The collapse of the Bronze Age Harappan, one of the earliest urban civilizations, remains an enigma. Urbanism flourished in the western region of the Indo-Gangetic Plain for approximately 600 y, but since approximately 3,900 y ago, the total settled area and settlement sizes declined, many sites were abandoned, and a significant shift in site numbers and density towards the east is recorded. We report morphologic and chronologic evidence indicating that fluvial landscapes in Harappan territory became remarkably stable during the late Holocene as aridification intensified in the region after approximately 5,000 BP. Upstream on the alluvial plain, the large Himalayan rivers in Punjab stopped incising, while downstream, sedimentation slowed on the distinctive mega-fluvial ridge, which the Indus built in Sindh. This fluvial quiescence suggests a gradual decrease in flood intensity that probably stimulated intensive agriculture initially and encouraged urbanization around 4,500 BP. However, further decline in monsoon precipitation led to conditions adverse to both inundation- and rain-based farming. Contrary to earlier assumptions that a large glacier-fed Himalayan river, identified by some with the mythical Sarasvati, watered the Harappan heartland on the interfluve between the Indus and Ganges basins, we show that only monsoonal-fed rivers were active there during the Holocene. As the monsoon weakened, monsoonal rivers gradually dried or became seasonal, affecting habitability along their courses. Hydroclimatic stress increased the vulnerability of agricultural production supporting Harappan urbanism, leading to settlement downsizing, diversification of crops, and a drastic increase in settlements in the moister monsoon regions of the upper Punjab, Haryana, and Uttar Pradesh.
AUTHOR SUMMARY
The collapse of the Bronze Age Harappan or Indus Civilization remains an enigma (1). The Harappans inhabited the Indus plain at the arid edge of the monsoonal belt and developed one of the earliest urban civilizations over a territory larger than the contemporary extent of Egypt and Mesopotamia combined (Fig. P1A). Between the Indus and Ganges watersheds, a now largely defunct drainage system, the Ghaggar-Hakra, was also heavily populated during Harappan times. Controlled by the monsoon and the melting of Himalayan snow, the variable hydrologic regime must have been a concern for Harappans, as it is today for a billion people living on the Indo-Gangetic Plain. We investigate how climate change affected this civilization by focusing on fluvial morphodynamics, which constitutes a critical gap in our current understanding of the Harappan environment. Shuttle Radar Topography Mission (SRTM) data combined with field surveys and radiocarbon and optically stimulated luminescence dating offer us a way to analyze fluvial landforms and date deposits of the Indo-Gangetic Plain. We reexamine archaeological site distribution in relation to climate changes and argue for a gradual decrease in flood intensity leading to a cumulative process of settlement shift eastwards, rather than a sudden collapse linked to a single arid event.
Fig. P1.
(A) Morphology of the Indo-Gangetic Plain (for altitudes, colors repeat every 10 m to 300 m). Convex downslope, aggradational landscapes have a light mask, whereas incisional landscapes have a red mask. Ghaggar-Hakra and Nara regions are indicated by G-H and N, respectively. Upper inset shows the longitudinal profiles for the Indus and its Punjabi tributaries. Lower inset shows the Harappan domain with modern regions. (B) Early/Mature Harappan sites with major urban centers: D = Dholavira; M = Mohenjo-Daro; G = Ganweriwala; H = Harappa; K = Kalibangan; R = Rakhigarhi. Interfluves are in gray and modern river courses in blue. (C) Late Harappan sites.
Climate Change and the Harappans.
Inundation agriculture during the winter was dominant along the Indus from the Harappan era until modern times. Hydroclimate reconstructions for South and Central Asia (Fig. 4) show that precipitation from both monsoon and westerly sources was at its lowest after approximately 4,000 y BP (3). Aridification would have diminished flood intensity and allowed agriculture to expand along rivers. However, our analysis of fluvial landscapes suggests that further drying was detrimental for the Harappans, who relied on annual floods to sustain their economy. Preservation of archaeological sites at locations seemingly vulnerable to flooding, erosion, or burial suggest that, as aridity intensified, monsoon-augmented floods became less frequent and/or less intense. The most spectacular case of climate-controlled landscape transformation is the Ghaggar-Hakra system, which became ephemeral and was largely abandoned after approximately 4,000 y ago. Posturban settlements are preferentially located near the regularly flooded region at the Indus confluence with the Punjabi rivers, or in eastern regions with more reliable rains. Diversification of agriculture towards summer rain-based crops and the increase in drought-tolerant crops at the end of the urban phase (see SI Text) reveal efforts to adapt to hydroclimatic stress. Although snowmelt continued to provide water to the Indus and its Himalayan tributaries, the Harappans did not develop canal irrigation. In contrast to inhabitants of Mesopotamia and Egypt, which were surrounded by arid lands, the Harappans had the option to migrate toward more humid regions of the Indo-Gangetic Plain. Migration toward the periphery could have contributed to the decline of urban centers in the core region of the Harappan domain. Without striving for deterministic explanations for the Indus Civilization collapse, it is likely that the unprecedented scale of hydroclimatic stress must have increased the vulnerability of floodwater farmers in Harappan society.
Settlement Dynamics in the Harappan Domain.
