Many aspects of the South Asian environment have been stable throughout the postglacial period but there have also been major changes, both natural and human induced. There is considerable debate as to whether the rainfall, the temperature, and the pattern of the monsoon winds in Indus times differed from those of today. Tectonic movements bringing earthquakes and landslides have been responsible for major alterations to the pattern of the landscape, and the activities of rivers have frequently sculpted changes on a smaller scale. Human exploitation, particularly for cultivation and animal husbandry, has wrought not only considerable destruction of vegetation and fauna, but also associated environmental degradation, while the natural drainage has been altered in some areas by the building of dams and the cutting of canals.
The “Lost Saraswati" River
The most dramatic change took place in the region south of the Indus River, where there is evidence that a great river system flowed in the Harappan period. Through ground survey and methods of remote sensing such as satellite photography, many stretches of dry riverbed have been traced in the Thar Desert and in the Indo-Gangetic divide, often as much as 10 kilometers wide, showing that they once held substantial rivers. The shells of freshwater molluscs found in their banks tell the same story. Today a number of small seasonal rivers rising in the Siwalik Hills occupy a narrow channel running for a short distance in some of these dry riverbeds. A massive concentration of Harappan and other prehistoric settlements has been discovered along the dry riverbeds, and it is clear that the drainage of the region has a complicated history. Many names are attached to different parts of the drainage, including the Hakra and the Ghaggar; so, for the sake of clarity, scholars frequently refer to the whole ancient system as the lost Saraswati. In the Vedas, the earliest sacred literature of the subcontinent, there are references to the Saraswati as a mighty river, but by the early first millennium BCE the Saraswati was said to disappear into the sand and to flow underground to join the Yamuna farther south.
Dating ancient river courses is a knotty problem, although the date of settlements along their banks is helpful, and establishing the sequence and significance of changes is a formidable task. There is not yet full agreement on the history of the Saraswati, and continuing work often brings changes to the generally accepted picture. At present it is thought (see Maps, 2, 4, and 5) that, during the Indus period and in earlier times, the rivers rising in the Siwalik Hills—including the Naiwals, the river still known as the Saraswati, the Ghaggar, the Wah, and the Drishadvati (now known as the Chautang)—were augmented by the waters of the Yamuna, flowing in the current bed of the Drishadvati, and of the Sutlej, now a tributary of the Indus, then flowing in the one of the western riverbeds that combine to form the Hakra. Both the Sutlej and the Yamuna are major rivers that rise in the Himalayas and that are swelled by snowmelt and monsoon rains; during the Indus period, the waters of these rivers combined to form a great waterway that flowed through the now arid region of Cholistan in the Thar Desert. The Beas River, which now joins the Sutlej, in Indus times flowed west to merge with the other rivers of the Punjab a little east of their confluence with the Indus. The precursor of what is now the Yamuna system was a minor river fed by a series of streams rising in the Himalayan foothills east of the Drishadvati.
In the Indus period the Saraswati river system may have been even more productive than that of the Indus, judging by the density of settlement along its course. In the Bahawalpur region, in the western portion of the river, settlement density far exceeded that elsewhere in the Indus civilization. Several sizable cities lay within this dense concentration, including the metropolis of Ganweriwala whose location is equidistant from the other great Indus cities of Harappa and Mohenjo-daro. Indus settlements are also densely concentrated along other parts of the river's course; while there are some fifty sites known along the Indus, the Saraswati has almost a thousand. Although this is a somewhat distorted figure since erosion and alluviation have between them probably destroyed or deeply buried many settlements in the Indus Valley itself, there can be little doubt that the Saraswati system did yield a great proportion of the Indus people's agricultural produce.
The course of the Saraswati west of Bahawalpur is still uncertain. Recent work has uncovered evidence of a possible inland delta near Fort Derawar, and it has been suggested that the Saraswati may have ended here, running into the sand in a fan of distributaries. There is also some evidence that the Saraswati split into two beds near Fort Derawar. For a stretch of around one hundred and fifty miles to the southwest no relics of ancient river channels can be traced, although the widespread presence of alluvium in the desert indicates that at some time, probably in the much earlier past, a river flowed in this region (which is beyond the reach of present or past Indus alluviation). However, what may have been the continuation of the Saraswati reappears as two rivers, now seasonal, flowing through the Thar Desert: the Raini Nullah and the Wahinda. To their west a continuation is offered by the bed now occupied by the Eastern Nara, a tributary of the Indus: It is not known whether this channel was fed by the Indus or by the Saraswati in the Harappan period. This river may have joined the Indus at the head of its delta to reach the Arabian Sea through what is now the Great Rann. Alternatively, the Eastern Nara may have flowed separately into the Great Rann.
During the second millennium BCE, there were significant alterations to the regional drainage, probably as the result either of a major tectonic event that shifted the course of the major rivers before they descended to the plains or of slightly altered gradients on the extremely flat plains. Some evidence indicates that changes occurred first to the Yamuna/Drishadvati. The main river (Yamuna) shifted its course eastward early in the second millennium, eventually reaching its current bed by the first millennium, while the Drishadvati bed retained only a small seasonal flow; this seriously decreased the volume of water carried by the Saraswati. The Sutlej gradually shifted its channel northward, eventually being captured by the Indus drainage. At first it was an independent tributary of the Indus, but in more recent times it joined the Beas before they both flowed into the Indus. The loss of the Sutlej waters caused the Saraswati to be reduced to the series of small seasonal rivers familiar today. Surveys show a major reduction in the number and size of settlements in the Saraswati region during the second millennium.
