the desert at different time during the early and middle Quaternary periods. Notable among these were the two presently extinct Himalayan streams, the Saraswati and the Drishadvati. Earlier, many scholars during the last one hundred years identified the Saraswati as the present dry bed of the Ghaggar along the northern margin of the Thar in Haryana and Rajasthan (Oldham, 1893; Ali, 1941). The Raini, the Wahinda and the Nara in Pakistan were identified as some of the shifted courses of that river. A dry stream bed between Hisar, Nohar and Bhadra was identified as the course of the Drishadvati, which is mentioned in ancient literature as a major tributary of the Saraswati. Satellite remote sensing by Ghose et al. (1979), Kar and Ghose (1984) and Kar (1986, 1993a) suggested that the early courses of these two Himalayan rivers were roughly through (1) the vicinity of Rajgarh, Hadyal, Ratangarh and the present misfit valley of the Jojri; (2) Nohar, Surjansar and Samrau; and (3) Sirsa, Lunkaransar and east of Bikaner. Subsequently, the Saraswati began to flow roughly through Nohar, Anupgarh, Sakhi (in India), Khangarh (in Pakistan), Ghantial, west of Shahgarh (in the extremely western part of Jaisalmer district in India) and then the lower courses of the Nara (in Pakistan). Further shifts took the Saraswati through the Raini and the Wahinda and the Hakra-Nara segment in Pakistan. Finally the river ceased to flow even through that course and met the Sutlej (mentioned as the Satadru in the early Indian literature) via Anupgarh to the west of Ahmadpur East (in Pakistan). The Drishadvati also gradually shifted northwestward, ultimately occupying the Narnaul- Hansi-Hisar-Bhadra-Nohar course to meet the Saraswati near Rangmahal. Vast alluvial plains were built by these streams, but their survival depended to a large extent on the contribution of the Sutlej in the sub-Himalayan plains, neotectonism and the fluctuating climate. The Luni system, originating from the Aravalli, once drained into this Himalayan system. The direction of shifts in the Saraswati-Drishadvati river system was roughly from east to west, so the above-mentioned northeast-southwest oriented courses could be the gradually shifted courses of the Saraswati system over a period of time. The configuration of a mapped palaeochannel between Tanot and Ghantiyalji in the difficult dune country in western part of Jaisalmer district (Fig. 2) was established through geophysical depth soundings and a potable aquifer was located (Kar and Shukla, 1993; Fig. 3). Subsequent field campaigns showed upstream extension of this potable aquifer between Tanot and Kishangarh (Singh et al., 1994). In the southern part of the Thar a number of palaeochannels of the Luni drainage system have been identified (Ghose, 1964; Ghosh, 1977; Pal, 1991; Kar, 1994, 1999a). Interpretation of a synthetic aperture radar imagery revealed some southwest-flowing to south-southwest-flowing palaeovalleys of the Luni in the alluvial plains between Jodhpur and Pali (Kar, 1999b; Fig. 4). The westernmost palaeovalley from near Kankani cuts across the present westward flow direction of two major tributaries of the Luni, the Guhiya and the Bandi, implying that the present channels are partly superimposed on the ancestral south-southwest-flowing channel of the Luni. The Quaternary Geomorphic Processes and Landform Development in the Thar Desert of Rajasthan 231

Figure 2. Palaeochannels of the Saraswati river in western part of Jaisalmer district

existence of the above palaeovalleys suggests that the Luni formerly used to flow through a number of easterly courses in the middle part of its present basin. The points of deflection from the present course were near Malkosni, Mortauka and Kankani. These possibly represent the successive shifted courses of the Luni. The trend of the SSW-oriented palaeovalley from Kankani to Sonai Lakha and beyond suggests that further south the river used to follow the tract between a set of two major NE-SW lineaments through Mandawas-Vayad (to the south of Garwara-Jetpur), and flow through the vicinity of Vayad, Nilkanth, Bhavrani and Balwara to the present course of the Jawai. Two lineaments tend to confine the SW-flowing course of the Jawai further south. In other words, this south-flowing early course of the Luni was partly structurally controlled. It also follows that the present course of the Luni from Kankani to Tilwara (where the river takes a sharp southward turn) and from there to the present confluence with the Jawai, is a recent one. It is likely that some other streams whose courses were subsequently occupied and modified by the Luni then drained the area. In the vicinity of Tilwara the south-flowing Lik river from Pokaran upland and beyond had a major contribution in the past (Kar, 1988a). Considering 232 Landforms Processes and Environment Management