Occurrence of Harappan sites within the incised valleys of Punjab (Fig. P1B and C) indicates that rivers were already entrenched by approximately 5,200 y ago. Numerous sites are located at the confluence zone of the Indus with the Punjabi rivers, where backwater flooding would have been common. Settlements on the Punjabi interfluves, including Harappa itself, tend to occur near their edges, close to fertile, annually flooded areas. Farther to the east, Harappan sites occur in the upper region of the Ghaggar-Hakra domain (Haryana and upper Punjab), but also along and within the incised valley of the Yamuna and on the Yamuna-Ganga interfluve. Settling of the Ghaggar-Hakra took advantage of the smaller floods typical for non-Himalayan monsoonal rivers. However, the largest agglomeration of mature Harappan sites, including the urban Ganweriwala, occurs on the lowermost Ghaggar-Hakra palaeochannel system in modern Cholistan. Proximity to both Ghaggar-Hakra and the well-watered confluence of the Indus with its Punjabi tributaries provides the best explanation for the continuity and high-density occupation there. Harappan discoveries in the Thar Desert adjacent to and along the Nara valley support our reconstructions of a better-watered past for this dry region. In upper Sindh (Fig. P1B and C), good preservation of Harappan sites on the alluvial plain suggests that the Indus mega-ridge has been relatively stable. In contrast, in the southwestern Sindh, any settlement on the alluvial plain that may have existed is now buried under fluvial deposits.
Morphodynamics of the Indo-Gangetic Plain.
Our analysis reveals a palimpsest of fluvial forms and deposits in the western region of the Indo-Gangetic Plain (Fig. P1A); however, one constant trait that is evident across the entire Harappan landscape is the change from an energetic fluvial regime earlier in the Holocene to increased stability of alluvial forms by Early Harappan times, as the climate became more arid (3). In contrast to the alluvial megafans of the eastern Indo-Gangetic Plain, we document that rivers in Punjab incised after approximately 10,000 y ago, but before approximately 3,900 y ago, probably driven by sediment load declining during centuries-long weak monsoon periods (4). Speculations have advanced the idea that the Ghaggar-Hakra, at times identified with the lost mythical Sarasvati River, was a large glacier-fed Himalayan river. Potential sources for this river include the Yamuna River, the Sutlej River, or both rivers, but the absence of wide, deeply entrenched valleys between the Sutlej and the Yamuna (Fig. P1A) demonstrates that large, glacier-fed rivers did not flow across the Ghaggar-Hakra region during the Holocene. However, our evidence shows widespread fluvial redistribution of sediment, suggesting that monsoon rains were able to sustain perennial rivers and explaining why Harappan settlements flourished along the entire Ghaggar-Hakra system. Further south, we document fluvial deposition during Harappan times both at the confluence region between the Indus and Ghaggar-Hakra systems and along channels running through the desert toward the Nara valley. The Nara itself, which runs parallel to the Indus, had active fluvial sedimentation approximately 2,900 y ago (Fig. P1A). Downstream, in Sindh, the Indus River built a unique distributive-type fluvial system that we term the Indus fluvial mega-ridge (Fig. P1A). The alluvial plain here is convex-up, showing maximum aggradation near the modern channel belt and tapering out toward the plain edges. Radiocarbon-dated deposits of old channel belts indicate that aggradation was minimal during the last approximately 3,000 y compared to earlier in the Holocene, which indicates that the development of the Indus fluvial mega-ridge was the direct consequence of late Holocene aridity (3). The reduction in sediment load compared to water discharge linked to monsoon weakening can explain the channel incision and stabilization.
Brief Harappan History.
The Harappan cultural tradition (1, 2) evolved during an Early Phase (5,200–4,500 y ago) from antecedent agricultural communities of the hills bordering the Indus alluvial plain to the west and reached its urban peak (Mature Phase) between approximately 4,500 and 3,900 y ago. The Harappans were agrarian, but developed large, architecturally complex urban centers and a sophisticated material culture coupled with a robust trade system. In contrast to Egypt and the neighboring hydraulic civilization of Mesopotamia, Harappans did not develop large-scale canal irrigation. Deurbanization after approximately 3,900 y ago was characterized by increasingly regional artifact styles and trading networks and the disappearance of the Harappan script (1, 2). While these changes are often referred to as “collapse,” archaeological evidence indicates a protracted and regionally varied process (1, 2).
B Bookhagen, D Fleitmann, K Nishiizumi, MR Strecker, RC Thiede, Holocene monsoonal dynamics and fluvial terrace formation in the northwest Himalaya, India. Geology34, 601–604 (2006).
The Harappan or Indus Civilization (1–8) developed at the arid outer edge of the monsoonal rain belt (9, Fig. 1) and largely depended on river water for agriculture (10). The Harappans settled the Indus plain over a territory larger than the contemporary extent of Egypt and Mesopotamia combined (Figs. 2 and 3). Between the Indus and Ganges watersheds, a now largely defunct smaller drainage system, the Ghaggar-Hakra, was also heavily populated during Harappan times (4, 5). Controlled by the Indian monsoon and the melting of Himalayan snow and glaciers (2, 11, 12), the highly variable hydrologic regime, with recurring droughts and floods, must have been a critical concern for Harappans, as it is today for almost a billion people living on the Indo-Gangetic Plain in Pakistan, northern India, and Bangladesh. In such challenging environmental conditions, both the development and the decline of the Harappan remain equally puzzling (13). We investigate how climate change affected this civilization by focusing on fluvial morphodynamics, which constitutes a critical gap in our current understanding of the Harappan in the way it affects habitability and human settlement patterns near rivers in arid regions.
