Counter-intuitive influence of Himalayan river morphodynamics on Indus Civilisation urban settlements

Abstract

Urbanism in the Bronze-age Indus Civilisation (~4.6-3.9 thousand years before the present, ka) has been linked to water resources provided by large Himalayan river systems, although the largest concentrations of urban-scale Indus settlements are located far from extant Himalayan rivers. Here we analyse the sedimentary architecture, chronology and provenance of a major palaeochannel associated with many of these settlements. We show that the palaeochannel is a former course of the Sutlej River, the third largest of the present-day Himalayan rivers. Using optically stimulated luminescence dating of sand grains, we demonstrate that flow of the Sutlej in this course terminated considerably earlier than Indus occupation, with diversion to its present course complete shortly after ~8 ka. Indus urban settlements thus developed along an abandoned river valley rather than an active Himalayan river. Confinement of the Sutlej to its present incised course after ~8 ka likely reduced its propensity to re-route frequently thus enabling long-term stability for Indus settlements sited along the relict palaeochannel.

Fig. 1

Topographic map showing northwestern India and Pakistan, key Himalayan rivers and the distribution of urban-phase Indus Civilisation settlements. Note how Indus urban-phase settlements are not necessarily located along modern Himalayan river courses. The most prominent cluster of sites occurs located on the drainage divide between the Sutlej and Yamuna rivers, an area devoid of perennial Himalayan drainage. Base digital elevation map is derived from NASA Shuttle Radar Topography Mission (SRTM). Site locations are from the compilation of urban-phase Indus settlement locations collated in Possehl. Inset locates figure in south Asia

Fig. 2

Trace of Ghaggar–Hakra palaeochannel on northwestern Indo–Gangetic plain. a Background shows Landsat 5 TM colour composite mosaic (bands 456). The Ghaggar–Hakra palaeochannel is visible as a sinuous, dark blue feature. Location of GS core sites adjacent to the Indus urban centre of Kalibangan, along with core sites at KNL1, MNK6, and SRH5, are also indicated. Location of key Indus urban settlements indicated by triangles. b Geomorphological map showing major alluvial landforms in the study region. Ch, Chandigarh; HFT Himalayan frontal thrust

Fig. 3

Topography of Ghaggar–Hakra palaeochannel. a Detrended relative elevation map of Sutlej–Yamuna drainage divide, derived from NASA Shuttle Radar Topography Mission (SRTM) 30 m DEM (2014 release) showing that Ghaggar–Hakra palaeochannel forms an incised valley. b Corresponding TM colour composite image (detail of Fig. 2) showing correspondence of Ghaggar–Hakra palaeochannel and incised valley. Locations of urban-phase Indus settlements along Ghaggar–Hakra palaeochannel are indicated

Fig. 4

Locations of core sites along Ghaggar–Hakra palaeochannel. Background images are derived from Landsat 5 TM colour composite satellite mosaic shown in Fig. 2. White circles show locations of cores with relationship to Ghaggar–Hakra palaeochannel (dark blue tone). Course of modern ephemeral Ghaggar River is indicated in yellow. a Vicinity of Kalibangan Indus urban centre showing locations of cores GS14, GS13, GS7, GS10 and GS11. Location of Thar Desert modern dune sample also indicated. b Location of core KNL1. Urban-phase Indus archaeological sites in area are indicated by white triangles. c Location of core MNK6. Locations of all drill sites tabulated in Supplementary Table 1

Fig. 5

Stratigraphic panel showing core stratigraphy, sedimentology and K-feldspar OSL ages at GS core sites along transect across Ghaggar–Hakra palaeochannel near to Indus urban centre of Kalibangan. Note how in cores GS7 and GS10, sediments show young OSL ages that are inset into surrounding older strata, indicating that the sediments with young ages infill an incised valley. Additional evidence for this comes from the abrupt age disjunction observed in core GS10 at ~ 16 m depth, which defines the base of the incised valley. Sampling points for U-Pb detrital zircon and ⁴⁰Ar/³⁹Ar detrital muscovite analysis are also indicated. Stratigraphic sections are arranged in elevation. Dashed lines indicate basal fluvial erosion surface (red) and base of youngest incised valley (blue). Note variable horizontal scale. bgl below ground level

