This paper deals with the impact of temporal variation of Land use/Land cover (LULC) and precipitation on runoff estimation for an un-gauged hilly watershed in Kodagu District of Karnataka, India. Sloped Terrain of the watershed and increasing trend of rainfall over the last decade, has been contributing to high runoff values which has ultimately damaged the rich flora and fauna of the region. The total area of the watershed is 459.31 km2. The land use/land cover change analysis has been performed for the years 2004, 2007, 2013, and 2018. Remote sensing and GIS techniques have been used to identify the watershed and for developing thematic maps. Very popular, soil conservation service - Curve number (SCS-CN) (USDA), method has been used in this study to estimate the seasonal runoff from the rainfall. The LULC maps have been generated by supervised maximum likelihood classification technique in order to get higher accuracy in the estimation of curve numbers. Class-wise runoff coefficients have been estimated to analyse the response of each class to rainfall. The runoff response for a watershed depends both on precipitation as well as land use/land cover classes. The impact of both the factors has been assessed in the estimation of runoff, and it has been observed that the impact of rainfall factor is more. The runoff values have been obtained for return periods of 10, 20, 30, 50, and 100 years by using the value of runoff coefficient averaged over 15 years. © 2020 American Society of Civil Engineers.