Assessment of Historical and Future Water Availability in Himalayan Tamor River Basin Nepal Utilizing CMIP5 CNRM Climate Model Experiments

: In this study, the assessment of water availability under climate changing 30 environment has been done in the Himalayan Tamor River Basin, Nepal using physically 31 based, spatially distributed, a continuous model 'Soil and Water Assessment Tool' (SWAT). 32 The hydrological simulation and projection have been performed in the historical (1996-2007) 33 and future times (e.g. 30s, 40s, 50s, 60s, 70s, 80s, 90s). The climate change impact assessment 34 on the hydrology of Tamor river basin has been performed utilizing the CMIP5 CNRM climate 35 model datasets (with RCP4.5 and RCP8.5). The model calibration and parameterization 36 uncertainty evaluation in the simulated and projected flows were done in SWATCUP using 37 SUFI2 algorithm. The results obtained from the model calibration (1996-2004) and validation 38 (2005-2007) showed a reliable estimate of daily streamflow for calibration period (R 2 = 0.85, 39 NSE =0.85 and PBIAS=-2.5) and validation period (R 2 =0.87, NSE =0.85 and PBIAS=-5.4). 40 The average annual water yield at the main outlet of the basin is computed as 1511.13 mm, and 41 the total annual quantity is recorded as 6.25 BCM. The average annual precipitation over the 42 seleced river basin is projected to be increased in all scenarios. The stations at higher altitude 43 show more temperature rise than those at a lower elevation and thus there would be minimal 44 snowfall has been projected in the basin by 2100 AD under both scenarios (RCP4.5 and RCP8.5). It is expected that the flow pattern in the future would be similar to the baseline 46 pattern under all scenarios. The baseflow will be dramatically increased in all scenarios, but 47 the lowest flow month would be shifted from March to February. Since the base flow during 48 lean months would be increased in future as projected by all scenarios, there would not be 49 adverse impacts on higher percentile flows. This study would be useful for the assessment of 50 the possibility of storage type or run-off-river type hydro-project in the basin in terms of water 51 availability. 52


22
Nepal Utilizing CMIP5 CNRM Climate Model Experiments 23 1 Bishal Pokhral, 2 Vishal Singh*, 1 S. K. Mishra, 2 Sanjay Kumar Jain, 2 Pushpendra Kumar 24 Singh, 2 Joshal Kumar Bansal  The hydrological simulation and projection have been performed in the historical (1996)(1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007) 33 and future times (e.g. 30s, 40s, 50s, 60s, 70s, 80s, 90s). The climate change impact assessment 34 on the hydrology of Tamor river basin has been performed utilizing the CMIP5 CNRM climate  The average annual water yield at the main outlet of the basin is computed as 1511.13 mm, and 41 the total annual quantity is recorded as 6.25 BCM. The average annual precipitation over the 42 seleced river basin is projected to be increased in all scenarios. The stations at higher altitude 43 show more temperature rise than those at a lower elevation and thus there would be minimal 44 snowfall has been projected in the basin by 2100 AD under both scenarios (RCP4.5 and