Fig. 6

Characteristics of sediments in cores. a–d Detailed sedimentary features of core recovered from GS10 at Kalibangan. Scale bar is 1 cm in all images. a Silty clay at 2 m depth, b interlaminated silt and very fine sand at 4 m depth, c red-brown clayey silt at 6.5 m depth, d grey micaceous fine sand at 17 m depth. e Core recovered from GS7 at Kalibangan at a depth of 10 to 0 m, from the centre of incised valley. Facies abbreviations: F2, red-brown silty clay. F3, red-brown very fine sand. Cm2, yellow-brown very fine sand. C4, grey fine, micaceous sand. Details of facies are summarised in Supplementary Table 2. The base of the section comprises unconsolidated grey micaceous fluvial sands. Above these there is an abrupt transition into brown very fine sands and silts, and toward the top red-brown silty clays indicative of very low-energy depositional environments are present. Sampling points of detrital zircon samples GS7 Zr1-3 are indicated, together with K-feldspar OSL ages

Fig. 7

Stratigraphic panel showing detailed core sedimentology in upper part of GS section across Ghaggar–Hakra palaeochannel at Kalibangan. OSL ages are indicated. Red arrows demarcate top of grey sands indicating termination of major Himalayan fluvial activity in each section. Sedimentary sections are arranged in elevation. bgl, below ground level

Fig. 8

Core stratigraphy, sedimentology and OSL ages at MNK6 and SRH5 drill sites along Ghaggar–Hakra palaeochannel. Sampling points for U-Pb detrital zircon and ⁴⁰Ar/³⁹Ar detrital muscovite analysis are also indicated. Arrows indicate basal fluvial erosion surface (red) and base of incised valley (blue). Note major age disjunction at 16 m depth in core MNK6, indicating a major episode of fluvial incision and defining the base of the incised valley. bgl below ground level

Fig. 9

Age distributions of detrital zircon and muscovite grains for core, modern river, and aeolian dune sand samples. a U-Pb detrital zircon age distributions. Modern Sutlej sand shows a peak at ~480 Ma that is not prominent in Yamuna, Ghaggar and Ganges modern river samples. All fluvial sand samples from cores show distributions that match modern Sutlej river sand, thus identifying Sutlej catchment as the source of the fluvial sand underlying the Ghaggar–Hakra palaeochannel. A palaeo-Yamuna River cannot be ruled out as an additional contributor to GS and KNL1 sands, but cannot be a contributor to SRH5. Only GS11-Zr6 shows a different distribution; this sand is interpreted as an aeolian deposit below the fluvial succession and shows a good match to the modern Thar Desert dune sand. Sample locations shown in Figs. 2 and 4, and detailed in Supplementary Tables 1 and 3. Sample points in cores shown in Figs. 5, 8 and Supplementary Fig. 4, and described in Supplementary Table 4. b ⁴⁰Ar/³⁹Ar detrital muscovite age distributions. Two prominent peaks at ~15–20 Ma and ~4–6 Ma are present in the core samples. Both populations are present in the modern Sutlej sample, but the younger population is not present in the modern Yamuna sample, implying that the Sutlej catchment must be a contributor to fluvial sediments in the core. A palaeo-Yamuna River cannot be ruled out as an additional contributor to the GS fluvial sands but could not have contributed to the SRH5 fluvial sediments. Sample points in cores described in Supplementary Table 5

Fig. 10

Topography of Sutlej–Yamuna plains showing modern Himalayan rivers occupy incised valleys. a Detrended relative elevation map, derived from SRTM 30 m DEM (2014 release), showing courses of the modern Sutlej, Beas and Yamuna rivers confined to regionally extensive incised valleys eroded into alluvial deposits of the Indo–Gangetic basin. Confinement prevents the rivers from readily avulsing across older fluvial fan surfaces. White box indicates area of detailed image in b. b Detail from Landsat 5 TM colour composite mosaic in Fig. 2 showing modern Sutlej incised valley near its outlet at Himalayan mountain front. Inferred palaeo-Sutlej course that joins Ghaggar–Hakra palaeochannel, a former Sutlej incised valley, is indicated, as is the likely river avulsion node