Three land utilization data sets of 1990,2000 and 2010 in the Lianshui basin have been chosen,and meteorological data from 1985 through 2014 have been divided into 1985 - 1994,1995 - 2004 and 2005 - 2014 meteorological background phases,which is convenient to set up seven combination simulation scenarios applied in the paper. SWAT,a distributed runoff model,is used to simulate runoff in different scenarios. The purpose of this paper is to study how the climate,land-utilization and land-cover changes in the Lianshui basin affects the runoff in the basin. PSO algorithm is used to calibrate model parameters by Kling-Gupta efficiency coefficient(KGE),as well as by observed data in Xiangxiang station. Furthermore,p-factor,r-factor,correlation coefficient(R~2),Nash-Sutcliffe efficiency coefficient(NS) and Percent bias(PBIAS) are selected to evaluate the uncertainty and accuracy of the simulation. The evaluation results indicated that the simulated accuracy and uncertainty,under different land utilization scenarios,were credible both in the calibration and validation phases. Consequently,by analytic comparison of the simulated values in different scenarios,it is clear that during 1985 to 2014,climate change had decreased basin runoff while land utilization change had increased basin runoff,but annual total runoff depth had decreased on the whole. Climate change had influenced the surface hydrology more significantly than land utilization change,thereby the contribution rate of climate change increased from 71.4% to 86.3%. At the same time,the contribution rate of land utilization change decreased from 28.6% to 13.7%. Therefore,under the background of local climate change,optimizing the spatial structure of land resources is one of the ways to scientific management of water resources in the basin.