Figure 3. Tanot-Ghantiyal geo-electric cross section

Figure 4. Present and palaeodrainage systems in the Luni-Bandi interfluves to the west of Bilara-Pali ( based on interpretation of radar imagery and field survey) Quaternary Geomorphic Processes and Landform Development in the Thar Desert of Rajasthan 233 that the straight line distance between the present south-flowing course of the Luni at Tilwara and the earlier SSW-flowing course at Kankani is more than 100 km, we may assume that either a very large flood or a major tectonic event provided the necessary energy to deflect the course of the Luni clockwise by about 40° from near Kankani, although validation of these premises will require further research. While several large tributaries of the Luni, originating from the Aravallis, used to flow for long distances and maintained their courses to the trunk stream, most of the smaller streams originating from the hills and rocky uplands within the desert used to flow for shorter distances. In the present <200 mm average annual rainfall zone, the Lik river, which is now a dying right bank tributary of the Luni, built up its own narrow flood plain and flowed southward beyond Tilwara (Kar, 1988a). The stream is now partly buried under high parabolic sand dunes. Another stream in the same dry belt originated in the uplands of Phalodi-Bap, flowed northward through the saline rann of Bap, the vicinity of Manchitiya, Naukh, Borana and Madasar, and ended up in its own fan deposits further north (Kar, 1986). Wide dry valleys, often cut across by thick sand sheets or high sand dunes, still mark the courses and carry water and sediments during extreme floods like those in 1990 and 1996. In the very dry norther part of Barmer district and adjoining areas of Jaisalmer district the passing of a deep tropical cyclone in August 2006 led to high intensity rainfall for three days (~ 300 mm). It partially excavated a hitherto unknown but possibly the longest erstwhile right-bank tributary of the Luni River from beneath a thick cover of aeolian sand sheets and advancing dune fronts, and flooded a mined gypsum playa where water stagnated for more than three years, causing loss of life and property. Despite a huge surge of water the flood flow was restricted to the buried course, even though it meant cutting through numerous dunes and other features (Kar et al., 2007; Kar, 2009). In Merta-Nagaur area in the east an upland formed dominantly of the Pre-Cambrian Bilara Limestone and Jodhpur Sandstone lies to the northwest of a major dry valley through Ren, Kheduli and Pundlu that could be linked to the proto-Saraswati system (Kar and Ghose, 1984). A number of giant dry valleys can be located on this upland from satellite imagery. These are the remnants of some major pluvial phases during the Pleistocene period. The valleys end abruptly at the margin of the surrounding sandy alluvial plains, and create enormous damages to life and property in the plains during spasmodic floods (as in 1996). This and a number of other low-level planes in the surroundings have loosely cemented gravel spreads. The linear arrangement of those gravel spread, a fining upward sequence in many of them and their spatial pattern suggest at least three generations of superimposed palaeochannels, sustained possibly by large, spasmodic flood discharges. Subsequent to the fluvial deposition the channel gravel were cemented. Relatively faster erosion in the surrounding alluvial plains then led to an inversion of relief, whereby the vestiges of the earlier channels are now left at a higher level than the alluvial surface built later on. Possibly, the lower alluvial surface was then explored by the proto-Saraswati system along the valley axis of Ren-Kheduli. An underfit channel now occupies this valley from Pundlu 234 Landforms Processes and Environment Management downstream and is sustained by seasonal flow through several smaller dry valleys from the Aravallis. There are other evidences of such inversion of relief due to lithification of channel bed material in the Thar. In the sandy plain between Jodhpur, Ratkuriya and Osian the channel gravel of former small ephemeral streams have been cemented as hard lithic calcretes through ground water mineralisation, and these calcretes are now left as higher surfaces due to selective erosion. Aeolian sediments occur both above and below many of these calcreted channel deposits. A number of other conglomerate beds in the Luni basin, especially between Pali and Sojat, and between Tilwara and Sindari bear testimony to early Quaternary fluvial episodes. One of these episodes was dated to ~80 kyr (Jain et al., 1999). It is likely that much of the alluviation took place before 40 kyr. Considering that periods of significantly higher rainfall during the last 150 kyr were approximately 120–130 kyr, 100–108 kyr and 75–85 kyr (as deduced from Prell and Kutzbach, 1987), stream activities during those periods might have been very high and of sustained nature to leave their imprints on the landscape. The transition from the Pleistocene to the Holocene was marked by increased precipitation. Palynological studies on lake sediments (Singh, 1971; Singh et al., 1974, 1990) suggest that the climate became gradually wetter from around 10 kyr. As we have mentioned earlier, the mean annual precipitation in 7.5–6.0 kyr was about twice the present value. The surviving streams of the period might have contributed significantly then to the process of alluviation. Ground water recharge to the buried palaeochannels was also increased. Geyh and Ploethner (1995) dated the ground water samples from one of the pre-Ghaggar Saraswati palaeochannels in the Cholistan desert of Pakistan and found that the last recharge of fresh ground water took place between 12.9 and 4.7 kyr (i.e. before pre-Harappan), with a break between 8 and 7 kyr BP. It is interesting to note that this and other recent dating of materials suggest that the Harappan and the pre-Harappan civilisations might have flourished under a decreasing rainfall regime. Studies by Singh et al. (1996), Rao and Kulkarni (1997) and Nair et al. (1999) on ground water along the palaeochannels of the Saraswati, which were identified earlier in the extreme southwestern part of Jaisalmer district (Ghose et al., 1979; Kar, 1986), confirm that the recharge to the aquifer did not take place from local rainfall. The rainfall in the desert showed a decreasing tendency from about 5 kyr (Singh et al., 1990). Tectonic activities during the Quaternary period played a significant role in the evolution of stream networks. Interpretation of satellite imagery of the Luni-Jawai plains indicated the presence of a number of lineaments across the plains. Striking relationship has been noticed between many of the NNE-SSW lineaments and drainage features like sudden branching and widening of channel bed (e.g. the Somesar, the Sukri, the Jawai and the Sagi), disappearance and reappearance of channels (e.g. the Mithri), and angular bends in the stream courses (e.g. the Sukri, the Ungti, the Bandi, the Jawai, etc.). Additionally, sudden incision of stream beds and formation of tributary gullies, or terracing along many streams like the Somesar, the Chhali, the Sukri, the Jawai, the Bandi and the Sagi, especially on the upstream side of the major lineaments, Quaternary Geomorphic Processes and Landform Development in the Thar Desert of Rajasthan 235 provide evidence for neotectonic movements. At least two terraces can be identified along many of the above streams, as well as along the Luni to the south of Sindari. The fall from the upper to the lower terrace is often between 1.5 and 2 m, while that from the lower terrace to the stream bed is between 1.0 and 1.5 m. Maximum terrace relief of 10 to 12 m has been noticed between Saila and Daman along the Jawai river (Kar, 1984, 1988b). Downstream of any major lineament the streams become braided and show signs of shifting till the next lineament across them favours downcutting and terracing (e.g. Bandi, Sukri, Somesar, Jawai). If we consider the areas separated by the major NNE-SSW lineaments as individual fault blocks, the pattern suggests that recent tectonic movements uplifted some of the blocks, effecting local base level changes along the streams, which cross them and helped in downcutting and terrace formation on the uplifted blocks. Unable to keep pace with the downcutting of the trunk streams the smaller tributaries on those blocks responded by the formation of gully networks. The adjacent blocks might have experienced relative down-faulting, so the streams on them responded by braiding and wider shifting of channels. Streams like the Ungti and the Mithri, which could not maintain their courses over these differently moving blocks were obliterated in parts and provided a segmented look. Since the amplitude of terrace relief and depth of stream dissection tend to increase from NNE to SSW, it may not be inappropriate to suggest that the lineaments represent some hinge faults which close to the northeast. The pattern of fluvial landform assemblages also suggest that the movements produced a horst and graben sequence across the Luni-Jawai plains, and possibly step faulting towards the west, the deepest segment lying especially between the lower reaches of the Jawai and the Luni in the west (roughly between Bhadwi, Jhab and Guda), and its linear extension further northeast through Balwara, Raithal, Motisara and Sanwarla (Kar, 1988b). Recent studies of subsurface lithofacies and Bouguer anomaly profiles confirm such possibilities (Bajpai et al., 1998). The satellite images taken after the flood of 1990 in Luni basin provide supportive evidence for neotectonically-controlled current channel processes. One of the best evidences was found in the Balwara-Debawas-Sanwarla sector where the ephemeral streams flowing eastward from the Siwana hills (e.g. the Luniwala and other smaller streams) and those flowing westward from the Aravallis (e.g. the Mithri and its distributary networks), were suddenly lost in their own microfans along a major lineament, in spite of the fact that the streams were in spate (Kar, 1994). All the major streams near the confluence of the Luni with the Jawai behaved abnormally while crossing a major NNE-SSW lineament and its subsidiaries. For example, the floodwater of the Sagi was divided through a number of channels while crossing a major NNE-SSW lineament near Dhani Goliya. Further north, the Bandi water was similarly divided when it negotiated another parallel NNE-SSW lineament. The Jawai widened its course near Pantheri where the river encountered this lineament. Downstream, near Dadal, the Jawai is characterised by anastomosing channel pattern, after it encounters another NNE-SSW lineament. The other notable controls were found in the case of the Luni to the south of Guda, where some of the former courses 236 Landforms Processes and Environment Management of the stream were partially revived and followed some of the major E-W lineaments through the vicinity of Punjaberi-Narsana-Kawatra, Piprala-Sewari, and Arniyali- Jhab-Dhani Goliya (Fig. 5). It revealed that the Luni once used to flow through a more westerly course in the area, and the course was partly controlled by E-W tectonic lineaments. A few other shifted courses of the Luni in Guda-Gandhav sector were identified by Pal (1991) from the signatures left after the flood of 1979. Satellite image interpretation also suggest that the major south-flowing palaeochannels of the Saraswati tended to follow some NNE-SSW lineaments, and are cut across by E-W lineaments. Some of these might have experienced neotectonism. Evidence of recent tectonic movements has also been found from the eastern fringe of the desert, especially around Sambhar lake and in the Kantli river basin (Dassarma, 1986, 1988).