55
The water resource is one of an essential input to the growth and well-being of the nation. 56 Nepal is a mountainous country with more than 6000 rivers, with a total average yearly flow 57 of about 225 billion cubic meters (WECS, 2011). As other mountain originated river of Nepal, 58 Tamor river is being planned for its extensive use especially in hydropower generation and 59 highly linked for inter-basin transfer being excess in the basin than demand. Nepal has realised 60 as a national topmost and highly prioritised issue of water resources for the country prosperity 61 and development (Chhetri et al., 2020;Sharma and Shakya, 2006). But the hydrology of the 62 rivers in Himalayan basin is predicted to be more vulnerable because of the seasonal, latitudinal 63 and altitudinal shifting of the freezing line due to the effects of the climate change (Singh et 64 al., 2021a; Jones, 1999). 65 The rate of snowfall can be determined for precipitation with the record of daily air 66 temperature. (Pradhanang et al., 2011). The rising of atmospheric temperature is the leading 67 cause of the declining the quantity of snowfall in the Himalayan regions which further result 68 in the rapid melting of snow-glaciers and reduces the snow cover durations (Singh et al., 2021a(Singh et al., , 2021b; Singh and Bengtson, 2004). The reduction of the snow-to-rain ratio causes wetter 70 monsoon and drier lean flow seasons which has a very adverse effect on the runoff river type 71 hydropower potential of the mountain rivers (Singh et al., 2021b;Agrawala et al., 2003). 72 Nowadays, several hydrological models are available for an efficient assessment of water 73 resources and to find out the impact of climate and soil properties on hydrology and water  The best model is that which gives results proximity to reality with the use of least possible 77 parameters and model complexity (Devia et al., 2015). Among different process-based semi-   the Narmada River basin, India and concluded that the climate change has more influence on 91 water yield and land-use change was affecting more on evapotranspiration and overland flow. Project. However, this is not a hydrological station and does not have a measured dataset of 111 any periods. Thus, the simulated discharge at the outlet obtained from this study would be very 112 useful for the generation of hydropower potential in the selected river basin. Similarly, the 113 predicted future streamflow will further help to understand the future pattern and future water 114 potential of the river. The water availability over the basin has been assessed by computing the

119
Koshi river is the largest river of Nepal, and among its seven major sub-basins, the Tamor sub-120 basin is the easternmost sub-basin that originates from the Himalayas of Kanchanjunga and 121 joins with Arun and Sunkoshi to form giant Saptakoshi which is the major tributary of Ganga 122 River in India. The basin area considered in this study is 4860.5 Km2 from Taplejung, 123 Terhathum and Panchthar districts of Province no. 1, Nepal. The basin has a very diverse 124 climate since the elevation ranges from 350 m to 7526 m above mean sea level within 150 km 125 length. The south-east monsoon is the most dominant climatic influence and is responsible for 126 almost monsoon in the basin. The difference between the warm and humid summer and the 127 extremely cold winter becomes remarkable with the increase in altitude from South to North.

128
The mean annual basin precipitation is around 2305 mm (a result of this study). Although, most 129 of the precipitation for this region is concentrated during the monsoon months, the snow-

144
It is always essential to use all relevant and high-quality data to obtain a good result. The 145 essential input datasets are digital elevation model (DEM), soil map, land use and land cover 146 (LULC) map, slope map, weather generator data and observed discharge were used for the 147 SWAT modeling simulation and calibration ( Figure 2).

148
In this study, the SRTM DEM of 30m × 30m resolution has been used to delineate the streams,    deviation of the measured data). SUFI-2 tends to cover most of the measured data with the 212 smallest possible uncertainty band. (Abbaspour, 2015). In general, hydrological models such 213 as SWAT incorporate many parameters, but only a few of these parameters have sensitive 214 impacts.
SWAT-CUP has two options for sensitivity analysis, and they are all-at-a time (AAT), i.e.

216
global and one at a time (OAT) sensitivity analysis. In this study, a global sensitivity analysis 217 was chosen since the AAT produces more reliable results than OAT (Abbaspour, 2015). Global 218 sensitivity analysis is determined based on t-stat and p-value (Abbaspour, 2015). At 95% 219 confidense interval, as per the t-stat test, if the value of parameter exceeds >2.0 or less than -220 2.0, the parameter can be defined as the significant sensitive. While, in case of p-value test,
Eq. 3 Where, Q is the variable and 'm' stand for measured and 'S' stands for simulated values, bar 231 stands for average, and i is the i th measured or simulated variable. The statistics recommended 232 by (Moriasi et al., 2007) will be used in this study to define SWAT model performance ratings.   The simulated daily discharge data provided by the model in each iteration was further  surface runoff can be seen in these subbasins such as subbasin 1 to subbasin 8 (Figure 6b).

380
From the above table, it was also clear that the water yield may not be highest for the sub-basin 381 with the highest runoff.

