Study of Streamflow Response to Land use Land cover Change in the Nethravathi River Basin, India

Abstract

The rainfall over Indian sub-continent is erratic and non-uniform in both South-west monsoon and North-east monsoon, which leads to high floods and severe drought in many places. However, India gets three-fourth of its annual rainfall during south-west monsoon season (June to September). The study of extreme events are significant in stochastic behaviour of rainfall pattern. In this study, frequency distribution method, GEV (Generalized Extreme Value) distribution, Mann-Kendall and Sen’s slope estimator are used for rainfall trend analysis over the Nethravathi basin, located in Western Ghats of Karnataka state, India. The rainfall data during the monsoon months (June to September) were analysed for a period of 1971 to 2010. The results from these methods have revealed that there is an increasing trend of frequency in class-1 and decreasing trend in class-2 and class-3 respectively. The interpretation of results was carried by the GEV distribution and nonparametric trend analysis (Mann-Kendall and Sen’s slope estimator test). The statistical techniques- Block Maxima (GEV) distribution, Mann-Kendall and Sen’s slope estimator test have demonstrated better results compared to frequency based method. Rainfall and land use land cover are considered to be the driving parameters of streamflow characteristics and cause considerable impacts on hydrologic regime of the watershed level. Therefore, an impact of LULC change on streamflow of Nethravathi river has been studied. Soil and Water Assessment Tool (SWAT) was used to construct the hydrologic model to study land use land cover change on streamflow in the Nethravathi basin. The sensitivity analysis was carried out based on Latin hypercube one factor-At-a-Time (LH-OAT) method using SWAT. The parameters Alpha-Bf, Canmax, Ch_K2, Ch_N2, Cn2, ESCO, Gwqmn, Revapmn, Gw_Dalay, Sol_K and Surlag are found to be most sensitive parameters for the Nethravathi river basin. Since SWAT require more number of input data to run, few to meteorological data are being monitored at the basin level or sub-basin level, it is not possible to use SWAT. Hence, an attempt has been made to propose a newly developed flow routing model called runoff coefficient routing model (RCRM), which is simple and require limited data such as precipitation, LULC and streamflow as compared to other models which require too many input data. The results of the newly developed RCRM model show better agreement with SWAT model in both calibration and validation period with R2 and NSE greater than 0.70. Therefore, it is concluded that the RCRM model is capable of predicting the streamflow at par with SWAT model. Hence, newly developed RCRM can be used to simulate and predict the streamflow in the data scarce region or basin. This study investigated the impacts of LULC changes on ET, streamflow and groundwater in the Nethravathi river basin using calibrated SWAT model. The impact results revealed that decrease in forest cover and increase agriculture& urban land, led to an increase in streamflow. It has also led to decreased ET and increased groundwater storage. This study provides useful information about impact of LULC change on streamflow, which may further helpful for flood mitigation and efficient water resources planning and management in the region. Further, the temporal variation in extreme precipitation events have been analysed for two decades (1991 - 2010). The analysis has shown extreme rainfall events have been reduced in the decade-2 (2001 - 2010) compared to decade-1 (1991 - 2000). Further, this study also analysed the impact of extreme precipitation on streamflow using SWAT model. Three hypothetical LULC scenarios have been developed based on the observed LULC change between 2003 and 2013 by satellite images along with field information. The three scenarios are Conversion to Agriculture (CA), Conversion to Built-up/Urban (CB) and Conversion to Wasteland (CW).The scenario-CB is found to be more sensitive as revealed by the result compared to scenarios-CA and scenarioCW. An attempt has been made to study the impact of vented dams and runoff-river type hydropower dams (without water storage) on streamflow. The model has shown the negative impact for vented dams as some portion of the streamflow is being used for agriculture or diverted or stored. Nevertheless no change was observed in the streamflow for runoff-river type hydropower dams as there is no storage or diversion of water being made. The present study results would benefit water managers, decision makes and developmental activities of the Nethravathi basin to implement protective measures for sustainable water resources in basin.