Aeolian landform development Whenever the aeolian processes dominated over the fluvial processes during the dry periods, a set of new landforms was created over the existing fluvial landforms, especially as sand sheets, sandy hummocks and sand dunes. This happened not only in the Thar, but also in its northern fringe in Punjab and Haryana, southern fringe in Gujarat alluvial plains, as well as in the eastern fringe beyond the Aravalli hill ranges. Parts of the colluvial plains and older alluvial plains were transformed into sandy undulating plains. Sand dunes and interdune plains are the other major aeolian landforms. The saline depressions are the results of a complex interplay between the fluvial and the aeolian processes. Presently significant aeolian erosion of the rocky/gravelly tract occurs in the very dry Pokaran-Jaisalmer-Ramgarh area, producing new materials for the aeolian deposition. Some of the classic examples of fluting and grooving on hard rocks are noticed in the limestone terrain around Jaisalmer where the saltating, wind-driven quartz particles grind the softer limestone surfaces during sand storms and generate finer particles. Ventifacts abound on the rhyolite, sandstone and limestone pediments and on the gravelly pavements. Over the millennia enormous quantity of medium to fine sand and silt has been produced from the hill-pediment-depositional plain complexes of the region through such aeolian erosion and through fluvial activities. The material has been recycled several times to construct the aeolian landforms. The sand dunes of the Thar can be broadly classified into two groups: the old and the new (Pandey et al., 1964; Singh, 1982). Most high sand dunes in the Thar are the old dunes. These include the presently stabilised and vegetated linear, parabolic, transverse, star, network and major obstacle dunes (Fig. 6). The average height of these dunes varies between 15 and 30 m. The parabolic dunes cover the maximum area of the dune-covered landscape. The average windward, flank and leeward slopes are 2° to 4° , 8° to 12° and 22° to 24° , respectively. The dunes occur in chains and a three to four tier arrangement is noticed almost everywhere. Hair pin parabolics and chevron pattern occur in the western fringe of the field, where 3–4 km long arms join Quaternary Geomorphic Processes and Landform Development in the Thar Desert of Rajasthan 237 Channel pattern across lineaments after floods, 1990, in Bhinmal-Guda-Bhawatra area Figure 5. 238 Landforms Processes and Environment Management Thar desert sand dunes Figure 6. Quaternary Geomorphic Processes and Landform Development in the Thar Desert of Rajasthan 239 at a sharp angle. Towards the northeast the arms become shorter (0.5–1.0 km long) and the sharp junction at nose is replaced by a curved one. The upper middle and crestal parts of the dunes account for about 37 per cent of the total sand volume in the dune chains (Kar et al., 1998b; Fig. 7). Further northeast the parabolics disintegrate into two major kinds of network dunes: (a) compound hooked dunes with transverse forms at right, and (b) large parabolic forms with network within (Kar, 1993b). The coalescing of forms has so long been explained by either joining of the individuals by transverse ridge formation (Singh, 1982), or through movement and overriding of individuals at different rates and times (Wasson et al., 1983). Our studies in the interdune plains where new bedforms are currently forming suggest that sand accumulation downwind of a parabolic chain can occur as isolated depositional lobes which subsequently tend to align themselves in the form of two near parallel sand streaks. A major zone of accumulation occurs further downwind, astride the path of the wind. It captures more sand through deflation in between the pair of streaks and also through the supply from the dune chain upwind. Gradually the nosal part of the dune is formed here. The sand streaks do also grow simultaneously and form the dune arms. In other words, the parabolic bedform can develop here without the need of migration of sets of dunes and in a manner quite different from the growth sequence of the coastal parabolic dunes (Kar, 1993b). Another implication of the finding is that the interdune plains in the parabolic dunefield, when subjected to sand drift, may witness within a few years’ time, new parabolic bedforms, even though these could be of lower height. Since the nosal part of the parabolic dune is a zone of net sand deposition and since the sand surcharged wind that blows at a relatively higher speed through the constricted corridor between the two arms of the parabolic dune, the wind is expected to decelerate and drop its load once it is out of the constricted corridor. Hence, it is most unlikely that these vegetated inland parabolic dunes, when subjected to vigorous erosion, will lead to the truncation of the nose altogether, leaving a pair of linear sand ridges. Instead, the eroded sand load will be dropped in front of the nose, either as another curved segment, or as low linear arms of another new parabolic bedform (Kar, 1996a). The linear dunes occur mainly in the western part of the Thar, especially in Jaisalmer region, and are oriented in the direction of the wind. The dunes were earlier thought to have originated from parabolic dunes (Verstappen, 1968; Singh, 1982), but studies confirm their development from streams of barchans in the high wind energy zone to the west of 150 mm isohyet, and from lee vortices behind major obstructions, or along the major stream valleys through funnelling effect (Kar, 1987, 1990a). The dunes are characterised by a broad convex summit. The length varies from more than 10 km in the extreme west of the field to 1–2 km in the east. The high transverse dunes occur mainly to the west of Bikaner and were formed astride the path of sand-laden wind. The slopes of the leeward, flank and windward sides of the dunes are 22° , 10° –12° and 3° –4° , respectively. The average spacing between transverse dune chains is 300 to 800 m. Simple and compound obstacle dunes were formed on the windward and leeward sides of hills. The dunes are highly dissected 240 Landforms Processes and Environment Management

