Comment on egusphere-2023-1021

Abstract

<strong class="journal-contentHeaderColor">Abstract.</strong> Physical numerical weather prediction models have biases and miscalibrations that can depend on the weather situation, which makes it difficult to postprocess them effectively using the traditional model output statistics (MOS) framework based on parametric regression models. Consequently, much recent work has focused on using flexible machine learning methods that are able to take additional weather-related predictors into account during postprocessing, beyond the forecast of the variable of interest only. Some of these methods have achieved impressive results, but they typically require significantly more training data than traditional MOS and are less straightforward to implement and interpret. We propose MOS random forests, a new postprocessing method that avoids these problems by fusing traditional MOS with a powerful ML method called random forests to estimate "weather-adapted" MOS coefficients from a set of predictors. Since the assumed parametric base model contains valuable prior knowledge, much smaller training data sizes are required to obtain skillful forecasts and model results are easy to interpret. MOS forests are straightforward to implement and typically work well, even with no or very little hyperparameter tuning. For the difficult task of postprocessing daily precipitation sums in complex terrain, MOS forests outperform reference machine learning methods at most of the stations considered. Additionally, they are highly robust to changes in the data size and work well even when less than a hundred observations are available for training.