Novel application of a process convolution approach for calibrating output from numerical models

Output from numerical models at high spatial and temporal resolutions is critical for modeling applications in a variety of disciplines. Prior to its use modeling, output climate must be brought finer resolution calibrated with respect observations. The calibration model output, referred as bias-correction, poses many statistical challenges. Here, we develop bias-correction method which systematic biases the mean standard deviation are corrected. In addition, employ novel process convolution approach correct bias dependence. We apply this temperature simulations generated by regional over Northeastern USA. goal study was historical (1976–2005) future (2006–2099) time periods while simultaneously preserving trends resulting carbon emissions scenarios. compare proposed quantile mapping (empirical mapping, EQM). resulted more effective correction seasonal dependence compared EQM.