Analyzing the effect of climate change on rainfall is crucial from the perspective of agriculture, water resources management, and disaster management. Motivated by the key features present in a gridded rainfall dataset published by the India Meteorological Department (IMD) for the years 1951–2022 at a spatial resolution of 0.25◦×0.25◦ across India, we propose a high-dimensional Bayesian skew-Gaussian conditional autoregressive model to analyze spatially-varying temporal patterns in Indian monsoon rainfall.We model the mean component in a semiparametric regression framework with spatially-varying coefficients and assume the skewness component to be also spatially-varying. We briefly discuss some theoretical properties of the proposed model. We draw inferences using Gibbs sampling by exploiting the hierarchical definition of a skew-Gaussian distribution as a random location-mixture of a Gaussian distribution. We study the posterior mean and the corresponding T-statistics of the overall change in rainfall amounts during the observational period at each grid cell. Our results indicate a significant negative trend in various parts of the Gangetic River basin, where agriculture is predominantly dependent on the availability of rainwater.