A Bayesian latent Gaussian conditional autoregressive copula model for analyzing spatially-varying trends in rainfall (2504.00766v1)

Abstract: Assessing the availability of rainfall water plays a crucial role in rainfed agriculture. Given the substantial proportion of agricultural practices in India being rainfed and considering the potential trends in rainfall amounts across years due to climate change, we build a statistical model for analyzing annual total rainfall data for 34 meteorological subdivisions of mainland India available for 1951-2014. Here, we model the margins using a gamma regression model and the dependence through a Gaussian conditional autoregressive (CAR) copula model. Due to the natural variation of the average annual rainfall received across various dry through wet regions of the country, we allow areally-varying gamma regression coefficients under a latent Gaussian model framework. The neighborhood structure of the regions determines the dependence structure of both the likelihood and the prior layers, where we explore both CAR and intrinsic CAR structures for the priors. The proposed methodology also imputes the missing data effectively. We use the Markov chain Monte Carlo algorithms to draw Bayesian inferences. In simulation studies, the proposed model outperforms some competitors that do not allow a dependence structure at the data or prior layers. Implementing the proposed method for the Indian areal rainfall dataset, we draw inferences about the model parameters and discuss the potential effect of climate change on rainfall across India.