Basic Techniques in Environmental Simulation
Environmental simulation modeling is defined as the generation of synthetic weather observations and forecasts by use of mathematical/ statistical models. Basic concepts in environmental simulation modeling are described, with emphasis on underlying statistical fundamentals, stochastic processes, and Markov processes. Four principle environmental simulation models and their application are described in detail. The treatment begins with the single- variable, single-station model, V1S1, and is extended to the two- variable, single-station model V2S1. The multivariate triangular matrix model, MULTRI, is then discussed; that model is capable of generating vectors of N correlated variables. A case study is presented showing the application of MULTRI to modeling point sky cover distributions at station pairs or at a single station for N lag times. The most complex model in the series of four is the 2-dimensional field simulation model, 2DFLD, capable of producing spatially correlated, synthetic, two-dimensional fields or networks or variables. Statistical methods used in developing environmental simu- lation models are described, with particular emphasis placed on how to fit probability distribution functions to weather variables.