There have been numerous efforts to reconstruct palaeohydrology of the Sutlej Yamuna interfluve in north-west India, which is no longer visible on the ground. A lot of attention have been given to the trace and catchment area of the Ghaggar-Hakra ancient river-bed, which has often been linked to the Indus Civilisation. Interest in the reconstruction of the paleo-hydrology of the Indus region goes back to the nineteenth century, but the increased availability of satellite imagery during the last 40 years has boosted interest in the subject. Although there has been many analyses attempted, previous remote sensing-based reconstructions of the palaeoriver networks of north-west India in particular have commonly employed single or limited time-span imagery, which limits their capacity to identify features in complex and varied landscape contexts. In particular, seasonal and environmental variability, spatially and temporally changing cultivation patterns and large-scale landscape modifications can probe unsurmountable barriers for the meaningful application of Remote Sensing in such a large area.

Figure 1_blog
Figure 1: Multitemporal composite of vegetation indices in the area between Patiala and Bathinda

To overcome these problems TwoRains has implemented a seasonal multi-temporal approach to the detection of palaeorivers over large areas. We have made use of twenty-eight years of Landsat 5 data – and a total of 2266 multi-spectral images, which have been bulk processed using Google© Earth Engine Code Editor and cloud computing infrastructure. An open-access journal article detailing the methodology and results obtained has been published in the journal Remote Sensing, where it can be accessed free of charge ( The code employed to generate these analysis has also been made publicly available (as supplementary material in the article) with the hope that other researchers can employ the same methods in their investigations.

The particular methods employed include:

(1) Long-term analysis of seasonal vegetation variability using an algorithm developed purposely for this research: Seasonal Multi-Temporal Vegetation Indices (SMTVI). This algorithm makes use of the seasonal vegetation response to soil surface and subsurface conditions over the last 28 years (figure 1); and

(2) Seasonal, multitemporal spectral decomposition techniques in particular Tasselled Cap Transformation and Principal Component Analysis (figure 2). These have been most useful in areas were vegetation did not have a much seasonal variability.

Figure 2_blog
Figure 2: Comparison of different seasonally filtered techniques for the analysis of multispectral data


As predicted in my previous bog post, the use of multitemporal data has been very successful and has allowed us to overcome seasonal cultivation patterns and long-term visibility issues related to recent crop selection, extensive irrigation and land-use patterns. The application of this approach on the Sutlej-Yamuna interfluve, has enabled the reconstruction of an unsuspectedly complex palaeoriver network comprising more than 8000 km of palaeochannels (figure 3). These results prove that the factors influencing water availability along the Ghaggar-Hakra basin are much more complex than previously thought. The palaeorivers that have been identified cover a large part of the northern sector – forming an almost continuous parallel pattern, which points to the changing nature of these channels and the likelihood that floods and avulsions were relatively common. The waters feeding the various palaeo-rivers originated from glacier-fed sources as well as monsoonal rain which is likely to have contributed to both perennial and ephemeral rivers, and the geographic source of watercourses ranges from the Himalayas to the Aravalli mountains. All these factors combine to create an extremely complex picture in which water availability and location is dependent upon a multiplicity of factors and is difficult to predict in the long term.

Figure 3_blog
Figure 3: Palaeorivers detected in the study area combining different seasonal multitemporal analyses

The results obtained have also enabled the definition of the morphology of these relict courses, which provides insights into the environmental conditions in which they operated. These new data will contribute to a better understanding of the settlement distribution and environmental settings in which this the Indus Civilisation operated. Next steps include getting more dates for when water flowed in these channels, and the accurate identification and location of archaeological sites across this vast area. Stay tuned.

To read the full paper, please visit: