This repoitory contains code for the Censored Spatial Gaussian Copula from my paper: Probabilistic Rainfall Downscaling: Joint Generalized Neural Models with Censored Spatial Gaussian Copula.
The first part of that paper is the JGNM with code here. It is a density estimation model for future rainfall based on temporal and spatial climate model predictors. It outputs three parameters for each location, with outputs looking like this:
Thes three parameters are used to get zero-gamma mixture densities for each location. The densities in question:
The second part is the spatial dependence with the Censored copula.
Here are a few examples of samples you can generate with our method.