Up the creek without a paddle: Intra-annual and longer term changes to river discharge in the Himalayas

Continuing from last weeks post on GLOFs, this post shall review the impacts that glacier distribution and mass may have on discharge of rivers whose headwaters are situated in the Himalayas. Supporting one-third of the world’s global population (Singh  et al., 2006) changes in glacial melt will alter the discharge of these rivers on both intra-annual and also longer-term timescales.

CONTRIBUTION FROM GLACIAL MELT

It is estimated that the River Ganges is replenished by meltwater from approximately 4,000 glaciers and over 3,300 are calculated to contribute to the annual discharge of the River Indus (Thayyen and Gergan, 2010). Rees and Collins (2006) demonstrated that Himalayan glaciers contribute to both intra-annual variations in river flow and also in addition to precipitation, longer-term changes in annual river discharge. The contribution of a glacier to river flow is basin-specific with the glacial melt component comparatively higher in glacierised basins than less glacierised basin, with the latter receiving a greater contribution from surface runoff (Rees and Collins, 2006).

However, although the extent glacial coverage in the basin will undoubtedly contribute to the initial contribution, and thus subsequent change, of the glacial melt component to the river discharge, the significance of the affect of changing glacier discharge will also depend on surface runoff input by precipitation. This applies to both intra-annual and longer-term river discharge variability.

INTRA-ANNUAL VARIABILITY

Thayygen and Gergan's (2010) study of the Dokriani glacier in Garwhal, Himalaya monitored the summer variability of runoff at three hydrometric and meterological stations at different altitudes in the basin between 1998-2004 (from 3,400masl just below the glacier snout to 2,360masl in an unglacierised region further down in the valley).  Over this six-year period, summer streamflow gradually decreased from 290 x10⁶m³ in 1998 to the lowest measured discharge of 123 x 10⁶m³ in 2004 (Thayyen and Gergan, 2010). Discharge from the uppermost station, 600 metres below the glacier snout also recorded a 50% decline in discharge over the same period. However, observations at the lowest station (Tela) indicate that the glacier component to the total river discharge downstream nearly doubled, from 18% in 1998 to 34% in 2004. Thus contribution of the glacial component to river flow increased despite the actual discharge from the glacier decreasing. Comparison against precipitation values between 1998-2004 showed a fall in summer precipitation. Therefore, glacier melt was acting as a buffer against reduced surface runoff, augmenting the years of low summer flow (Thayyen and Gergan, 2010). This is an key point as it highlights the importance of glacial discharge to maintaining river discharge during low flow periods, thus supplying communities with a freshwater source when the demand is the greatest during the summer months. This is underlined by the lowest specific runoff values recorded in the study with the lowest specific yield from the glacierised catchment 15mm day^-1, 4mmday^-1 higher than the non-glacierised (precipitation-dependent) catchment (Thayyen and Gergan, 2010).

Singh et al. (2006) also demonstrated the intra-annual importance of glacier melt to river flow. Applying the SNOWMOD snowmelt model to same glacier, Dokriani, between 1997 and 1998, the model showed a linear increase in stream discharge with a rise in temperature for both of years, with glacial melt and rainfall contributing 87% and 13% respectively. The model also highlighted variation in runoff within the summer months with the largest percentage changes in runoff occurring in September and the smallest in July in 1997, whilst the reverse was recorded in 1998. Related to 1998, the highest percentage change in June (23%) attributed to a later onset of the monsoon (Singh et al., 2006). However, in the subsequent months, the impact of precipitation changes decrease and the glacial component becomes more important and thus air temperature, which influences glacial melt becomes the main factor. Using the limited duration of data available, the model projected that a 2°C rise in temperature would cause summer discharge to increase by approximately 28% (Singh et al., 2006).

LONGER TERM VARIABILITY

In addition to augmenting low flow during the summer months, changes in glacier melt will affect long-term annual discharge of rivers with headwaters in the Himalayas.

Bhutiyani et al., (2008) studied the flow of four rivers in the northwestern Himalaya from 1961-2004, with an extended study for one of the rivers, Satluj River, from 1922-2004. In the absence of specific catchment data, regional temperature and precipitation were used to analysed the changes in discharge in each of the rivers obtained from hydrographs. Initial observations from the basins showed insignificant changes in the relatively less glacierised Beas and Ravi river discharge, whilst Chenab, the most glacierised catchment showed a significant increase in discharge between 1961-2004. However, precipitation from the summer monsoon in the Beas catchment decreased, demonstrating that glacial melt is again buffering against reductions in river flow. Temperature in the northwest region of the Himalayas has being warmed by approximately 4.4°C in the winter months in the last two decade (Bhutiyani et al., 2008). This has caused an increase in glacial melt as the transient snowline increases in altitude causing greater ablation rates at the glacier snout. The short-term implications of increased glacial discharge, as observed in Beas, Ravi and Chenab is either a maintenance of river discharge as precipitation decreases or, as in the case of Chenab where precipitation was relatively constant, an increase in river flow.