Fig. 1.
Large-scale morphology of the Indo-Gangetic Plain (for altitudes, pattern of colors repeats every 10 m to 300 m in height; higher landscape in gray). Convex downslope, aggradational landscapes (e.g., Tista = t; Kosi = k; Gandak = g) have a lighter mask, whereas incisional landscapes have a red mask. (B) Along-channel longitudinal profiles for the Indus and its Punjab tributaries. (C) Hydrographs for the Indus and confluent Punjab tributaries (Panjnad River). (D) Profiles across the extended Indus plain (in meters above sea level). River channel locations are identified on the profiles, as are locations and ages of studied fluvial sedimentary deposits in Sindh and Punjab (ages in thousands of years).
Fig. 2.
(A) Morphology of the western Indo-Gangetic plain with interfluves (in gray mask), incised valleys (no mask), terrace edges (as dashed black lines), and active and fossilized river channels (in blue). Legend further indicates sampling locations and types. (B) Pre-Harappan sites with modern region names, chronological information (youngest fluvial deposits at all sites), and selected town names.
Fig. 3.
Settlements on morphological units of the western Indo-Gangetic plain (see Fig. 2 for color conventions and legend). (A) Early and Mature Harappan sites, with names of some major urban centers: D = Dholavira; M = Mohenjo-Daro; G = Ganweriwala; H = Harappa; K = Kalibangan; R = Rakhigarhi. (B) Late Harappan (red) and Painted Gray Ware (white) sites.
Brief Harappan History
The Harappan cultural tradition (2–4) evolved during an Early Phase (5,200–4,500 y ago) from antecedent agricultural communities of the hills bordering the Indus alluvial plain to the west and reached its urban peak (Mature Phase) between approximately 4,500 and 3,900 y ago. The Harappans were agrarian but developed large, architecturally complex urban centers and a sophisticated material culture coupled with a robust trade system. In contrast to the neighboring hydraulic civilization of Mesopotamia (14) and to Egypt (15), Harappans did not attempt to control water resources by large-scale canal irrigation (16). Deurbanization ensued after approximately 3,900 y ago and was characterized by the development of increasingly regional artefact styles and trading networks as well as the disappearance of the distinctive Harappan script (2–4; 17). While this is often referred to as “collapse,” archaeological evidence indicates a protracted and regionally varied process (2, 4, 17). Many settlements exhibit continuity, albeit with reduced size, whereas many other riverine sites are abandoned, especially along the lower Indus and the Ghaggar-Hakra course (3–5; SI Text). Between 3,900 and 3,000 y ago, there was a proliferation of smaller, village-type settlements (2–4, 6, 18), especially in the Himalayan foothills and the western part of the Ganges basin along the Yamuna River and on the Yamuna-Ganga interfluve (Fig. 3B).
Socio-economic theories have been invoked to address the collapse of urban Harappan society, including foreign invasions, social instabilities, and decline in trade (4). Alternatively, environmental factors were suggested to play a decisive role in the decline (1, 2, 8); among these, regional aridification, hydrological changes such as the drying or capture of the Ghaggar-Hakra system (3, 4, 7, 19, 20), as well as human-induced environmental degradation (21) have been advanced. Despite almost a century of research, a clear perspective on the role played by fluvial dynamics in influencing the fate of the Harappan civilization has been hampered by a lack of high-resolution topographic data and sedimentary chronologies. Shuttle Radar Topography Mission (SRTM) data (22) combined with field surveys and radiocarbon and optically stimulated luminescence dating offer us a way to analyze fluvial landforms and date deposits of the Indo-Gangetic Plain. In this context, we reexamine archaeological site distribution to understand how climate-controlled changes in river dynamics affected the Harappans (SI Text).
Morphodynamics of the Indo-Gangetic Plain
The Indo-Gangetic Plain (Fig. 1) was built during the Cenozoic with sediments derived primarily from the Himalayas (23). Our digital elevation model shows a trend from aggradation in the eastern part of the Indo-Gangetic Plain toward incising rivers in the west (24), probably driven by the westward weakening of the monsoonal rains along the Himalayas (9, 25). In the eastern Indo-Gangetic Plain, alluvial megafans were actively built during the Holocene by rivers with large, highly seasonal sediment discharge and low stream power (25). In contrast, rivers in the western Indo-Gangetic Plain (Fig. 1) are largely degradational after emerging from the Himalayan foothills. Wide, shallowly incised valleys separated by interfluves characterize the Indus and its tributaries in Punjab (26) as well as the Ganges and its westernmost tributaries (Fig. 1).
