Changes in flow and sediment load of poorly gauged Brahmaputra river basin under an extreme climate scenario


 Brahmaputra River Basin (BRB), the largest contributor of sediment load in Ganges–Brahmaputra–Meghna delta, is highly vulnerable to future climate change. Several studies assessed the effects of climate change of BRB on river flow but an assessment on sediment load has not been conducted. Changes in sediment load in the future need to be assessed to control and manage sediment flows in large catchments properly. The present study focuses on developing a hydrological and sediment routing model of BRB using the HEC-HMS model to estimate future sediment load together with the flow for the RCP 8.5 climate scenario. Modified Universal Soil Loss Equation and Engelund Hansen method of HEC-HMS have been applied for the sediment transport of BRB. The model has been calibrated using daily runoff for the period 1983–1996 and validated for the period 1997–2010, respectively. The uncertainty in the percentage change in seasonal sediment load during the pre-monsoon season is higher than that of the monsoon season. However, the contribution of the sediment load of pre-monsoon is very much lower than the monsoon season. The percentage changes in mean annual sediment load compared to the baseline period for the 2020s, 2050s and 2080s are 34, 67 and 115%, respectively.

Ghosh & Dutta () used a single projected climate scenario and found that the projected climate change can increase the peak flow of the Brahmaputra by about 28%.
In a study by Gain et al. (), a daily discharge time series has been constructed by using a discharge-weighted ensemble based on the inputs from the 12 General Circulation Models (GCMs). Then, these data have been used to a global hydrological model and the results show that climate change is likely to improve dry season conditions in the Lower BRB. Ghosh & Dutta ()  for the early century , mid-century  and end century . In this study, the change in monthly flow for August has increased to 4, 5 and 8% for the years of 2020s, 2050s and 2080s, respectively. Darby et al. () found an increase of 25-30% in sediment load from the BRB considering variants of the SRES A1B emissions scenario. Alam () analyzed the impact of climate change on streamflow of BRB using a physically-based semi-distributed hydrological model, SWAT. Another study by Alam et al. () investigated the sensitivity of the discharge of the Brahmaputra river to climate change and found that changes in discharge are almost linearly related to changes in precipitation and temperature. An increase in mean annual streamflow in the 2020s, 2050s and 2080s for BRB has been found in this study. Considering land use and climate change scenarios, Pervez & Henebry () assessed changes in the future freshwater availability in the BRB. Mohammed et al. () considered bias-corrected data from an ensemble of 11 climate projections with the Representative Concentration Pathway (RCP) 8.5 from the Coordinated Regional Climate Downscaling Experiment-South Asia (CORDEX-SA) database. They found an increase and a decrease in mean monthly discharges during pre-monsoon months and postmonsoon months, respectively, for the BRB. Results show that floods are likely to become more frequent in the future, and their magnitude will also become more severe.
The average timing of floods is projected to shift earlier compared to the present hydrological regime.
To make the best use of water resources of the BRB, engineers, as well as policymakers, need to know the future changes in both sediment load and flow. However, little is known regarding the impact of climate change on the sediment regime of the BRB. Sediments contain essential nutrients and material for ecosystems and agricultural lands and it is vital from environmental, economic and social perspectives (Apitz ). The quantity and quality of sediments are affected due to natural variability in hydrological conditions, as well as changes in land use, water use and climate (Chalov et al. ). Changes in sediment loads in the future considering alteration of climatic variables need to be assessed to control and manage sediment flows in large catchments properly, especially where people's livelihood depends on the river systems and their natural processes. Therefore, it is obvious to understand how a change in the global climate could affect regional water supplies and the amount of sediment load. A one-dimen-  : 1982-2000) and Flood Action Plan 24 (period: 1993-1995) data, plotted together with secondary data and found a downward trend This study focused not only on changes in discharges from the BRB in the 2020s, 2050s and 2080s but also on the sediment yield from this basin in the future considering climate change. The RCP 8.5 climate change scenario, which represents the highest CO 2 concentrations in the atmosphere and the highest rise in global temperatures out of the four concentration scenarios defined within the Fifth Assessment Report (AR5) of the IPCC (IPCC ), has been considered for impact analysis on the flows and sediment load of the BRB. The key objectives of this present study are to (1) calibrate and validate the HEC-HMS model for the BRB using available observed data of the baseline period, (2) use the calibrated models to simulate future river discharges and sediment load for the respective climate projections, (3) analyze the simulated future discharges to estimate the changes in monthly, seasonal and annual flow and (4) analyze the simulated future sediment loads to estimate the changes in monthly, seasonal and annual sediment loads.

Study area
The BRB covers a catchment area of about 543,400 km 2 (Joint Rivers Commission Bangladesh-JRCB ) of which 50% lies in China, 36% in India, 7% in Bangladesh and 7% in Bhutan. Figure 1 shows the watersheds and major river networks of the BRB. The BRB is categorized into three different physiographic zones: Tibetan Plateau, Himalayan Belt and the floodplain (Immerzeel ). The major river of this catchment, the Brahmaputra-Jamuna, is originated in the Himalayas and passes through China, India

Geospatial data
Geospatial data, such as elevation, land cover and soil types, are collected from different global geospatial repositories.
To delineate the catchment area and river network for (BUET) has provided these RCP8.5 scenarios data.

Model setup over the study domain
The watershed and stream network of BRB has been delineated using Arc Hydro tools (a version that works with Arc GIS 10.1). The output files from terrain pre-processing have been used to create input files for the HEC-HMS models using the HEC-Geospatial Hydrologic Modeling Extension (HEC-GeoHMS). Thiessen polygon method is used to achieve accurate estimation of the spatial distri-

Future climate projections
An 11-member ensemble from three Regional Climate

Calibration and validation of HEC-HMS model
In a sensitivity analysis of SMA parameters, the percentage changes in simulated volume have been plotted against the percentage variation of each SMA parameter and it is shown in Figure 3. In this study, soil percolation has been found as the most sensitive parameter for simulated streamflow during the calibration period. On the other hand, Ground Water 2 (GW2) storage has been found as the least sensitive parameter. Each parameter has been ranked according to their sensitivity for the change in simulated volume, as shown in Table 4. Therefore, among the SMA parameters, the value of soil percolation has been calibrated first and Ground Water 2 (GW2) storage has been adjusted last during the calibration process through manual calibration.
The model has been simulated for a 30-year baseline period from 1980 to 2009 to assess changes in flow and sediment load in the future. Therefore, data up until 2010 have      and 54% for the dry season, respectively.

Change of mean annual sediment load
Differences in mean annual sediment load are described in the box plots as shown in Figure 11. The percentage changes