The Changing Course of the Indus
The lower Indus drainage has also had a complex history. While the river is prone to change along most of its course, today the central section of its bed is anchored by the gorge in the Rohri-Sukkur Hills, the Sukkur gap, through which it flows. As recently as the time of Alexander the Great (fourth century BCE), however, it skirted these hills to the west. Throughout its Holocene history the Indus has had a major tributary, the Eastern Nara, which, over time, has also changed both its course and the place of its confluence with the Indus. The progressive creation of a substantial delta on the Arabian Sea, reducing the rivers' gradient and increasing their length, has been a major cause of the rivers' changing course. While there is still debate about which of the palaeochannels traced in Sindh was occupied by which river at which time, a likely sequence of changing courses has been established.
In the early Holocene period, the Indus probably followed the course of the Jacobabad and Shahdadkot palaeochannels, flowing east to the south of the Marri-Bugti Hills and south along the eastern edge of the Kirthar Range. It then swung west and flowed along the Sanghar palaeochannel, joining the Eastern Nara, and the combined river flowed into the Arabian Sea a little northwest of Kutch. It may have held this course until around 4000 BCE, but by the Indus period it was flowing in the more gently curving Warah palaeochannel east of its earlier course. This crossed the course of the modern Indus a little south of Mohenjo-daro and flowed south in the Samaro-Dhoro palaeochannels, being joined by the Eastern Nara well to the south. It is possible that during both periods the river occupying the bed of the Eastern Nara was the lower course of the Saraswati rather than a branch of the Indus as it is today. Further alterations in the rivers' course occurred by or after 2000 BCE, and by the fourth century, when Alexander visited the region, the river was closer to its modern course.
In 1819 an earth tremor in Kutch created a huge natural dam 75 miles long and up to 16 miles wide that disrupted the flow of the Eastern Nara branch of the Indus. After seven years, however, the river's floodwaters created a breach and the Nara resumed its flow to the sea. Dramatic short-term events on this scale are likely to have taken place also in the past, and some scholars hypothesize that a similar event caused Mohenjo-daro to be swamped by floodwaters, initiating its decline: This theory finds little support, however.
The Arabian Sea Coast
It is thought that the Harappan settlements in the inhospitable Makran coastal region, such as Sutkagen-dor and Balakot, were located to engage in sea trade and the exploitation of coastal resources. Now well inland, these were closer to the coast in Harappan times when, according to the results of geological studies, the Makran coastline ran north of its present-day location. Intriguing evidence from Amri on the Indus, now well inland, suggests the possibility that an arm of the sea stretched much farther inland in the early third millennium: Molluscs in Early Harappan levels are almost all marine species, but by the Harappan period riverine species were increasing and were as common as marine species by the Late Harappan period. This may relate to other evidence that suggests that the sea level around 3000 BCE was between 2 and 5 meters higher in this region than today's level and that it fell dramatically after around 2200-2000 BCE. The region has also been subject to considerable tectonic uplift in the postglacial period.
From the Karachi region, the coastline in the early postglacial period seems to have run roughly east to around the mouth of the Luni River in Gujarat. The mouth of the Indus lay well to the east of its present location. Throughout most of the Holocene period the Indus River has been building a massive delta, its annual deposition of silts at its mouth pushing the coastline progressively west and southwest. In modern times, however, a major dam and a number of major canals have been built for irrigation, removing significant volumes of water upstream and reducing the quantity of silt deposited at the river mouth, with the result that today the delta is being eroded away slightly faster than it is being replaced by annual alluviation.
The Harappan Indus delta was therefore emptying into what is now the swampy Ranns of Kutch. At that time, however, this was open water, separating mainland South Asia from a large island that is now Kutch. Tectonic activity may have raised the region since Harappan times, and the Indus and other rivers also deposited silt, filling in the area of the Ranns. To the south, the Saurashtra peninsula may also have been separated from the mainland by a tidal channel that ran through what is now the Nal Depression; this has been filled in through time by alluvium deposited by rivers of Rajasthan and Gujarat. On the other hand, pollen evidence shows that Nal Lake in its center was a freshwater lake in Harappan times.
Climate Changes
Pollen evidence from the saline Sambhar, Didwana, and Lunkaransar Lakes and from the freshwater Pushkar and Gajner Lakes in Rajasthan has been interpreted as showing that the temperatures and summer rainfall during the middle Holocene period were higher than those of today and that rainfall rose considerably around 3000 BCE, continued at the elevated level until the early second millennium, and then declined. Settlement around 3000 BCE in Kashmir, a region that for long periods of the past was too cold for human habitation, suggests the onset of a period of warmer, wetter conditions in the third millennium BCE. Opinions on these matters, however, are divided, and many scholars hold the view that temperature, rainfall, and monsoon patterns changed little during the Holocene period. Wetter conditions recorded in the lakes of Rajasthan, for example, could be the result of the rivers then flowing through the region rather than of an increase in rainfall. Nevertheless, a number of fluctuations in temperature and rainfall, as evidenced at certain times in various parts of the world, are thought to reflect global climate changes, due to factors such as changes in the tilt of the earth in its solar orbit and occasional major catastrophes such as massive volcanic eruptions. Evidence from many areas indicates that warmer and wetter conditions than those of today pertained globally in the mid-Holocene period. Thereafter, the picture is more complicated, with increased rainfall in some areas coinciding with decreases in others. However, warmer, drier conditions seem to have been widespread during the second millennium.
World History