The Old Beas Survey (2, 27; 28) previously documented incision of the terminal Pleistocene sediments on the Beas-Ravi interfluve at and near Harappa followed by stable conditions and occupation levels after approximately 7,700 y ago. On the interfluves in Punjab, we dated the latest fluvial channel deposits to approximately 10,000 y ago (Fig. 2B; SI Text), confirming that large-scale fluvial sedimentation ceased at the beginning of the Holocene. Within the entrenched river valleys, we mapped abandoned river channel belts and terraces (26), which indicate periodic but progressive incision (29, 30). Our dates on incised valley deposits vary in age between 700 and 3,900 y (Fig. 2B; SI Text). Thus, rivers in Punjab started to incise after 10,000 y ago, but before 3,900 y ago. As documented along the Himalayan course of the Sutlej River (12), the easternmost tributary of the Indus, century-long phases of sediment load decline caused by weak monsoons were responsible for incision, primarily in the early Holocene between approximately 10,000 and 8,700 y ago. The presence of Harappan and even earlier settlements within these incised valleys (vide infra) also argues for major incision predating the Harappan. During Harappan times, the alluvial landscape in Punjab offered suitable terrain for floodwater farming within incised valleys and important protection against large floods on interfluves.
The similarity of incision profiles (29, 31) for Indus tributaries across Punjab (Fig. 1) reflects comparable hydrological histories during the Holocene (25). On the other hand, high water discharge and sediment load (29) explain the relatively steep longitudinal profile of the Indus. We note the sharp contrast between the degradational character of the tributaries of the Indus and the Ganges in the western Indo-Gangetic Plain and the lack of wide incision valleys along the Ghaggar-Hakra interfluve (Figs. 1 and 2A). Numerous speculations have advanced the idea that the Ghaggar-Hakra fluvial system, at times identified with the lost mythical river of Sarasvati (e.g., 4, 5, 7, 19), was a large glacier-fed Himalayan river. Potential sources for this river include the Yamuna River, the Sutlej River, or both rivers. However, the lack of large-scale incision on the interfluve demonstrates that large, glacier-fed rivers did not flow across the Ghaggar-Hakra region during the Holocene. Existing chronologies (27, 28) and our own age on the bank of Sutlej (SI Text) identified deposits of Late Pleistocene age, indicating that the interfluve formed instead during the last glacial period. Provenance detection (32) suggests that the Yamuna may have contributed sediment to this region during the last glacial period, but switched to the Ganges basin before Harappan times.
The present Ghaggar-Hakra valley and its tributary rivers are currently dry or have seasonal flows. Yet rivers were undoubtedly active in this region during the Urban Harappan Phase. We recovered sandy fluvial deposits approximately 5,400 y old at Fort Abbas in Pakistan (SI Text), and recent work (33) on the upper Ghaggar-Hakra interfluve in India also documented Holocene channel sands that are approximately 4,300 y old. On the upper interfluve, fine-grained floodplain deposition continued until the end of the Late Harappan Phase, as recent as 2,900 y ago (33) (Fig. 2B). This widespread fluvial redistribution of sediment suggests that reliable monsoon rains were able to sustain perennial rivers earlier during the Holocene and explains why Harappan settlements flourished along the entire Ghaggar-Hakra system without access to a glacier-fed river (5, Fig. 3A). Similar, strictly monsoonal rivers maintaining a groundwater-fed base flow are now active only on the more humid Ganga basin (34). We also document renewed fluvial deposition on the lower Ghaggar-Hakra system approximately 700 y ago, which indicates that seasonal monsoon flows intensified episodically during the late Holocene and may provide an explanation for the high concentration of medieval fortified sites in this region (5).
Farther to the south, the five Punjab tributaries of the Indus merge to form the Panjnad River, before joining the Indus (Figs. 1 and 2A). Incision (4–5 m deep) between the two confluences and further south along the greater Indus separates vertically the modern floodplain and the southernmost extension of the Ghaggar-Hakra interfluve in the Cholistan region. Dunes younger than 1,500 y old on the edge of the expanding Thar Desert have begun to cover this region of the interfluve, but sediment originating from the Indus-Punjab system, the Ghaggar-Hakra, or from both of these river systems was deposited as late as 4,250 y ago (Fig. 2B; SI Text). Zircon dating of sand in this confluence region indicates inputs from both Beas and Sutlej drainage basins (32). Continuing to the southwest on the Ghaggar-Hakra interfluve, we document well-watered lands in the region of Pat, where channels ran parallel with the Indus and joined the Nara valley; their fluvial deposits at Fakirabad, among the dunes of the expanding desert, are even younger at approximately 3,350 y old. Further south, the Nara valley, which would be currently dry if not for modern irrigation, also had active fluvial sedimentation approximately 2,900 y ago (Fig. 2B; SI Text).
Downstream in the province of Sindh, the Indus River built a unique distributive-type fluvial system that we term the Indus fluvial mega-ridge (Fig. 1). The alluvial plain here is convex up (35–37), showing maximum aggradation near the modern channel belt and tapering out toward the plain edges (SI Text). The cross-sectional relief of the ridge is very subdued (over 100 km wide and 10–15 m high; Fig. 1) and the river is stable on its apex because the thalweg is incised as deep as the ridge. Fossil channel belts and associated crevasse splays occur on both sides of the modern Indus course (Fig. 2A; SI Text). Radiocarbon-dated fluvial deposits of old channel belts in lower Sindh indicate that aggradation was minimal during the late Holocene (between 2.4 m/kyr near the mega-ridge top down to less than 1 m/kyr near its edge; Fig. 2B; SI Text). This relative stability of the late Holocene landscape suggests that large avulsions of the Indus were rare and distributary channels acted mainly as overspills, as documented for the historical period (38, 39). In contrast, at our Matli floodplain drill site on top of the Indus mega-ridge (Fig. 2B; SI Text), sedimentation rates were at least three times higher between approximately 7,200 and 2,700 y ago compared to the last approximately 2,700.
