Lithologic Controls on Parameters of Conceptual Rainfall-Runoff Model and Runoff Characteristics: A Case Study of the Xinanjiang Model

This study investigates lithologic control on parameters of the conceptual rainfall-runoff model and runoff characteristics through the construction of Xinanjiang rainfall-runoff models in different subwatersheds with different lithologies located in the Xiangxihe River watershed. The Xinanjiang models are calibrated based on a genetic algorithm and validated using the observed data. All the simulations for these four subwatersheds have NSE values larger than 0.8 during the calibration period and NSE values larger than 0.7 during the validation period. Root mean square error (RMSE) values range from 0.030 to 2.780 m(3)/s. Compared with the other three subwatersheds dominated by limestone and dolomite, Gaojiaping subwatershed mainly composed of granite is characterized by the significantly smaller areal mean free water storage capacity and larger outflow coefficient of free water storage to the interflow. Gaojiaping subwatershed has a dramatically higher-level surface runoff proportion and a remarkably lower-level groundwater runoff proportion than the other three subwatersheds-Sanchakou, Cangpinghe, and Xiangguxi subwatersheds. The concept of a geological characteristic value of watershed is proposed to reflect the water-bearing property of rock stratum for the whole watershed through an area-weighted average. This geological characteristic value in the karst subwatershed is larger than that of the nonkarst subwatershed. The relationships between the parameters of the Xinanjiang model and the geological characteristic values are obtained, and the variation of four categories of parameters with the geological characteristic values is analyzed. The surface runoff proportion and groundwater runoff proportion are negatively and positively correlated with the geological characteristic value, respectively. The established empirical equations can well reflect the relations between the geological characteristic values and runoff proportions. The importance of this finding emerges from the direct link between the index of hydrogeological survey in hydrogeology and proportions of different runoff components, providing a simple and effective way for water resources assessment in ungauged watersheds.