Regional Climate Modelling over Europe at Glacial Times - PhDData

Access database of worldwide thesis




Regional Climate Modelling over Europe at Glacial Times

The thesis was published by Velásquez Álvarez, Patricio Andrés, in September 2022, University of Bern.

Abstract:

The climate on Earth has continuously fluctuated throughout the world’s history under the influence of internal and external forcing factors. A key challenge for climate science is the understanding of the different drivers and mechanisms that define the climate of the past and its fluctuations. Glacial climate states are of great interest in climate research as their conditions are highly different compared to today’s climate. Climate modelling functions as a complementary tool to further investigate the role of forcing factors such as surface conditions in glacial climates. Global climate models are used to describe the Earth’s system; however, they show wide disagreement when simulating the climate of the past over the continents. This disagreement may be related to a variety of factors, including the coarse model resolution and an incomplete representation of Earth system processes. The application of regional climate models improves the representation of these processes due to their higher spatial resolution. Still, the accuracy of the simulated regional climate strongly depends on the representation of the surface conditions in the models. Even though the surface conditions become more realistic, deviations can still be evident in the simulations, especially in precipitation. These biases may impact the results obtained through hydrological and glacier modelling that follows next in the modelling chain. Accordingly, the central goal of this thesis is to investigate the role of the glacial surface conditions in the European glacial climate using the regional climate model WRF. Two studies are carried out to achieve this central goal. An additional study presents a method to adjust deviations in simulated precipitation at glacial times, e.g., the simulated precipitation of the previous two studies.

The first study assesses the importance of resolution and land{atmosphere feedbacks on the climate of Europe. To that end, a more accurate glacial land cover is generated using an asynchronous coupled land{atmosphere modelling experiment that combines a global climate model, a regional climate model, and a dynamic vegetation model. The regional climate and land cover models are run at high (18 km) resolution. The asynchronous coupling shows that the land{atmosphere coupling achieves quasi-equilibrium after four iterations. Simulated climate and land cover agree reasonably well with independent reconstructions based on paleoenvironmental proxies. This study determines the importance of land cover on the climate of Europe at the Last Glacial Maximum (LGM) using a sensitivity simulation with an LGM climate but present-day land cover. Results show that the LGM land cover leads to colder and drier summer conditions around the Alps and warmer and drier climate in southeastern Europe. This finding does not only demonstrate that LGM land cover plays an important role in regulating the regional climate, but it also indicates the need of using realistic glacial land cover estimates to accurately simulate the regional glacial climate.

The second study investigates the sensitivity of the glacial Alpine hydro-climate to northern hemispheric and local ice-sheet changes. Therefore, sensitivity simulations are carried out with a 2 km horizontal resolution over the Alps for the LGM and the Marine Isotope Stage 4 (MIS4). During winter, the findings show wetter conditions in the southern part of the Alps under LGM conditions compared to present day. This wetting can be traced back to dynamical processes, i.e., changes in the wind speed and direction. In summer, drier conditions are found in most of the Alpine region during LGM. These drier conditions can be attributed to thermodynamic processes, i.e., lower temperatures. The MIS4 climate shows enhanced winter precipitation compared to the LGM, which is explained by its warmer climate compared to the LGM, i.e., by thermodynamics. An increase of the northern hemispheric ice-sheet thickness leads to a significant intensification of glacial Alpine hydro-climate conditions, which is mainly explained by dynamical processes. Changing only the Fennoscandian ice sheet is less influential on the Alpine precipitation, whereas modifications in the local Alpine ice-sheet topography significantly alter the Alpine precipitation. These findings demonstrate that the northern hemispheric and local ice-sheet topography are of great importance at regulating the Alpine hydro-climate.

The third study presents a new correction method for precipitation over complex terrain that explicitly considers orographic characteristics. This method offers a good alternative to the standard empirical quantile mapping (EQM) method during colder climate states, in which the orography strongly deviates from the present-day state, e.g., at the LGM. The new method and its performance are presented for Switzerland using regional climate model simulations at 2 km resolution for present day and LGM conditions. In present-day conditions, the comparison between simulations and observations shows a strong seasonality and, especially during colder months, a height dependence of the bias in precipitation. The new method is able to fully correct the seasonal precipitation bias induced by the global climate model. A rigorous temporal and spatial cross-validation with independent data exhibits robust results. The application of the new bias-correction method to the LGM demonstrates that it is a more appropriate correction compared to the standard EQM under highly different climate conditions as the latter imprints present-day orographic features into the LGM climate.

The last chapter of this thesis is dedicated to highlight some key results of the studies of this thesis and to outline possible follow-up studies and potential benefits for other studies and the scientific community.



Read the last PhD tips