Figure 7. Form characteristics of a parabolic dune, Thar desert (based on GPS measurements) by rills and gullies. A number of fossil dunes occur along the wetter eastern margin of the Thar, especially between Rohtak, Sultanpur, Bandikui and Lalsot in the northeast and Idar-Langhnaj in the south. These areas now receive more than 550 mm of average annual rainfall, but were affected by aridity during the last glacial period. The dunes define the maximum eastern limit of the past aeolian activities during the dry phases (Goudie et al., 1973). In contrast to the old dunes a number of new dunes are being formed presently in the desert. Under the natural set up the dunes are formed in high wind energy regime of the west, especially as 1 to 8 m high barchans and 20 to 40 m high megabarchanoids. These are crescentic in shape. Smaller barchans move fast. Near Pokaran the barchans move at an average speed of 30 m per year (Kar, 1998). The brink line of the dune is Quaternary Geomorphic Processes and Landform Development in the Thar Desert of Rajasthan 241 usually located downwind of the crest. When the dune is in its formative stages the crest line and the brink are almost at the same place, but as the barchan captures more sand from the upwind direction, the distance between the brink line and the crest increases. The simple barchanoids are first noticed near the 120 contour of wind erosion index, where they are usually 1–3 m high and are formed in dry channels or in other flat sandy terrain with a thick deposit of loose sand. In the western part the barchanoids are 15–20 m high, but are not completely devoid of vegetation. Each dune consists of a high arcuate segment in its upwind part which is partly vegetated; a field of disjointed chains of low barchanoids which occur downwind of the arcuate segment and across the path of the dominant wind in the direction of wind. The characteristic horns of the barchans are missing in these dunes (Fig. 8). Compound megabarchanoids occur in areas where the index value is 480 or more (i.e. extremely high category). The pattern and development of bedforms within these fields of large crescentic dunes suggest that these are the zones where sand is collected from the surroundings by wind, processed and then released along linear strips, either along the crests of the old linear dunes, or as narrow zones of barchans (Kar, 1990a). Further downwind the barchan strips themselves are coalesced along their path of movement to ultimately grow into linear dunes. In the eastern and northern parts of the desert wind strength is not sufficient now to form these dunes under natural conditions. However, at places where the natural stability of the old aeolian landforms have been disturbed by human activities, especially around the settlements and in deep-ploughed agricultural fields, localised colonies of barchans can be noticed. In Ganganagar-Hanumangarh-Rawatsar area such low barchans are numerous in the alluvial plains. Their formation is related to destruction of the erstwhile low network dunes, followed by land levelling for irrigated agriculture. The dunes mostly occur at the sites of the old, low dunes in the sandy plains, which are at a slightly higher elevation than the adjoining alluvial plains. Mobility of the dunes is very less. Sandy undulating plains are characterised by sand sheets of 50 to 300 cm thickness, as well as low sand streaks and shrub coppice dunes (nebkhas). The height of such hummocks seldom exceeds 5 m. The average slope of the sandy undulating plains varies from 1° to 3° . Like the crescentic dunes the low sandy undulations and loose sand sheets are of recent origin. In many cases these are associated with high human activities.

Past aeolian processes For better appreciation of the aeolian processes in the past, we shall first provide a short overview of the current processes and their manifestations on the sandy terrain. Presently there is a distinct rainfall gradient in the Thar from east to west. The efficiency of aeolian processes increases with decreasing rainfall from east to west, as well as with the increasing wind speed in that direction. The strength of the summer monsoon wind (SW wind) between March and July counts, rather than the feeble winter wind (NE wind). In normal years, aeolian processes are most efficient between 242 Landforms Processes and Environment Management

Figure 8. Morphology of a recently forming barkhanoid chain to the northeast of Ramgarh (based on aerial photo interpretation)

March and June, or early July. The maximum aeolian activity and development of new aeolian bedforms without human interventions takes place mostly in the western part of the desert (Kar, 1993b, 1998). Wind erodibility pattern (based on sediment character and vegetation) and human pressures determine the areas of localised acceleration to aeolian processes in the eastern fringe areas (Kar, 1996a). Given the past spatial patterns of monsoon distribution in the region, as derived from different simulations, it is likely that the pattern of rainfall gradient seen at present, was also present during similar climatic phases in the past. With the change in climate the isohyets were shifted across the longitudes, but the pattern of rainfall variation remained almost the same across the desert. Wind speed, especially the SW monsoon wind speed, also pulsated with the monsoon. Therefore, the areas under different categories of wind erosion fluctuated over the time, but the broad pattern Quaternary Geomorphic Processes and Landform Development in the Thar Desert of Rajasthan 243 remained almost the same. Aeolian stratigraphic records suggest that winter winds were not a major factor of dune building in the past. Additionally, prolonged dry weather periods, a possible reduction in vegetation type and cover, and a higher albedo favoured increased aeolian activity. Earlier studies suggested that the major dry phases favoured aeolian processes, and implied that aridity was a major factor of dune building. More recent studies, involving luminescence dating of the sediments, reveal that aeolian processes were pronounced for shorter time windows during the dry phases, and that maximum sand accretion took place during the periods when monsoon wind started building up (Chawla et al., 1992; Kar et al., 1998a). LGM was possibly the time of much less aeolian activity, because the wind strength was very less effective. Studies on oceanic cores suggest that the periods of sustained higher rainfall might have lagged the inception of higher wind speed by years or centuries. This is analogous to the synoptic conditions, which we notice in the desert at present. The SW wind peaks up from March onward, and the maximum speed in the central and western parts is obtained in June-July. The rain starts by the end of June, or early July. The wind falls from August, and by September the monsoon recedes. The wind, therefore, gets a small period between March and early July to move the sand and build the dunes. More sand shifting takes place in the years when the summer wind is relatively stronger and is preceded by a series of drought years, which reduce the vegetation cover. Possibly the pattern remained so in the past also; only the relative strength of SW wind changed. If we consider such a scenario at century and millennium scales in the past, the wind got fewer periods to build up the landforms. The opportunities occurred especially during the transition from a dry climatic phase to the better monsoons and during the periods of drought within those better monsoon phases. Such periods were dominated by stronger monsoon wind but rainfall amount was meager, a scenario analogous to that of the severe drought of 1984-87 in the Thar. The view that the aeolian processes were more efficient during the periods of stronger SW monsoon gets support from the trend of the old dunes and the modern wind (Kar, 1993b). Most high sand dunes in the Thar were last formed between 11 kyr and 18 kyr. It is customary to suggest that linear, parabolic and transverse dune types are the vestiges of a past dry climate, but as we have shown in the preceding section, these dunes are forming even now and under different geomorphic settings. If the present is a key to the past then the mode of formation of the present dunes should provide clues to the past processes. Systematic study of a number of deep aeolian sand-dominated sedimentary sequences across the desert and its wetter fringe and their luminescence dating are now helping to reconstruct the past aeolian histories. The growth of one lee side linear dune, behind a quartzite hill near Didwana, has been sequentially dated near its keel from about 200 kyr to about 3 kyr BP (Wasson et al., 1983; Dhir et al., 1994). Several aeolian units were also dated to around 100–130, 65–75 and 30–55 kyr. The time brackets are approximate, and mark the periods of enhanced sand mobilisation (Dhir et al., 2010). The last 20 kyr saw at least three major periods of sand accretion (Singhvi and Kar, 1992, 2004; Kar et al., 1998a, 2001; Thomas et 244 Landforms Processes and Environment Management al., 1999). The last major sand accretion took place from 2 to 0.5 kyr in the west. In the east it took place from 2 to ~0.7 kyr, especially at some favoured locations. Most parts of the eastern fringe area was, however, almost free from sand accretion after ~3 kyr (Singhvi et al., 1994). Studies on a transverse dune in the less than 200 mm rainfall zone revealed that the dune crest advanced by 0.3 to 0.9 m per year during the period 2 kyr–0.5 kyr, but now the rates of dune mobility and sand accretion are at least four-fold higher than the geological rates, possibly because of high human and cattle pressures (Kar et al., 1998a).

