Effect of changes in climate variables on hydrological regime of Chenab basin, western Himalaya

In high altitude, scarcely gauged basins, climate change impact assessment on river discharge is important for sustainable management of water resources. These basins are sources for irrigation and hydropower generation in the region. Expected changes in precipitation and temperature can affect the basin ’ s hydrological regime which will have consequential impacts on the dependent sectors. For quantifying the impacts of major climatic variables on hydrological processes, this paper examined bias-corrected GCM outputs coupled with a hydrological model – HBV for Chenab basin. Trend analysis shows that precipitation would decrease after the short-term period and temperature is expected to increase throughout the century. Simulated river discharge is expected to increase throughout the 21st century under both RCP 4.5 and RCP 8.5 scenarios. It is also observed that there would be a shift in seasonal discharge patterns with increased pre- and post-monsoon contributions. Increase in snow and ice melt contribution to the overall discharge is also expected and would range between 50 and 59% until 2100. This study concluded that expected increase in discharge volume coupled with shift in seasonal discharge pattern would impact the basin water management and thus it is important to consider the impact of climate change on the hydrological regimes of basins.


INTRODUCTION
Increasing surface temperature and variable precipitation patterns are likely to occur over the next century and the impacts of climate change are of major concern worldwide (Bajracharya et al.  (Jeelani et al. ) and these changes will impact the hydrological regime by affecting the timing and volume of river discharge in the region (Uprety et al. ). The Indus basin, one of the most important transboundary basins of the region, has significant discharge contribution from snow and ice melt. With the warming trend, the Indus basin in particular is more vulnerable to climate change (Shrestha et al. ). It is thus imperative to understand the potential impacts of climate change on the hydrological regime of the basins in the Himalayan region for sustainable water resources management. Especially, the Indus basin, which sustains the lives of millions of people, requires proper planning and management of water resources with due consideration of climate change (Lutz et al. ).
In this context, a study analyzed the CMIP5 global climate model and reported an increase in maximum and minimum temperature in the Indus basin region, under both RCP 4.5 and RCP 8.5 scenarios for the time period of the 2030s and 2070s with respect to the baseline period of 1971-2005 (Gebre & Ludwig ). In this study, the multimodal results of average monthly maximum temperature ranges from 1 C to 7 C indicating an increasing trend for both the periods under both RCP scenarios. A similar pattern was observed for minimum temperature. Another study by Su et al. () analyzed CMIP5 outputs for three 4.5,and 8.5 for mid and late century and suggested increasing temperature in a consistent manner over the entire basin. Under the RCP 2.6, RCP 4.5, and RCP 8.5 scenarios, in the mid-21st century, annual mean temperature will increase by 1.21 C, 1.93 C, and 2.71 C, respectively. Precipitation projection is more variable than temperature. A study by Huang et al. () reported that the trend for annual precipitation over the entire basin, except some parts in the north and south, is decreasing for the three RCPs scenarios (RCP 2.6,RCP 4.5,and RCP 8.5 (Anjum et al. ). Also, it is important to understand the expected changes in the hydrological regime of the basin, which is the source of hydropower production, because any alteration in the hydrological regime will have immediate consequences for the hydropower projects (Schaefli ).
This study aims to assess the impacts of climate change on the hydrological regime of Chenab basin which is one of the westward flowing tributaries of the Indus basin. To achieve this objective, we used the semi-distributed Hydrologiska Byråns Vattenbalansavdelning (HBV) hydrological model forced with bias-corrected downscaled CORDEX data for the South Asia domain using the IPCC scenarios RCP 4.5 and 8.5. The model simulations were used to estimate discharge for Chenab basin for short- (2007-2035), mid-(2036-2065), and long-term (2066-2100) future scenarios. This paper also investigates the trends of bias-corrected projected temperature and precipitation changes in the future and simulated changes in snow and ice melt contribution to the basin.

Study area
Chenab River is one of the five main rivers of the great Indus system and is formed by the merging of two streams, the Chandra and the Bhaga at Tandi, and traverses through the state of Himachal Pradesh before passing through the state of Jammu and Kashmir. The Chenab basin is located between 30 N and 34 N and 74 E and 78 E (Jain et al.  The baseline or historical CORDEX data for the period 1971-2005 and future data for the period 2006-2100 are used for this study. CORDEX data for South Asia domain (WAS-44i) for two experiments, RCP 4.5 (medium stabilization scenario) and RCP 8.5 (high baseline emission scenario), were procured for two variablesdaily precipitation and daily average temperature. The procured datasets (historical and future projections) were processed in R software, clipped to the region of interest (study area basin boundary) and converted to ASCII format to be used as input for the HBV model. Change factor is a simple and popular procedure for the bias correction method (Diaz-Nieto & Wilby ). This is a simple downscaling method using the average values of observations and predictions. Change factor is widely used due to its simplified approach where average change factor is scaled to each day. There are limitations attached to this bias correction method as well. However, studies indicate that all bias correction methods are able to reduce biasness to some extent, with variable performances depending on the location of basin and choice of method (Luo et al. ).
In the current study, the change-factor method, the most commonly used method for bias correction, has been applied. This method estimates the change coefficient by indicating the difference between future predictions and baseline observations and adjusts the baseline observations using this coefficient to reflect a future climate.
In this study, the change factor for precipitation and temperature was estimated using Equations (1) and (2), respectively, and used to scale the daily observed precipitation and temperature data from IMD to produce future climate time series: where P obs,d is observed daily data procured from IMD;  The MK test is widely used for the analysis of trend in climatologic and in hydrologic time series (Khaled ).