We speculate that the development of the Indus fluvial mega-ridge was also the direct consequence of late Holocene aridity (12, 40–42). Hydroclimate in the western Indo-Gangetic Plain is influenced by both the Indian summer monsoon system and westerly winter disturbances bringing humidity from the Mediterranean, Black, and Caspian Seas (9, 11). Most sediment is delivered to the Indus by floods after high-intensity monsoon storms (43), but the bulk of Indus water discharge is dependent on snow melt (11). The weakening of the monsoon after approximately 5,000 y ago compared to the slower decline in winter precipitation originating in western Asia (40, 42, 44, Fig. 4) must have resulted in a reduction in sediment load compared to water discharge, causing channel incision and stabilization (12; SI Text) and leading to longer intervals of decoupling between channels and the alluvial plain. The subdued relief of the fluvial ridge, resulting from less frequent breaches and overspills as well as cohesive banks (35), which are typical for arid regions, are not favorable to avulsions. Rarity of large scale avulsions reinforces deposition close to existing channel belts and allows for the slow growth of the mega-ridge. Within the deep active channels on top of the ridge, effective conveyance of sediments toward the coast for the build-up of a new deltaic depocenter in western lower Sindh (45, 46) must have diverted most of the sediment away from the Indus alluvial plain in the late Holocene.
Fig. 4.
Palaeo-hydroclimate reconstructions and sedimentary activity chronology for the Harappan domain. (A) Indian monsoon precipitation reconstruction (35). (B) Synthetic reconstructions for the Indian monsoon domain (black) and westerly precipitation (37) (gray). (C) Sedimentation ages for fluvial forms in: Sindh (i.e., Indus fluvial mega-ridge); Nara Valley; Indus-Panjnad-Ghaggar-Hakra confluence; Punjab incised valleys; Punjab interfluves (doabs); and the lower and upper Ghaggar-Hakra (G-H) system regions, respectively. OSL dates are represented by triangles and calibrated AMS radiocarbon dates by circles; error bars are shown if larger than the symbol size. Fluvial deposits are shown in black- and white-filled symbols (for high vs. low sedimentation rate deposits, respectively), whereas dunes are represented by gray-filled symbols. Ages for the upper Ghaggar-Hakra system are from Saini et al. (33). The temporal extent of each Harappan phase is indicated by vertical dashed lines (E = early; M = mature; L = late).
Our analysis reveals a palimpsest of fluvial forms and deposits in the western region of the Indo-Gangetic Plain; however, one constant trait that is evident across the entire Harappan landscape is the change from a more energetic fluvial regime earlier in the Holocene (before approximately 5,000 y ago; Fig. 4) to increased stability of alluvial forms by Early Harappan times, and even drying up of some river channels during and after Harappan times.
Settlement Dynamics in the Harappan Domain
The distribution of Harappan sites within the incised valleys of Punjab (Fig. 3A) provides clear evidence that rivers were already entrenched by 5,200 y ago or earlier (Fig. 2B). Numerous sites are present in the incised valley at the confluence zone of the Indus with the Punjab rivers. In this region of confluences, regardless of their past geometries, backwater flooding would have been common because the Indus and its tributaries in Punjab reach their flood stages asynchronously (Fig. 1). Settlements on the Punjab interfluves, including Harappa, also tend to be located near their edges, close to the fertile, annually flooded areas in the incised valleys (Fig. 3A; 28). Farther to the east, Harappan sites have been reported along a network of smaller monsoon-fed rivers in the upper region of the Ghaggar-Hakra domain (i.e., Haryana and upper Punjab; 19, 47, 48) as well as along and within the incised valley of the Yamuna and the Yamuna-Ganga interfluve (49, 50). As the climate continued to become drier during late Harappan times (Fig. 4), the number of sites increased in the upper Punjab and Haryana, especially on interfluves near the Himalayan piedmont where monsoonal rains are more consistent (Fig. 3B).
The largest agglomeration of mature Harappan sites, including the urban Ganweriwala, occurs on the lowermost Ghaggar-Hakra interfluve (5, 49) in modern Cholistan. The proximity to both the Ghaggar-Hakra valley and the well-watered Indus-Punjab river confluence region provides the best explanation for the unusual continuity and high-density occupation of the lower Ghaggar-Hakra interfluve (Figs. 2 and 3). Recent Harappan discoveries in the Thar Desert adjacent and along the Nara valley (51) support our reconstructions of a better-watered past for this dry region as well. As channels of the Ghaggar-Hakra dried through the Late Harappan Phase, fewer sites occur on the lower part of the Sutlej-Yamuna interfluve (4, 5), with the notable exception of sites closest to the Indus where access to water remained somewhat reliable (Fig. 3B). In contrast, the number of sites in the upper region of the interfluve increased as they did beyond the Yamuna to the east (4, 18, 50; SI Text, Fig. 3B). Later, Painted Gray Ware sites were reestablished along the middle Ghaggar-Hakra course (Fig. 3B), suggesting a possible reinvigoration of seasonal river flows during the Iron Age (5). The lack of Painted Gray Ware sites at the distal end of the Ghaggar-Hakra system (4, 5), however, supports our chronostratigraphic evidence that river flow had become ephemeral and did not reach as far as the Indus.
