New Studies: Indus River flow variability and trends


Annual and decadal changes in stream flow have important implications for food production, hydroelectric power generation, and basic water availability. The Indus River, which is sourced from glacier melt, snowmelt, and rainfall in the mountains of Pakistan, India, and China, supplies water for a complex irrigation system that yields 90% of the food produced in Pakistan. Approximately 85% of the total discharge in the Indus River Basin occurs between May and September. Two recent studies (Sharif et al. 2012; Cook et al. 2013) contribute to our understanding of the variability and trends in discharge in the Upper Indus River Basin.

Using tree-ring records collected in the Upper Indus River Basin, Cook et al. (2013) reconstructed more than five centuries of summer (May–September) stream flow in the Indus River. The reconstructed stream flow records suggest that large decadal changes in stream flow are normal in the Upper Indus River Basin; however, the causes of these past changes are not identified. The study found that reconstructed stream flows were 8% lower than the overall mean in the 112-year period between 1572 and 1683 and 11% lower than the overall mean during the shorter 27-year period between 1637 and 1663. A repeat of such declines in the future would have a significant impact on the availability of water for irrigation and hydroelectric power generation in the region, and ultimately on the residents.

Indus River
Indus River
Photo Credit: Alex Treadway

However, the reconstruction by Cook et al. (2013) needs to be considered carefully. Indeed, it is based on a statistical analysis calibrated over a relatively short and recent period from 1975–2004 (30 years) and validated between 1962 and 1974 (13 years). The resulting empirical relationship is then applied to reconstruct the Upper Indus River Basin discharge since 1452, the age of the oldest tree ring analysed so far. This method provides an understanding of climate evolution in this region; however, whether or not this empirical relationship between tree rings and Upper Indus River Basin discharge remained stable in the past cannot be accurately confirmed. Consequently, the results obtained through this study must be considered highly uncertain. Furthermore, the approach used in the study is based on mean summer flows and does not resolve large-scale flood events, such as the Indus River floods of 2010.

The 40-year observational record of Indus River discharge at the Partab Bridge exhibits a significant increase in mean summer flows after 1988 (Cook et al. 2013), but such changes are not unprecedented in the context of the 500-year reconstructed stream flow records. In contrast, Sharif et al. (2012) find that stream flows in higher-elevation basins, which contain a greater proportion of glacier melt, show trends of decreased summer discharge and increased spring discharge. These two findings are compatible with recent observations of advancing or stable glaciers in the Karakoram region (Gardelle et al. 2012) and increases in winter precipitation and decreases in summer temperatures in the Upper Indus River Basin (Fowler and Archer 2006; Bocchiola and Diolaiuti 2012). Anomalously high snowfall totals will lead to increases in overall discharge, but in high-elevation basins a thick snow cover and decreased temperatures will limit ice melt contributions to stream flow. Caution is urged, however, in connecting the tree-ring chronology with the discharge record and recent glacier change, given the uncertainty in the methods used and the different timescales of these phenomena.

While these two recent studies highlight the ongoing hydrological changes in the Upper Indus River basin, the mechanisms responsible for the observed climatic changes and subsequent response in glaciers and stream flows in the Upper Indus River Basin remain unknown. Ongoing research by ICIMOD and its partners will endeavour to advance our understanding of glacier mass balance, recent glacier change, and precipitation gradients in high-altitude basins and improve the hydrological models used to explore future changes in Upper Indus River Basin discharge.


  • Bocchiola, D; Diolaiuti, G (2012) Recent (1980–2009) evidence of climate change in the upper Karakoram, Pakistan. Theoretical and Applied Climatology doi:10.1007/s00704-012-0803-y
  • Cook, ER; Palmer, JG; Ahmed, M; Woodhouse, CA; Fenwick, P; Zafar, MU; Wahab, M; Khan, N (2013) ‘Five centuries of Upper Indus River flow from tree rings.’ Journal of Hydrology 486: 365–375, doi:10.1016/j.jhydrol.2013.02.004
  • Fowler, HJ; Archer, DR (2006) Conflicting signals of climatic change in the Upper Indus Basin. Journal of Climate 19(17): 4276–4293
  • Gardelle, J; Berthier, E; Arnaud, Y (2012) Slight mass gain of Karakoram glaciers in the early Twenty-first Century. Nature Geoscience 5(5): 322–325
  • Sharif, M; Archer, DR; Fowler, HJ; Forsythe, N (2012) ‘Trends in timing and magnitude of flow in the Upper Indus Basin.’ Hydrology and Earth System Sciences Discuss 9: 9931–9966