Formation of saline depressions The saline depressions with ephemeral lakes (or the ranns) are a significant component of the landscape assemblage in the Thar desert. Some of the major ranns occur near Didwana, Tal Chhapar, Pachpadra, Thob, Bap, Kanod, Jamsar and Lunkaransar. The ranns are characterised by alternate layers of silt-clay and sand dominated layers, as well as by gypsum-rich layers (Singh et al., 1974; Rai, 1990; Kajale and Deotare, 1993). Studies in the Bap rann suggest that sedimentation began there during the terminal Pleistocene (Deotare et al., 1998), but the other ranns, like those at Sambhar, Didwana and Pachpadra, may hold older sediments also. Early hypotheses on their formation like aeolian transport of salts from the Rann of Kachchh (Holland and Christie, 1909) or recession of the Tethys sea (Godbole, 1952), have been disproved by later studies (Ramesh et al., 1993). Aggarwal (1957) first emphasised a riverine connection for the lakes at Pachpadra and Didwana. Ghose (1964) suggested salt deposition at the confluence of streams as the cause of their formation. Recent studies (Kar, 1990b) revealed that many lakes in the eastern part of the desert lie at the wind shadow zone of major topographical barriers like hills with associated linear dunes, which favoured strong deflation and created the hollowed basins (e.g. Degana, Didwana, Tal Chhapar, Parihara). In some cases, like that at Degana, courses of former streams which used to flow away from the deflation zone, were obstructed by the linear dunes which were advancing from the flanks of the hill, and were guided to the basin. Such events initially create a fresh water lake, but with time it turns into a saline rann (Fig. 9). Similar process-form interaction was noticed near Jaidu in western Thar. Blockage of ephemeral stream valleys by advancing sand dunes, followed by accumulation of water and salt, are also important factors of rann formation. In the Jaldhari-Deunga tract near Mohangarh the Jaisalmer Limestone beds, dipping at 2° to 5° northwards, are now being etched and grooved continuously by the sand-surcharged SW wind. This created a number of small, wind- aligned depressions along the dip slope, separated by micro-escarpments (Fig. 10). Enlargement of the depressions and breaching of the micro-escarpments will create an elongated depression in the rocky topography, parallel to the Kanodwala Rann, and then a new rann. It will be interesting to know if the evolution of the Mitha Rann, the Khara Rann, the Kanodwala Rann and the Kharariwala Rann in the vicinity was partly controlled by such wind erosion. The ranns are partly oriented in the Quaternary Geomorphic Processes and Landform Development in the Thar Desert of Rajasthan 245 Stream capture by advancing sand dunes and deflation in wake of hills to form lakes near Degana Figure 9. 246 Landforms Processes and Environment Management Wind-scoured depressions to the east of Kanodwala rann. Figure 10. Quaternary Geomorphic Processes and Landform Development in the Thar Desert of Rajasthan 247 direction of the dominant SW wind, although a dry valley in the rocky terrain also links these. Erosion along zones of weakness along a supposedly SW-NE trending lineament and possible groundwater-related weathering could be the associated processes of rann formation here. A number of other factors, including tectonic disturbances, were also responsible for the formation of many other ranns. Tectonism appears to have played a crucial role in the development of Sambhar lake, while a complex process involving disruption of fluvial regime by neotectonic activities, as well as aeolian activities might have aided the formation of a number of smaller ranns, like at Sanwarla, Nilkanth, Nosra and Daman in Luni-Jawai plains (Kar, 1988b, 1995).

Other processes A slow process of rock weathering is continuing throughout the geological periods, and is preparing material for the subaerial erosional processes. Forms like weathering pits, honeycombs, spheroidal features, tafonis and alveolar caverns in sandstone, granite and rhyolite are related to moisture availability that, surprisingly, is adequate even in the low rainfall areas of the desert. Desert varnish occurs in the average annual rainfall belt of 180–250 mm, and is possibly related to some microbial activities. Case hardening and box weathering features in sandstone and limestone are noticed in the less than 250 mm rainfall zone. These latter features, although common on smaller outcrops, have also been noticed at giant, landscape scales in Jaisalmer- Ramgarh area. At places it has led to the formation of ferricreted surfaces of considerable size. Splitting of boulders, granular disintegration and other forms of mechanical weathering are common in the drier parts of the desert. In the gravelly pavements to the southwest of Pokaran gravels of 2 to 30 cm length abound on the surface and occur in a matrix of fine sand and silt-sised particles. Larger boulders are also occasionally encountered. Numerous cleaved gravels/ boulders and the aerodynamically formed single or multi-faceted ventifacts on their surfaces bear testimony to a long period of insolation weathering and wind erosion, with little down-slope transportation. The profiles usually show a near-surface concentration of the gravel and other coarse particles, followed down the profile by sand and silt- rich sediments, and then the parent conglomeratic material. Such near-surface concentration of coarser fragments has been noticed in many other deserts, including the deserts of Arabia and Australia. While a part of it can be explained as a lag gravel, related to the winnowing of the finer top sediments, the lack of a gradational contact between the surface gravels and the parent formation below suggests that some other processes like long cycles of wetting and drying of the fine sediments and consequent forcing upwards of the gravels through desiccation cracks could be involved. In some areas the large gravels tend to be sorted along polygonal cracks. This mechanism is akin to the formation of stony gilgai topography in the Australian desert (Mabbut, 1977). Carbonate enrichment of sediments, especially in the subsurface, is a major process. 248 Landforms Processes and Environment Management

It produces a variable thickness of calcic horizon with powdery to nodular carbonates (kankar). Such pedogenic carbonates are common in the semi-arid areas also, but become less frequent with higher rainfall. Surface induration of carbonates as thin patinas occur over the rocky pediments, which seal the joints and fractures, and enhance run-off. A thicker induration of calcrete, partly covered by later aeolian sand sheets, is widespread in the western part of the desert in Jaisalmer-Barmer area where the vast pedeplains on rhyolite, sandstone and limestone have such induration. Stratigraphic records from the desert show thick calcretes involving sand, alluvium and conglomerate. Some of these are groundwater calcretes, but some others could have a complicated history (Dhir, 1995; Achyuthan and Rajaguru, 1998). With induration and hardening the deposits put resistance to subsequent erosion, as compared to the surrounding sandy plains. Ultimately the process creates an inversion of relief. Typical examples of such inversion of relief due to calcretisation of fluvial deposits are noticed in Kherapa-Anwana-Nandiyan Khurd area to the east of Jodhpur, where a number of 2–5 m thick, massive cream to buff coloured calcrete bands in sinuous pattern define the small palaeochannels which used to drain earlier into a SSW-flowing ephemeral stream, the Jojri.