CLIMATE CHANGE AND FUTURE PROJECTIONS
Since the test is a non-parametric one, it does not require the data to be normally distributed. In addition, the test is even suitable for inhomogeneous time series as it is less sensitive to abrupt breaks. The null hypothesis (H 0 ) for this test is that there is no trend (the data are independent and randomly ordered), tested against the alternative hypothesis (H 1 ), which assumes that there is a trend.
where T j and T i are the annual values in years j and i, j > i, respectively.
For n ! 10, the statistic S is approximately normally distributed and the test statistic Z s is calculated to measure the significance of the trend. If p-value is less than the significance level α (alpha) ¼ 0.05, H 0 is rejected and the result is said to be statistically significant with the presence of a trend in the time series. If the p-value is To further estimate the magnitude of the trend, Sen's slope estimator was used. Sen's slope is a non-parametric method that is not much affected by gross data errors and outliers. The Sen's slope (β) is calculated as the median of all the slopes estimated between all the successive data points time series (x) as: where Δy is the change in climate variable due to the change in time, Δt between two subsequent observations.
The MK test results (  CORDEX data indicated significant trends for average temperature for mid-and long-term projections, while for precipitation the trends were significant mainly for shortterm projections, which means that in the long run, the basin is expected to have a higher temperature but no   for short-, medium-, and long-term periods, respectively.
The results indicate that the discharge in Chenab River by mid-century will be around 1.5 times the present discharge under RCP 4.5. However, the rate of increase stabilizes thereafter until the end of the century, while the discharge in Chenab River will be almost double its present value by the end of the century under RCP 8.5.
Correlation of discharge, temperature, and precipitation data (Figure 6(a)-6(d)) indicates that initial increase in discharge could be attributed to the increase in precipitation The simulated future discharge at the outlet of the basin is expected to increase significantly throughout the century, with monsoon season continuing to be the peak discharge period.
Projected precipitation for monsoon season indicates a majorly decreasing trend for all the time series for both the RCPs (except short-term in RCP 4.5), but an increase in volume of monsoon discharge is observed, which can be attributed to the increased contribution from snow and ice melt. This is further supported by the projected increase in temperature in all time series. Projected temperature increase for the monsoon season with respect to the baseline is expected to be in the range of 4-12% for RCP 4.5 and 7-24% for RCP 8.5.  contribution is expected during 2031-2065 and is around 56% of the annual discharge, and remains within this range until the end of the century. The contribution during the short time series is around 50%, which is almost equivalent to the baseline period. Under RCP 8.5, snow and ice melt contribution increases from baseline to short-and medium-term periods to 54-55%, but reaches its peak contribution of 59% of annual discharge towards the end of the century.
Overall, by the end of the century, snow and ice melt contribution is significantly higher as compared to the base- Increasing melt contribution in winter months not accompanied by an equivalent increase in snowfall also signifies that net snow accumulation in the basin may be approaching towards a net ablation scenario during the winter period by the end of the century.

CONCLUSIONS
This study assessed the impact of climate variables (temperature and precipitation) on major hydrological components (discharge, snow and ice melt contribution, seasonal discharge) in the high altitude, scarcely gauged Chenab basin in India, by examining bias-corrected GCM outputs coupled with a hydrological model. The trend analysis results of projected climatic variables show that precipitation is largely expected to decrease after the short-term period while temperature is expected to increase throughout the century, over the basin.
Hydrological simulation under climate change scenarios for both RCP 4.5 and RCP 8.5 scenarios indicates increasing discharge compared to the baseline period throughout the 21st century; however, discharge attains its peak by mid century under RCP 4.5 but continues to rise until the end of the century under RCP 8.5. A shift in seasonal discharge pattern is also observed with increased pre-and post-monsoon contributions in annual discharge. Further, an increase in temperature also indicates a consequential impact on snow and ice cover of the basin, resulting in their increased contribution to the Chenab River flows throughout the century.
Increase in discharge volume as well as its shifting seasonality might have an impact on the basin water management, irrigation, reservoirs, and hydropower projects. An increase in water could present a threat of flooding in low-lying areas and this needs to be managed in advance by creating storage reservoirs.
Hydropower projects need to be better prepared for climate change-induced variability in discharge pattern and associated events such as flooding to keep their economic feasibility viable. Run-of-river projects operating and planned over Chenab basin are also vulnerable to variation in discharge, especially increase/shifts in peak flows. Thus, hydrological simulations under climate change scenarios provide valuable information about future discharge patterns which are useful to prepare well in terms of water management.

CONFLICT OF INTEREST
We wish to confirm that there are no known conflicts of interest associated with this publication and there has been no significant financial support for this work that could have influenced its outcome.