Rainfall modeling using latent Gaussian random fieldsan estimation method

International audienceWe propose a method for estimating the parameters in a latent Gaussian field used for modeling daily rainfall. For the rainfall variable, a monotonie transformation is applied to achieve marginal normality, thus, defining a latent variable, with zéro rainfall values corresponding to censored values below a threshold. Methodology is presented for model estimation and validation illustrated using accumulated daily rainfall data from a network of 14 stations in the Southern Sweden. Performance of the model is judged through its ability to accurately reproduce a sériés of temporal and spatial dependence measures