Man as an agent of landform development During the last half of the Twentieth Century Thar desert has witnessed tremendous rise in population and the uses of land and its resources, especially for agriculture. There has been a four-fold increase in human population between the 1961 and the 1991 census. The present density is 84 km–2. There has also been a significant rise in the livestock population. To cater to the needs of the increasing human and livestock populations, pressure on the existing resources of the land is increasing. During the last two decades use of tractor for ploughing on dune slopes and other sandy/shallow marginal lands have spread even into the dry and more windy western part where the wind erosion index is more than 120 (very high). Overgrazing and other forms of vegetation destruction are continuing. Ground water is being exploited on a wider scale, without much concern for the quality of the water. A giant irrigation scheme, the Indira Gandhi Nahar Project (IGNP), has come up in the western part of the desert. It has transformed the agricultural scenario in the region, but unregulated flood irrigation, land levelling in the dune landscape in the hope of good supply of canal water, and other forms of mismanagement of land are continuing. As we have noted earlier, the landforms in the region are in a very fragile state due to the prevailing climate and terrain characteristics. Consequently, such activities are accelerating the natural geomorphic processes to such a degree that the human-induced rates of mobility of sediments from the fluvial and aeolian landforms are more than the normal rates, and there are more spread of waterlogging and salinity-alkalinity (Kar, 1996b). For example, many of the erstwhile ‘stabilised’ old dunes have become reactivated and have a thick cover of recent sand, which is advancing. New smaller dunes, especially barchans, are also forming in most parts of the sandy plains where none Quaternary Geomorphic Processes and Landform Development in the Thar Desert of Rajasthan 249 existed earlier. Gully erosion is increasing in the sandy plains of Sikar area. In the canal-dominated Suratgarh-Rawatsar-Hanumangarh-Pilibangan tract waterlogging and salinity-alkalinity have engulfed large areas, especially along a maze of palaeochannels. Some of these trends can be halted or reversed if the engineering structures and associated land use alternatives are suggested on the basis of a sound understanding of land-forming processes and vulnerability of the terrain to induced pressures.

References Achyuthan, H. and Rajaguru, S.N. 1998. Micromorphology of Quaternary calcretes around Didwana in Thar desert of Rajasthan, Annals of Arid Zone, 37: 25–35. Aggarwal, S.C. 1957. Pachbhadra and Didwana Salt Sources, Govt. of India Press, Delhi, 365p. Ali, S.M. 1941. The problem of desiccation of the Ghaggar plain, The Indian Geographical Journal, 16: 166–178. Allchin, B., Goudie, A. and Hegde, K.T.M. 1978. The Prehistory and Palaeogeography of the Great Indian Desert, Academic Press, London, 370 p. Arora, R., Rajaguru, S.N. and Ghate, S. 1992. Ferricretes in district Jaisalmer, Rajasthan. (Abstracts) Abstracts Volume, International Symposium on Evolution of Deserts, Physical Research Laboratory, Ahmedabad: 238–239. Bajpai, V.N., Kumar, T. and Tandon, S.K. 1998. Subsurface sediment accumulation patterns and their relationships with tectonic lineaments in Luni river basin, Rajasthan (Abstracts) Abstracts Volume, IGCP (349/413) Workshop on Palaeoclimatic Records on Arid and Semi-arid Regions during the Late Quaternary, Deccan College, Pune: 3. Bryson, R.A. 1989. Modelling the NW India monsoon for the last 40000 years, Climate Dynamics, 3: 169–177. Bryson, R.A. and Baerries, D.A. 1967. Possibilities of major climatic modifications and their implications: Northwest India: A case for study, Bulletin of the American Meteorological Society, 43(3): 136–142. Chatterji, P.C. 1977. Geological history of the northwestern arid zone of India. In: P.L. Jaiswal (editors), Desertification and its Control, Indian Council of Agricultural Research, New Delhi: 58–68. Chawla, S., Dhir, R.P. and Singhvi, A.K. 1992. Thermoluminescence chronology of sand profiles in the Thar desert and their implications, Quaternary Science Review, 11: 25– 32. Dasgupta, S.K. and Chandra, M. 1978. Tectonic elements of west Rajasthan shelf and their stratigraphy, Quarterly Journal of the Geological, Mineralogical and Metallurgical Society of India, 50: 1–16. Dassarma, D.C. 1986. Neotectonism in Rajasthan: Its manifestations and effects. Proceedings, International Symposium on Neotectonics in South Asia, Survey of India, Dehra Dun: 282–288. Dassarma, D.C. 1988. Post-orogenic deformation of the Precambrian crust in northeast Rajasthan. In: Roy, A.B. (editor), Precambrian of the Aravalli Mountain, Rajasthan, India, Geological Society of India, Bangalore: 109–120. Deotare, B.C., Kajale, M.D., Kshirsagar, A.A. and Rajaguru, S.N. 1998. Geoarchaeology 250 Landforms Processes and Environment Management

and palaeoenvironmental studies around Bap-Malar playa, district Jodhpur, Rajasthan, Current Science, 75: 316–320. de Noblet, N., Braconnot, P., Joussaume, S. and Masson, V. 1996. Sensitivity of simulated Asian and African summer monsoons to orbitally induced variations in insolation 126, 115 and 6 kBP. Climate Dynamics, 12: 589–603. Dhir, R.P. 1995. The genesis and distribution of arid zone calcretes, Memoir, Geological Society of India, 32: 168–190. Dhir, R.P., Rajaguru, S.N. and Singhvi, A.K. 1994. Desert Quaternary formations and their morphostratigraphy: Implications for the evolutionary history of the Thar, Journal of the Geological Society of India, 43: 435–447. Dhir, R.P., Singhvi, A.K., Andrews, J. E., Kar, A., Sareen, B. K., Tandon, S. K., Kailath, A. and Thomas, J. V. 2010. Multiple episodes of aggradation and calcrete formation in Late Quaternary aeolian sands, central Thar Desert, Rajasthan, India. Journal of Asian Earth Sciences, 37: 10–16. Geyh, M. and Ploethner, D. 1995. An applied paleohydrological study in Cholistan, Thar desert, Pakistan. In: Adar, E.M. and Leibundgut, C. (editors) Applications of Tracers in Arid Zone Hydrology, IAHS Publication 232, Institute of Hydrology, Wallingford: 119– 127. Ghose, B. 1964. Geomorphological aspects of the formation of salt basins in western Rajasthan. Proceedings, Symposium on Problems of Indian Arid Zone, Central Arid Zone Research Institute, Jodhpur: 79–83. Ghose, B., Singh, S. and Kar, A. 1977a, Geomorphology of Rajasthan desert. In: Jaiswal, P.L. (editor), Desertification and its Control, ICAR, New Delhi: 69–76. Ghose, B., Singh, S. and Kar, A. 1977b. Desertification around the Thar—A geomorphological interpretation, Annals of Arid Zone, 16: 290–301. Ghose, B., Kar, A. and Husain, Z. 1979. The lost courses of the Saraswati river in the Great Indian Desert—New evidence from Landsat imagery. The Geographical Journal, 145: 446–451. Ghosh, R.N. 1977. Photogeological studies on ancient water regimes of Rajasthan rivers. In, Agrawal, D.P. and Pande, B.M. (editors), Ecology and Archaeology of Western India, Concept Pub. Co., Delhi: 157–166. Godbole, N.N. 1952. The salinity of Sambhar lake, Bulletin of the National Institute of Sciences of India, 1: 89–93. Goudie, A.S., Allchin, B. and Hegde, K.T.M. 1973. The former extensions of the Great Indian Sand Desert, The Geographical Journal, 139: 243–257. Heron, A.M. 1953. The Geology of Central Rajputana, Memoir, Geological Survey of India, 79: 389. Holland, T.H. and Christie, W.A.K. 1909. The origin of the salt deposits in Rajputana, Records, Geological Survey of India, 38: 154–186. Jain, M., Tandon, S.K., Bhatt, S.C. and Singhvi, A.K. 1999. Alluvial and aeolian sequences along the river Luni, Barmer district: Physical stratigraphy and feasibility of luminescence chronology methods, Memoir, Geological Society of India, 42: 273–295, Bangalore. Kajale, M.D. and Deotare, B.C. 1993. Field observations and lithostratigraphy of three salt lake deposits in Indian desert of western Rajasthan, Bulletin of the Deccan College Post-Graduate & Research Institute, 53: 117–134. Quaternary Geomorphic Processes and Landform Development in the Thar Desert of Rajasthan 251

