Precipitation is important factor affecting vegetation and controlling key ecological processes. In order to quantify spatial patterns of precipitation in Chongqing tobacco planting region, China, under ArcGIS platform, three multivariate geostatistical methods including cokriging, small grid and regression kriging, coupled with auxiliary topographic factors extracted from a 1:100000 DEM were applied to predict spatial distribution of precipitation for January (the least month), June (the richest month) and the whole year. The results showed that cokriging was the best for prediction precipitation of January, which could explain 58% of the total variation. Small grid simulation with IDW interpolation exhibited higher accuracy for both June precipitation and annual precipitation, which explained 72% and 61% of the total variation respectively. Generally, multivariate geostatistics accounted for most of the spatial variability in mean precipitation and especially could exhibit great improvement for estimating precipitation in areas where topography has a major influence on the precipitation.