This study examines the effects of urban development pressure on agricultural land prices in Korea. It investigates the determinants of agricultural land prices, focusing particularly on urban planning variables on agricultural land use. Since agricultural land price is in general closely related to the spatial characteristics of an area, it adopts a mixed GWR (geographically weighted regression) model in order to identify local and global effects of independent variables on agricultural property values. Results of a mixed GWR model prove to be superior to those of a global model (OLS) and standard spatial econometrics models (SAR, SEM, SAC) in terms of model fits and stability of the parameters estimated. The model notes that spatial dependency and heterogeneity are particularly important in examining the variations of agricultural land price. The empirical results provide strong evidences that such factors as man-made and natural features are closely relevant in determining agricultural property values.