Coupled stochastic weather generation using spatial and generalized linear models

摘要: We introduce a stochastic weather generator for the variables of minimum temperature, maximum temperature and precipitation occurrence. Temperature are modeled in vector autoregressive framework, conditional on Precipitation occurrence arises via probit model, both spatially correlated using spatial Gaussian processes. Additionally, local climate is included by varying model coefficients, allowing evolving relationships between variables. The method illustrated network stations Pampas region Argentina where nonstationary historical correlation challenge existing approaches.