382
After a successful calibration of the model by using SWAT-CUP and re-run of the calibrated 383 model in Arc-SWAT model for the study period, a baseline flow for the calculation is received.

384
Precipitation and temperature inputs were provided for the different study period in future 385 whereas remaining inputs were used available by SWAT auto simulation. Table 3 here**

387
All precipitation and temperature stations within the basin used by the SWAT model were 388 analysed after bias correction. The precipitation and temperature datasets were corrected with 389 reference to the observed data and a comparison has made against the data obtained from the 390 model as well as that obtained after bias correction. After bias correction, the R 2 for both 391 precipitation and temperature is increased, and the RMSE is decreased. Further, the bias-392 corrected data showed improved mean and standard deviation than raw data. Table 3 presents 393 the mean, R 2 , RMSE, and standard deviation (Std. Dev.) of the observed data, raw data 394 (downloaded data before bias correction), and corrected data of Taplejung stations of the basin.

395
The result for other stations also showed the same pattern and trend. Figure 7a and Figure 7b  Precipitation is likely to be decreased in most of the scenarios from April to June and in 420 December, but the relative decrease percentage will be quite low with compared to that of 421 increasing periods. The result summarises that the mean monthly precipitation of the basin will  (Table 5). The worst-case among all scenarios would be for a change in average 465 annual minimum temperature in the 2090s by RCP8.5 where the value from the observed 466 baseline period would be shifted from -2.50C to +5.30C. Table 5 here** 468 The result shows that the percent rise in temperature will be very high in the stations at higher 469 altitude. So, it is expected that the increased temperatures would push the permanent snow line 470 northern upwards to the higher altitude and less precipitation would take place in the form of 471 snowfall. Obviously, the increment is higher under RCP 8.5 than under RCP 4.5 for all decades.

472
And it is also found that under RCP8.5, the per cent change in temperature in the 2060s will be 473 even more than that in 2090s under RCP4.5. Further, the daily minimum temperature will be  Table 6. 485 **Insert Table 7 here** 486 The changes in major water balance components mentioned above, with relative to the baseline 487 observed data (Figure 10). The result shows that the average annual basin precipitation and shows that the average annual water quantity will be increased in the 2030s and 2060s but 501 would be slightly decreased in 2090s under both scenarios (Figures 10c and 10d).  Table   551 8. The analysis shows that the firm flow corresponding to 90% dependable flow at the outlet is Hydropower projects are designed based on Q40 in Nepal, and Q40 flow is projected to be 569 increased on the future, which would enhance the hydropower production capacity in future. observations showed that the average monthly maximum and average monthly minimum 578 temperature for all future scenarios than the baseline period for all the stations will rise 579 significantly. The percent rise of temperature will be more for the stations at a higher altitude.

580
The average maximum and minimum temperature of Taplejung station at 1732 m are projected 581 to be increased by 3.800C and 3.820C. Further, the percent rise would be more for the minimum 582 temperature than the maximum temperature. It is obvious that the RCP8.5 hits more adversely 583 than RCP4.5. It is predicted that the percentage change in temperature in the 2060s under 584 RCP8.5 would be even more than that in 2090s under RCP4.5. The increase in termperature will have significant impacts on Tamor river basin, especially over the Northern region which 586 is mostly corresponded to snow and glaciers.

587
The result of the study shows that the average annual water quantity would be increased in the not be many impacts on higher percentile flows. Excluding the probability of exceedance from 610 3% to 33%, all percentiles in future scenarios are predicted to be increased than that on baseline 611 flow. Generally, Run of River type Hydropower projects are designed based on Q40 in Nepal, 612 and Q40 flow is projected to be increased on the future, which would enhance the hydropower 613 production capacity in future.

614
The assessment of water availability at the main outlet of the basin would be very useful for    Flow chart for the water balance study of the baseline and the future periods.           Percentile monthly ow at Majhitar outlet for observed and future scenarios.