Development of statistical downscaling model based on Volterra series realization, principal components and ridge regression
Singh, P., Shamseldin, A.Y., Melville, B.W. et al. Development of statistical downscaling model based on Volterra series realization, principal components and ridge regression. Model. Earth Syst. Environ. 9, 3361–3380 (2023). https://doi.org/10.1007/s40808-022-01649-3
Abstract
By analysing the statistical relationships between local rainfall data and regional climate variables we can identify predictors and model future climate change impacts. The combined application of principal component analysis (PCA) and ridge regression improved the performance of a statistical model called SDCRR, which was reliable enough to capture appropriate information from predictors in the Manawatu River basin, in the North Island of New Zealand.
The results of this study show the SDCRR model has better performance than the widely used statistical downscaling model (SDSM).