Refine
Has Fulltext
- yes (2)
Is part of the Bibliography
- yes (2) (remove)
Document Type
- Journal article (1)
- Doctoral Thesis (1)
Language
- English (2) (remove)
Keywords
- Modell (2) (remove)
Institute
- Institut für Geographie und Geologie (2) (remove)
The Mediterranean area reveals a strong vulnerability to future climate change due to a high exposure to projected impacts and a low capacity for adaptation highlighting the need for robust regional or local climate change projections, especially for extreme events strongly affecting the Mediterranean environment. The prevailing study investigates two major topics of the Mediterranean climate variability: the analysis of dynamical downscaling of present-day and future temperature and precipitation means and extremes from global to regional scale and the comprehensive investigation of temperature and rainfall extremes including the estimation of uncertainties and the comparison of different statistical methods for precipitation extremes. For these investigations, several observational datasets of CRU, E-OBS and original stations are used as well as ensemble simulations of the regional climate model REMO driven by the coupled global general circulation model ECHAM5/MPI-OM and applying future greenhouse gas (GHG) emission and land degradation scenarios.
Climate models are the tool of choice for scientists researching climate change. Like all models they suffer from errors, particularly systematic and location-specific representation errors. One way to reduce these errors is model output statistics (MOS) where the model output is fitted to observational data with machine learning. In this work, we assess the use of convolutional Deep Learning climate MOS approaches and present the ConvMOS architecture which is specifically designed based on the observation that there are systematic and location-specific errors in the precipitation estimates of climate models. We apply ConvMOS models to the simulated precipitation of the regional climate model REMO, showing that a combination of per-location model parameters for reducing location-specific errors and global model parameters for reducing systematic errors is indeed beneficial for MOS performance. We find that ConvMOS models can reduce errors considerably and perform significantly better than three commonly used MOS approaches and plain ResNet and U-Net models in most cases. Our results show that non-linear MOS models underestimate the number of extreme precipitation events, which we alleviate by training models specialized towards extreme precipitation events with the imbalanced regression method DenseLoss. While we consider climate MOS, we argue that aspects of ConvMOS may also be beneficial in other domains with geospatial data, such as air pollution modeling or weather forecasts.