Sediment-transport rates from decadal to millennial timescales across the Indo-Gangetic Plain: Impacts of tectonics, climatic processes, and vegetation cover

Abstract

The Himalayan foreland basin, manifested as the Indo-Gangetic Plain (IGP) on the surface, has been filling up with sediment, carried by the Himalayan river network. Since 1960, several sediment budget records have been established for this region, estimated by different approaches, and averaged over different time scales. Here, we have compiled 196 records of sediment load, sediment yield, and erosion rate published between 1960 and 2022 for different catchments across the IGP. We categorised the data as long-term (multi-centennial to multi-millennial) and short-term (annual to multi-decadal) records, based on the timescale they are averaged upon. We find that there is a clear gradient of sediment yield across the IGP where long-term sediment yield decreases from ∼5800 t/yr/km2 (Sutlej basin) in the west to ∼1200 t/yr/km2 (Ganges basin) in the east and the short-term yield decreases from ∼1800 t/yr/km2 (Sutlej basin) in the west to ∼100 t/yr/km2 (Ganges basin) in the east. Our analysis shows that the driving factors of sediment generation and dispersion vary throughout the region and among the integration timescales. The region south of the Main Frontal Thrust, specifically the IGP, is highly populated and the short-term records suggest an anthropogenic bias. However, we show that only a low but measurable percentage of the annual sediment load is trapped by anthropogenic interventions, such as dams and barrages, when we compare the short- and long-term sediment loads. Among 172 sampled catchments, both short (n = 87) and long-term (n = 60) loads from basins with area >100 km2, show nearly log-linear correlation with their respective basin sizes. However, in case of the Ganga-Brahmaputra basin, the slope of the best fit line for the short-term loads is 30% lower compared to long-term loads. The short-term load estimations for the same basin size are consistently lower than the long-term loads. We attribute this discrepancy on the lag in response time of the vast IGP dispersal system. We note that the variability of long-term erosion rates from west to east correspond well (R2 = 0.77, n = 6) with the variability of mean normalized channel steepness indices for all major catchments to the south of the Main Frontal Thrust region. To the north of the Main Frontal Thrust, the variation of long-term erosion rates correlates well with channel steepness within a single catchment, but not in a similar fashion across different catchments from west to east. We observe that the linear relationship between long-term erosion rate and channel steepness index for individual basins to the north of the MFT is influenced by the presence of high vegetation density; however, any correlation between long-term erosion rates with mean annual rainfall and vegetation cover is absent. Our analyses suggest that short-term records are more susceptible to vegetation cover changes in steep terrains, but landscape erosion rates at these timescales are also driven by various degrees of landscape steepness. We show that short-term records are influenced by transient sediment storages on the IGP that respond to past tectonic and climatic forcing. This underlines the importance of understanding transient sediment storages when interpreting short-term records in the context of recent climatic changes and anthropogenic impacts.