In Sindh, in the southern part of the Harappan domain, many archaeological sites are found on landforms that were not affected by fluvial erosion or deposition, such as the western Pakistan ranges and bedrock inselbergs raised above the alluvial plain. However, in upper Sindh, good preservation of early and mature Harappan sites on the alluvial plain (45), including Mohenjo-Daro, suggests that the Indus mega-ridge has been relatively stable in that region since Harappan times. In contrast, in lower Sindh, just a few Harappan sites have been discovered on the eastern delta plain (52), supporting the idea of a switch of the fluvio-deltaic depocenter to the west (45, Figs. 2 and 3). Any settlement on the alluvial plain that may have existed in the southwestern part of Sindh is probably buried under later fluvial sediments. Sites in Kutch, a coastal region that was never dependent on flood agriculture but instead relied on trade, remained active into late Harappan times (4, Fig. 3B), when settlements were far fewer on the Indus plain.
Although erosion or burial of archaeological sites is an important process in fluvial environments, preservation of Harappan archaeological sites in potentially dynamic contexts such as entrenched valleys, river confluences, and accretionary distributive fluvial forms indicates relatively low fluvial activity across the wider alluvial plain since the time settlements were established. In this context, however, the preferential occurrence of Harappan urban sites on interfluves, but in close proximity to floodable, agriculturally viable land (20, 24), suggests awareness of devastating floods that can occur on Himalayan rivers. Settling of the Ghaggar-Hakra river system is the best example of an adaptation strategy that takes advantage of smaller floods along monsoonal rivers, where monsoon rains are not augmented by meltwater. However, most floods, whether in entrenched valleys in Punjab, along overspills in Sindh, or along the smaller rivers of the Ghaggar-Hakra system, must have been predominantly benign in character to be able to foster intensive agriculture and also regular enough not to require canal irrigation.
Diminished monsoon rains and less reliable flooding during late Harappan times could explain the expansion of settlements near the headwaters of the Ghaggar-Hakra system, where small-scale floods probably remained more dependable and where rains allowed summer-cropping on interfluves. Agriculture, land development, and continuous habitation can skew the distribution of archaeological sites, but such biases should be minimal in desert areas like Cholistan. In upper Punjab, Haryana, and Uttar Pradesh, where historical and modern destruction of sites has been intensive, the increase in the number of sites during the late Harappan shows an unambiguous trend. In contrast, in lower Punjab and Sindh, regions of similarly intense history of human activities, the number of sites decreased over the same interval.
Climate Change and the Harappans
For the arid Indus region, river floods have always been far more important and reliable for agriculture than rainfall. As during the Harappan era (7, 10, 20), inundation agriculture during the Rabi season (winter crops) was dominant along the Indus and its tributaries (10, 38, SI Text) prior to the development of the largest irrigation system in the world in 20th century Pakistan. Hydroclimate reconstructions for South and Central Asia (39–41) show that precipitation from both monsoon and westerly sources that feed rivers of the western Indo-Gangetic Plain (20) decreased since approximately 5,000 y ago, and was at its lowest after approximately 4,000 y BP (Fig. 4). This drying of the Indus region supports the hypothesis that adaptation to aridity contributed to social complexity and urbanization (1, 52). An intense drought period around 4,200 y ago, linked to major disruptions in Egypt and Mesopotamia (53–56), coincides instead with the flourishing of Harappan urbanism within the limits of available chronology. This event, recorded at the start of the aridification process in South Asia, had affected the total Indus discharge (57), but aridification may have diminished the intensity of floods and allowed inundation agriculture to expand along the Indus and its tributaries (20). However, our analysis of fluvial landscapes suggests that further drying was detrimental for the Harappans, who relied on annual floods to sustain their economy. Good preservation of numerous archaeological sites at locations seemingly vulnerable to flooding, erosion, or burial suggest that, as aridity intensified, monsoon-augmented floods became less frequent and/or less intense. The most spectacular case of climate-controlled landscape transformation is the Ghaggar-Hakra system, which became ephemeral and was largely abandoned.
Settlements of the posturban period are preferentially located near the easily flooded region at the confluence of the Indus with rivers in Punjab or in eastern regions with more reliable monsoon rains. Diversification of agriculture towards Kharif (summer) rain-based crops and the increase in drought-tolerant crops like millets (SI Text) at the end of the urban phase (58) reveal intense efforts to adapt to hydroclimatic stress at the arid outer edge of the monsoonal rain belt. The climate-controlled decline in floods concurrent with the reduction of monsoon rains probably led to a decrease in surpluses for supporting that part of the population that was not involved directly in agriculture, which in turn led to diminished social complexity. Although snowmelt continued to regularly provide water to the Indus and its Himalayan tributaries (57), the Harappans did not develop canal irrigation. In contrast to inhabitants of Mesopotamia and Egypt, which were surrounded by arid lands, the Harappans had the option to migrate east toward more humid regions of the Indo-Gangetic Plain. Migration toward the periphery could have contributed to reduced population in the core region of the Harappan domain and the decline of urban centers. The unprecedented scale of hydroclimatic stresses must have increased the vulnerability of Harappan society, but does not provide a simple, deterministic explanation for the transformations in site size, distribution, and interrelationships across the whole civilization area (2, 4, 6, 17, 59). The longevity of the decentralized late Harappan phase and continued habitation documented in places in Sindh and lower Punjab suggests that the perennial flows of the Indus and its Himalayan tributaries still flooded agriculturally viable lands, albeit less extensively than earlier. As a lesson from the past, the possible return to stronger monsoon-augmented floods (60), similar to the disastrous events of 2010 in Pakistan, may render current flood controls and irrigation systems vulnerable and require the large-scale adaptation of modern society in the western Indo-Gangetic Plain.
Methods
We used SRTM (22) data to map the Indo-Gangetic Plain, focusing on its western region (see SI Text for full methods). We identified active and abandoned river courses, incised valleys and interfluves, terraces, fluvial ridges, alluvial fans, piedmonts, and dune fields. To avoid misinterpreting the morphology due to post-Harappan alterations of the Indo-Gangetic Plain via natural and anthropogenic processes, especially over the past century, we only interpreted features that retain a clear, large-scale SRTM topographical expression. Using SRTM-based digital elevation models, satellite photography, and geological maps, we selected key locations in Sindh and Punjab (Pakistan) to investigate the stratigraphy of the uppermost Indo-Gangetic Plain fill. Chronologies for fluvial and aeolian deposits in newly collected cores, trenches, and large-scale exposures in sand quarries were constructed based on accelerator mass spectrometry (AMS) radiocarbon and optically stimulated luminescence dating (SI Text). Location and age estimates of archaeological sites (SI Text) provided additional chronological information to assess the dynamics of fluvial environments.
Note
The authors declare no conflict of interest.
Acknowledgments.
We thank the personnel and Dr. M.M. Rabbani, then Director General of the National Institute of Oceanography in Karachi, for logistical support in Pakistan. We thank Dr. Rita Wright for sharing information on sites of the Harappa and Lower Beas Regional Survey. This study was supported by grants from the National Science Foundation, the Leverhulme Trust, Woods Hole Oceanographic Institution, the University of Aberdeen, and Louisiana State University.
DQ Fuller, M Madella Indian Archaeology in Retrospect. Protohistory, eds S Settar, R Korisettar (Manohar Publishers, New Delhi) vol. II, 317–390 (2002).
B Bookhagen, DW Burbank, Towards a complete Himalayan hydrological budget: the spatiotemporal distribution of snow melt and rainfall and their impact on river discharge. J Geophys Res115, F03019 (2010).
M-LH Miller Agriculture and Irrigation in Archaeology, eds C Stanish, J Marcus (Cotsen Institute of Archaeology Press, Los Angeles), pp. 92–128 (2006).
A Karim, J Veizer, Water balance of the Indus river basin and moisture source in the Karakoram and western Himalayas: implications from hydrogen and oxygen isotopes river water. J Geophys Res107, 4362, doi:. (2002).
B Bookhagen, D Fleitmann, K Nishiizumi, MR Strecker, RC Thiede, Holocene monsoonal dynamics and fluvial terrace formation in the northwest Himalaya, India. Geology34, 601–604 (2006).
RM Adams Heartland of Cities: Surveys of Ancient Settlement and Land Use on the Central Floodplain of the Euphrates (University of Chicago Press, Chicago, 1981).
M Kumar Linguistics, Archaeology and the Human Past, Occasional Paper 7, eds T Osada, A Uesugi (Research Institute for Humanity and Nature, Nakanishi Printing Co. Ltd, Kyoto), pp. 1–75 (2009).
R Sinha, S Sarkar, Climate-induced variability in the Late Pleistocene-Holocene fluvial and fluvio-deltaic successions in the Ganga plains, India. Geomorphology113, 173–188 (2009).
J Schuldenrein, RP Wright, R Mughal, MA Khan, Landscapes, soils, and mound histories of the Upper Indus Valley, Pakistan: new insights on the Holocene environments near ancient Harappa. J Archaeol Sci31, 777–797 (2004).
HS Saini, SK Tandon, SAI Mujtaba, NC Pant, RK Khorana, Reconstruction of buried channel-floodplain systems of the northwestern Haryana Plains and their relation to the ‘Vedic’ Saraswati. Current Sci India97, 1634–1643 (2009).
MR Gibling, SK Tandon, R Sinha, M Jain, Discontinuity-bounded alluvial sequences of the southern Gangetic plains, India: aggradation and degradation in response to monsoonal strength. J Sed Res75, 369–385 (2005).
DJ Harbor, SA Schumm, MD Harvey The Variability of Large Alluvial Rivers, eds SA Schumm, B Winkley (American Society of Civil Engineers), pp. 161–176 (1994).
AK Gupta, DM Anderson, JT Overpeck, Abrupt changes in the Asian southwest monsoon during the Holocene and their links to the North Atlantic Ocean. Nature421, 354–356 (2003).
H Wulf, B Bookhagen, D Scherler, Seasonal precipitation gradients and their impact on erosion in the Northwest Himalaya. Geomorphology118, 13–21 (2010).
