Impact of human activities on the Lower Karnali river flow through Bardia National Park, Nepal. An assessment of the changes in spatial and temporal patterns of the Lower Karnali river flow and its possible effect on sub-tropical tall grasslands with a 2D hydrodynamic model in HEC-RAS.

Abstract

Sub-tropical tall grasslands of Nepal are a very important habitat for many different species including the endangered tiger. Fragments of these tall grasslands are found along the east branch of the Lower Karnali river, which flows through Bardia National Park (BNP), located in the southwest of Nepal. The existence of these tall grasslands is presumed to be linked to the fluvial processes of the river and also to the groundwater table in the floodplains. This makes them vulnerable to large changes in the Lower Karnali river system. A large change occurred around 2009, when a shift from east to west in the main channel occurred. This probably had a natural origin as the lower Karnali system is a fluvial mega fan. Human activities also affect the system. In the lower Karnali gravel is extracted to use as a construction material. Furthermore, irrigation water is being extracted and this will increase in the future due to the construction of two new irrigation projects, the modernization of the Rani Jamara Kulariya Irrigation Scheme (MoRJKIS) and the Bheri-Babai diversion multipurpose project (BBDMP). In this study, it was evaluated whether these human activities may change the spatial and temporal patterns of the water flow at Lower Karnali river in such a way that it may impact the tall grasslands found in BNP. Therefore, a 2D hydrodynamic model was developed of the Lower Karnali River with the HEC-RAS 5.05 software to simulate the water flow over the two branches. The model was calibrated with available downstream discharge data of 2016 and satellite imagery. It was validated with downstream discharge data collected on a field trip in 2018. With a scenario analysis, the effect of human activities on three river flow variables (flood extent, water level, and discharge) in BNP was examined. The simulated human activities were: 1) gravel extraction of 28.1ˑ104 m3 at the upstream part of the west branch and 2) irrigation upstream water extraction of 100 m3/s which is 80% of the maximum capacity of the two irrigation projects. The uncertainty of the model was considerable. Therefore, the absolute results were unreliable and only relative changes in results under different conditions were evaluated. The flood extent, water levels and discharge decreased with lower inflow discharges. In the current state the average dropping rates of water levels after the monsoon were found to be 3 cm/day. The average discharge distribution of water over the west and east branch was found to be respectively 87% and 13% at low inflow discharges and at high discharges to be respectively 49% and 51%. According to the scenario analysis, the human activities will not considerably change the three flow parameters during high discharges. During low discharges the three flow variables were only considerably changed by the irrigation, which decreased each of these variables. This study found that the spatial and temporal patterns are only changed by irrigation activities. Since, the irrigation water extraction did not considerable change the spatial patterns in BNP during floods this does not affect the mechanism between high flows and the vegetation, including the disturbance by fluvial processes. The irrigation extraction does considerably change the spatial patterns during low discharges. In particular the water levels were lowered, which may also effect the ground water table. However, due the irrigation the dropping rate in the post monsoon would only increase with an average of 0.2 cm/day. This is neglible for the tall grasslands.