A stochastic weather generator (SWG) is a statistical tool used to create synthetic sequences of daily weather data (e.g., precipitation, temperature, solar radiation) that mimic the statistical properties of observed historical weather patterns. Using a probabilistic model the tool generates random sequences of meteorological data that replicate the local weather pattern as observed at each meteorological station. With the simulated daily data as a basis, realistic temporal sequences such as dry and wet periods, hot or cold spells, can be produced while preserving the statistical properties like mean, variance, and autocorrelation of the historical data.

Usually precipitation sequences are generated first, and other data sequences are derived using statistical relationships between these data and precipitation, with different relationships used for wet and dry days.  The estimation of precipitation involves first modelling the occurrence of wet and dry days using a Markov chain procedure, and then modelling the amount of precipitation falling on wet days,  sampled randomly from an appropriate distribution, such as a Gamma distribution. By using different random seeds, a large number of sequences can be generated, all of which have the same statistical properties as the original data used to calibrate the statistical model.

SWGs are of particular use in areas where meteorological records are short or contain missing data. In addition, by simulating possible future weather conditions based on the modified climate variables (e.g., temperature increase, altered rainfall patterns) projected by GCMs, SWGs can be used to explore the impacts of climate change on agriculture, water resources, and ecosystems. The latter application is built into the Statistical Downscaling Model (SDSM) which contains a fairly user-friendly weather generator.

The ICARDA Agro-Climate Tool uses GEM6, a Richardson-type SWG. A good summary description of general design, analysis steps and uses of stochastic weather generators, as well as some references,  can be found on this IPCC website.