F Chen, et al., Holocene moisture evolution in arid central Asia and its out-of-phase relationship with Asian monsoon history. Quaternary Sci Rev27, 351–364 (2008).
CA Petrie, et al., Investigating changing settlement dynamics on the plains: the 2009 survey and excavations at Masudpur (Hissar District, Haryana). Puratattva39, 38–49 (2009).
RD Joshi, B Madhu, R Jassu Frontiers of the Indus Civilization, eds BB Lal, SP Gupta (Indian Archaeological Society jointly with Indian History & Culture Society, London, 1984).
Q Mallah Linguistics, Archaeology and the Human Past, Occasional Paper 3, eds T Osada, A Uesugi (Research Institute for Humanity and Nature, Nakanishi Printing Co Ltd, Kyoto), pp. 27–76 (2008).
M Staubwasser, F Sirocko, PM Grootes, M Segl, Climate change the 4.2 ka BP termination of the Indus valley civilization and Holocene south Asian monsoon variability. Geophys Res Lett30, 1425, doi:. (2003).
RW Law Occasional Paper 11: Linguistics, Archaeology and the Human Past (Research Institute for Humanity and Nature, Nakanishi Printing Co. Ltd, Kyoto, 2008).
BN Goswami, V Venugopal, D Sengupta, MS Madhusoodanan, PK Xavier, Increasing trend of extreme rain events over India in a warming environment. Science314, 1442–1445 (2006).
Cover image: Pictured is the giant nudibranch, Dendronotus iris, with brightly colored spines and gills. This soft-bodied sea slug feeds on the burrowing anemone and sequesters toxins from its prey, which are then used as a defense. These animals are also capable of swimming to escape from predators. James M. Newcomb et al. review the evolution of swimming behaviors in shell-less mollusks at the behavioral and neuronal levels, as discussed at the Arthur M. Sackler Colloquium, “In the Light of Evolution VI: Brain and Behavior.” See the article by Newcomb et al. on pages 10669–10676 in supplement 1. Image courtesy of Charuni A. Gunaratne.
We thank the personnel and Dr. M.M. Rabbani, then Director General of the National Institute of Oceanography in Karachi, for logistical support in Pakistan. We thank Dr. Rita Wright for sharing information on sites of the Harappa and Lower Beas Regional Survey. This study was supported by grants from the National Science Foundation, the Leverhulme Trust, Woods Hole Oceanographic Institution, the University of Aberdeen, and Louisiana State University.
Community Surface Dynamics Modeling System (CSDMS) Integration Facility, Institute of Arctic and Alpine Research (INSTAAR), University of Colorado, Boulder, CO 80309-0545
To whom correspondence should be addressed. E-mail: [email protected].
Author contributions: L.G., P.D.C., M.G.M., and D.Q.F. designed research; L.G., P.D.C., M.G.M., D.Q.F., S.C., J.A.D., T.S., G.A.T.D., A.R.T., K.G., R.A., A.A., F.F., S.V., and J.P.M.S. performed research; L.G., P.D.C., M.G.M., D.Q.F., S.C., J.A.D., T.S., G.A.T.D., A.R.T., K.G., R.A., and J.P.M.S. analyzed data; and L.G. and D.Q.F. wrote the paper.
Competing Interests
The authors declare no conflict of interest.
Metrics & Citations
Metrics
Note: The article usage is presented with a three- to four-day delay and will update daily once available. Due to ths delay, usage data will not appear immediately following publication. Citation information is sourced from Crossref Cited-by service.
Citation statements
Altmetrics
Citations
Cite this article
L. Giosan, P.D. Clift, M.G. Macklin, D.Q. Fuller, S. Constantinescu, J.A. Durcan, T. Stevens, G.A.T. Duller, A.R. Tabrez, K. Gangal, R. Adhikari, A. Alizai, F. Filip, S. VanLaningham, J.P.M. Syvitski, Fluvial landscapes of the Harappan civilization, Proc. Natl. Acad. Sci. U.S.A.
Emotions coordinate our behavior and physiological states during survival-salient
events and pleasurable interactions. Even though we are often consciously aware of
our current emotional state, such as anger or happiness, the mechanisms giving ...Emotions are often felt in the body, and somatosensory feedback has been proposed
to trigger conscious emotional experiences. Here we reveal maps of bodily sensations
associated with different emotions using a unique topographical self-report method.
In ...
Giant meteorite impacts during Earth’s early history likely had significant effects
on early life. We studied the effects on the surface environment and life of a Paleoarchean
impactor ~50 to 200× larger than the famous K-Pg impactor. The ...Large meteorite impacts must have strongly affected the habitability of the early
Earth. Rocks of the Archean Eon record at least 16 major impact events, involving
bolides larger than 10 km in diameter. These impacts probably had severe, albeit temporary,
...
An influential study suggests that Black newborns experience much lower mortality
when attended by Black physicians after birth. Using the same data, we replicate those
findings and estimate alternative models that include controls for very ... The racial gap in infant mortality is a pressing public-health concern, and [B. N.
Greenwood et al., Proc. Natl. Acad. Sci. U.S.A.117, 21194–21200 (2020), 10.1073/pnas.1913405117] suggest that Black newborns are more likely to survive if cared for by ...