Kar, A. 1984. Assumptions regarding former stream behaviour and neotectonism from fluvial terraces along the Somesar river in western Rajasthan, India, Annals of Arid Zone, 23: 161–170. Kar, A. 1986. Remote sensing of buried former streams in the extremely arid terrain of Jaisalmer, India, for water and salinity, Proceedings, Seventh Asian Remote Sensing Conference, Seoul: B2/1–9. Kar, A. 1987. Origin and transformation of longitudinal sand dunes in the Indian desert, Zeitschrift fur Geomorphologie, 31: 311–337. Kar, A. 1988a. The Lik river of the Rajasthan desert: Its source, courses and present condition. In: Tewari, A.K. (editor) Desertification: Monitoring and Control, Scientific Publishers, Jodhpur: 249–255. Kar, A. 1988b. Possible neotectonic activities in the Luni-Jawai plains, Rajasthan, Journal of the Geological Society of India, 32: 522–526. Kar, A. 1989. Terrain characteristics of Jaisalmer district, Geographical Review of India, 51: 48–59. Kar, A. 1990a. The megabarchanoids of the Thar: Their environment, morphology and relationship with longitudinal dunes, The Geographical Journal, 156: 51–61. Kar, A. 1990b. A stream trap hypothesis for the evolution of some saline lakes in the Indian desert, Zeitschrift fur Geomorphologie, 34: 37–47. Kar, A. 1992. Geomorphology of the Thar desert in Rajasthan. In: Sharma, H.S. and Sharma, M.L. (editors) Geographical Facets of Rajasthan, Kuldeep Publications, Ajmer M.L.: 298–314. Kar, A. 1993a. Drainage desiccation, water erosion and desertification in northwest India. In: Sen, A.K. and Kar, A. (editors), Desertification and its Control in the Thar, Sahara and Sahel Regions. Scientific Publishers, Jodhpur: 49–72. Kar, A. 1993b. Aeolian processes and bedforms in the Thar desert, Journal of Arid Environments, 25: 83–96. Kar, A. 1994. Lineament control on channel behaviour during the 1990 flood in the southeastern Thar desert, International Journal of Remote Sensing, 15: 2521–2530. Kar, A. 1995. Geomorphology of arid western India, Memoir, Geological Society of India, 32: 168–190. Kar, A. 1996a. Morphology and evolution of sand dunes in the Thar desert as key to sand control measures, Indian Journal of Geomorphology, 1: 177–206. Kar, A. 1996b. Desertification processes in arid western India. In: Tadakuni Miyazaki and Atsushi Tsunekawa (editors) Towards Solving the Global Desertification Problem, 4, National Institute for Environmental Studies, Tsukuba: 20–29. Kar, A. 1998. Present-day mobile crescentic dunes in the Thar desert, India. In: Alsharhan, A.S. Glennie, K.W. Whittle, G.L. and Kendall, C.G.St.C. (editors), Quaternary Deserts and Climatic Change. A.A. Balkema, Rotterdam and Brookfield: 155–164. Kar, A. 1999a. Neotectonic and climatic controls on drainage evolution in the Thar desert. In: B.S. Paliwal (editor), Geological Evolution of Northwest India, Scientific Publishers, Jodhpur. 246–259. Kar, A. 1999b. Discovery of a palaeodrainage system from the radar imagery of southeastern Thar desert and its significance, Memoir, Geological Society of India, 42: 229–235. 252 Landforms Processes and Environment Management

Kar, A. 2009. An analysis of climate change through quantification of forms and processes in the Thar Desert. In Kumar, A., Kushwaha, R. A., Thakur, B. (eds.), Earth System Sciences, Concept Publishing Co., New Delhi. 2: 363–381 Kar, A. and Ghose, B. 1984. Drishadvati river system in India: An assessment and new findings, The Geographical Journal, 150: 221–229. Kar, A. and Shukla, J.P. 1993. Ground Water Exploration along the Saraswati-Drishadvati Rivers System in Western Rajasthan: Kishangarh-Ghantiyal Tract, Report (1992-93). CAZRI and GWD, Jodhpur: 15p. Kar, A., Felix, C., Rajaguru, S.N. and Singhvi, A.K. 1998a. Late Holocene growth and mobility of a transverse dune in the Thar desert, Journal of Arid Environments, 38: 175–185. Kar, A., Moharana, P. C., Singh, S. K., Goyal, R. K. and Rao, A. S. 2007. Barmer flood, 2006: causes and consequences. In Faroda, A. S. and Joshi, D. C. (eds.), Flood of August 2006 in Arid Rajasthan: Causes, Magnitude and Strategies. State of Art Report 29. INCOH, Roorkee. 26–39. Kar, A., Singhvi, A. K., Rajaguru, S. N., Juyal, N., Thomas, J. V., Banerjee, D. and Dhir, R. P. 2001. Reconstruction of the Late Quaternary Environment of the Lower Luni Plains, Thar Desert, India. Journal of Quaternary Sciences, 16 (1): 61–68. Kar, A., Singhvi, A. K., Juyal, N.and Rajaguru, S. N. 2004. Late Quaternary aeolian sedimentation history of Thar Desert. In Sharma, H. S. Singh, S. and De, S. (eds.), Geomorphology and Environment. ACB Publications , Kolkata. 105–122. Kar, A., Tsunekawa, A. and Miyazaki, T. 1998b. Potentiality of global positioning system in sand dune measurement: A case study from the Thar desert, India. In: Alsharhan, A.S., Glennie, K.W., Whittle, G.L., and Kendall, C.G.St.C. (eds.) Quaternary Deserts and Climatic Change, A.A. Balkema, Rotterdam and Brookfield: 433–438. Mabbutt, J.A. 1977. Desert Landforms, Australian University Press, Canberra: 340p. Misra, V.N., Rajaguru, S.N., Agrawal, D.P., Thomas, P.K., Husain, Z. and Dutta, P.S. 1980. Prehistory and palaeoenvironment of Jayal, western Rajasthan, Man and Environment, 4: 19–31. Nair, A.R., Navada, S.V. and Rao, S.M. 1997. Isotope study to investigate the origin and age of groundwater along palaeochannels in Jaisalmer and Ganganagar districts of Rajasthan, Memoir, Geological Society of India, 42: 315–319. Narayanan, K. 1964. Stratigraphy of the Rajasthan shelf, Proceedings, Symposium on Problems of Indian Arid Zones. Government of India, UNESCO and CAZRI, Jodhpur: 92–100. Oldham, C.F. 1893. The Saraswati and the lost river of the Indian desert. Journal of the Royal Asiatic Society, 34: 49–76. Overpeck, J., Anderson, D., Trumbore, S. and Prell, W. 1996. The southwest Indian monsoon over the last 18000 years. Climate Dynamics, 12: 213–225. Pal, G.N. 1991. Quaternary landscape and morphostratigraphy in the lower reaches of the Luni basin. In: Desai, N., Ganapathi, S. and Patel, P.K. (editors), National Seminar on Quaternary Landscape of Indian Subcontinent, Department of Geology, University of Baroda, Vadodara: 79–90. Pandey, S., Singh, S. and Ghose, B. 1964. Orientation, distribution and origin of sand dunes in the central Luni basin, Proceedings, Symposium on Problems of Indian Arid Quaternary Geomorphic Processes and Landform Development in the Thar Desert of Rajasthan 253