Although in the short-term (four decade) period of these three rivers, glacial retreat causes a rise in stream discharge which would be beneficial to local communities, the longer term (1922-2004) study at Satluj shows a reversal of this increase in a longer time frame. Between 1922-2004 the river discharge fluctuated from above normal to below normal discharge (Bhutiyani et al., (2006). From 1945-1990 above normal discharge was observed within the Satluj river. Whilst the initial rise (1945-67) coincided with relatively static temperature changes and a rise in precipitation, the later phase (1968-90) occurred during a period of increasing temperature and a decline in monsoon precipitation suggesting glacial melt was buffering river discharge against falling surface run-off input (Bhutiyani et al., (2008). Since 1990, discharge has begun to decrease. Bhutiyani et al., (2008) argues that this decline is due to a reduction in the glacier-melt component of discharge which reached its maximum during the 1968-1990 period and has now thinned considerably. The key observations from the longer-term study suggest that in the short term, assuming precipitation remains constant, river discharge will increase. However, in the long-term once glaciers have reached a maximum, their contribution to river flow will begin to decline and subsequently river discharge will also decrease. The current observed weakening of the monsoon will only exacerbate river discharge changes further, with a reduction in flow likely to occur sooner as glacial melt is augmenting against reductions in surface runoff input. This will have significant impacts on communities within the Himalayan region, and will be expanded upon further in my next post.

REGIONAL DISPARITIES.

The complex regional climate with the Himalayas, mean the impacts of future warming on river flow will vary across the mountain chain, The glacio-hydrological regimes in the Himalaya differ related to the decrease in the influence of the monsoon from the east (monsoonal) to the drier west (Rees and Collins, 2006). Using a temperature-index based model, Rees and Collins (2006) modelled the changes in river discharge of two hypothetical catchments, one in the west and one in the east. By using hypothetical catchments, the study could maintain other parameters such as glacial coverage and basin size that could otherwise influence the response of the river to changing glacial melt rate. A baseline scenario using climate data from 1961-90 and a warming scenario of 0.0.6°C were run for a 150 year period starting from 1990. As shown in Figure 1, in the short-term river discharge increases due to rising temperatures increasing glacial melt discharge cause the hypothetical glacier to retreat. Once the rate of new ice exposed by the increasing snowline is no longer offset by the melt of the glacial ice below this line, then the glacial discharge begins to decrease indicated by a fall in river discharge (Rees and Collins, 2006). The diagram also shows that the impacts of the retreating glacier are more distinct in the drier western catchment (Figure 1a), and the initiation in the decrease is sooner and rate comparitively greater in the western catchment compared to in the east where the monsoon suppresses the changes in glacial melt.



Figure 1: Variation in annual mean flow under the 0.06°C yr^-1 warming scenario for a)the western catchment and b) the eastern catchment (Source: Rees and Collins, 2006:2166)



CONCLUSIONS

Review of some of the current literature on hydro-glaciological regimes in the Himalayas demonstrates the important contribution of glacial-melt to both intra-annual and longer-term changes in Himalayan river flow. In addition to augmenting periods of low summer flow, glacial melt also buffers against yearly variations in precipitation across the mountain chain. In the short-term, rising temperatures are anticipated to cause an increase in the glacial melt component to river discharge, with the subsequent change in short-term river discharge (remaining static or increasing) depending on the precipitation regime in that area (declining or stable/increasing). Intra-annual changes to river discharge will also depend on summer precipitation, and under the current pattern of a weakening monsoon, glacial melt is becoming increasing important in augmenting against low summer flow. If current global warming continues, in the long term it is predicted that after a peak in river discharge, discharge will subsequently begin to fall, a pattern simulated in both the east and west Himalaya. This will have significant implications for countries bordering the Himalayas, with over one-third of the world’s population currently dependent on Himalayan rivers for freshwater, and this value is growing. Therefore, changes to river discharge are of growing political and social concern and significant investment will be needed to prepare for declining river discharge in the future.   

Reference:

Bhutiyana, M. R., V. S. Kale and N. J. Pawar (2008) ‘Changing streamflow patterns in the rivers of northwestern Himalaya: Implications of global warming in the 20^th century’ Current Science, 85, 5: 618-626.

Rees, H.G. and D. N. Collins (2006) ‘Regional differences in response of flow in glacier-fed Himalayan rivers to climatics warming’, Hydrological Processes, 20, 10: 2157-2169.

Singh, P., M. Arora and N. K. Goel (2006) ‘Effect of climate change on runoff of a glacierised Himalayan basin’, Hydrological Processes, 20: 1979-1992.

Thayyen, R. J., and J. T. Gergan (2010) ‘Role of glaciers in watershed hydrology: a preliminary study of a “Himalayan catchment”, The Cryosphere, 4: 115-128.