Zone. CAZRI, Jodhpur: 84–91. Pant, G.B. and Hingane, L.S. 1988. Climatic changes in and around the Rajasthan desert during the 20th century, Journal of Climatology, 8: 391–401. Pareek, H.S. 1984. Pre-Quaternary Geology and Mineral Resources of north-western Rajasthan, Memoir, Geological Survey of India, 115: 99 p. Prell, W.L. and Kutzbach, J.E. 1987. Monsoon variability over the past 150,000 years, Journal of Geophysical Research, 92 (D7): 8411–8425. Raghav, K.S. 1992. Quaternary history of a part of the northeast fringe of the Thar desert of India. Annals of Arid Zone, 31: 1–8. Rai, V. 1990. Facies analysis and depositional environment of Pokaran saline rann, district Jaisalmer, Rajasthan, India, Journal of the Geological Society of India, 36: 317–322. Rakshit, P. and Sundaram, R.M. 1998. Calcrete and gypsum crusts of the Thar desert, Rajasthan: Their geomorphic locales and use as palaeoclimatic indicators, Journal of the Geological Society of India, 51: 249–255. Ramesh, R., Jani, R.A. and Bhusan, R. 1993. Stable isotopic evidence for the origin of salt lakes in the Thar desert, Journal of Arid Environments, 25: 117–123. Rao, S.M. and Kulkarni, K.M. 1997. Isotope hydrology studies on water resources in western Rajasthan, Current Science, 72: 55–61. Roy, A.B. (ed.). 1988. Precambrian of the Aravalli Mountain, Rajasthan, India. Geological Society of India, Bangalore, 439 p. Sen, S. 1970. Some problems of Precambrian geology of the central and southern Aravalli range, Rajasthan, Journal of the Geological Society of India, 11: 217–231. Singh, A.P., Babel, M.K. and Soni, V. 1994. A Report on Geophysical Resistivity Exploration to Trace the Path of Buried Saraswati River and its Tributaries, District Jaisalmer. Ground Water Department, Jodhpur: 14p. Singh, D., Singh, Y.P., Bairwa, S.P., Dass, P., Porwal, C.P. and Mathur, K.M. 1996. Isotope study of water resources in arid areas of Jaisalmer, Annals of Arid Zone, 35: 313–317. Singh, G. 1971. The Indus Valley culture seen in the context of post-glacial climatic and ecological studies in north-west India, Archaeology and Physical Anthropology in Oceania, 6(2): 177–189. Singh, G., Joshi, R.D., Chopra, S.K. and Singh, A.B. 1974. Late Quaternary history of vegetation and climate of the Rajasthan desert, India, Philosophical Transactions of the Royal Society, 267B: 467–501. Singh, G., Wasson, R.J. and Agrawal, D.P. 1990. Vegetational and seasonal climatic changes since the last full glacial in the Thar desert, northwestern India, Review of Palaeobotany and Palynology, 64: 351–358. Singh, S. 1982. Types and formation of sand dunes in Rajasthan desert. In: Sharma, H.S. (ed.), Perspectives in Geomorphology. Concept Pub. New Delhi: 165–183. Singh, S., Kar, A. and Vats, P.C. 1997. Landforms of the arid ecosystem of western India. In: Singh, S. and Kar, A. (editors) Desertification Control in the Arid Ecosystem of India for Sustainable Development, Agro Botanic Publishers (India), Bikaner: 36–45. Singhvi, A.K. and Kar, A. (eds.) 1992. Thar Desert in Rajasthan. Geological Society of India, Bangalore: 191 p. Singhvi, A.K. and Kar, A. 2004. The aeolian sedimentation record of the Thar Desert. Proceedings, Indian Academy of Sciences (Earth and Planetary Science), 113 (3): 371–401. 254 Landforms Processes and Environment Management

Singhvi, A.K. and Krbetschek, M.R. 1996. Luminescence dating: A review and a perspective for arid zone sediments. Annals of Arid Zone, 35: 249–279. Singhvi, A.K., Banerjee, D., Rajaguru, S.N. and Kishan Kumar, V.S. 1994. Luminescence chronology of a fossil dune at Budha Pushkar, Thar desert: Palaeoenvironmental and archaeological implications, Current Science, 66: 770–773. Sinha Roy, S., Malhotra, G. and Mohanty, M. 1998. Geology of Rajasthan. Geological Society of India, Bangalore: 278 p. Sirocko, F. 1991. Deep-sea sediments of the Arabian Sea: A paleoclimatic record of the Southwest-Asian summer monsoon, Geologische Rundschau, 80(3): 557–566. Sirocko, F., Sarnthein, M., Erienkeuser, Lange, H., Arnold, M. and Duplessy, J.C. 1993. Century-scale events in monsoonal climate over the past 24,000 years, Nature, 364: 322–324. Thomas, J.V., Kar, A., Kailath, A.J., Juyal, N., Rajaguru, S.N. and Singhvi, A.K. 1999. Late Pleistocene-Holocene history of aeolian accumulation in the Thar desert, India. Zeitschrift fur Geomorphologie, Supplementband 166 (in press). Verstappen, H. Th. 1968. On the origins of longitudinal (seif) dunes, Zeitschrift fur Geomorphologie, 12: 200–220. Wadhawan, S.K. 1988. Evolution of Quaternary aeolian deposits of Jodhpur and Barmer districts, Rajasthan, India. In, M.P. Patel and N. Desai (editors), Proceedings, National Seminar on Recent Quaternary Studies in India, M.S. University, Baroda: 64–78. Wadhawan, S.K. and Kumar, V. 1996. Subsurface Quaternary aeolian stratigraphy in the Ghaggar basin of Thar desert, India. Journal of Arid Environments, 32: 37–51. Wadhawan, S.K. and Sural, B. 1992. Quaternary aeolian stratigraphy and neotectonics in Rajasthan and Gujarat, Records, Geological Survey of India, 125(7): 18–19. Wasson, R.J., Rajaguru, S.N., Misra, V.N., Agarwal, D.P., Dhir, R.P., Singhvi, A.K. and Rao, K.K. 1983. Geomorphology, late Quaternary stratigraphy and palaeoclimatology of the Thar dunefield, Zeitschrift fur Geomorphologie, Supplementband 45: 117–151. Zonneveld, K.A.F., Ganssen, G., Troelstra, S., Versteegh, G.J.M. and Visscher, H. 1997. Mechanisms forcing abrupt fluctuations of the Indian Ocean summer monsoon during the last deglaciation. Quaternary Science Reviews, 16: 187–201.

View publication stats