Knist, Sebastian: Land-atmosphere interactions in multiscale regional climate change simulations over Europe. - Bonn, 2018. - Dissertation, Rheinische Friedrich-Wilhelms-Universität Bonn.
Online-Ausgabe in bonndoc: https://nbn-resolving.org/urn:nbn:de:hbz:5n-52545
@phdthesis{handle:20.500.11811/7666,
urn: https://nbn-resolving.org/urn:nbn:de:hbz:5n-52545,
author = {{Sebastian Knist}},
title = {Land-atmosphere interactions in multiscale regional climate change simulations over Europe},
school = {Rheinische Friedrich-Wilhelms-Universität Bonn},
year = 2018,
month = dec,

volume = 86,
note = {Interactions between the heterogenous land surface and the atmosphere play a fundamental role in the weather and climate system through their influence on the energy and water cycles. Global climate models (GCMs) currently have coarse horizontal grid resolutions in the order of 100 km. With their higher resolution regional climate models (RCMs) better resolve mesoscale processes in the atmosphere and better represent the heterogenous land surface properties. Thus, RCMs are able to provide more detailed characteristics of regional to local climate. This thesis conducts regional climate simulations in multiple resolutions for the European domain of the Coordinated Regional Climate Downscaling Experiment (EURO-CORDEX) and a central European domain (3kmME) with the RCM WRF downscaling both ERA-Interim reanalysis and GCM MPI-ESM-LR (RCP4.5) climate change scenario data. The analysis focusses on land-atmosphere interactions to gain a better understanding of the regional water cycle components, the involved multi-scale processes, their sensitivities and variabilities both under present-day climate and future climate change conditions. Furthermore, the added value of the convection-permitting 3kmME simulations, being one of the first sets of decade-long convection-permitting regional climate simulations over Central Europe, is investigated.
A comparison of summertime land-atmosphere coupling strength is carried out for a subset of the ERA-Interim-driven EURO-CORDEX model ensemble (1989 to 2008). The coupling strength is quantified by the correlation between the surface sensible and the latent heat flux, and by the correlation between the latent heat flux and 2m temperature and compared to European FLUXNET observations and gridded observational Global Land Evaporation Amsterdam Model (GLEAM) data, respectively. The RCM simulations agree with both observational datasets in the large-scale pattern characterized by strong coupling in southern Europe and weak coupling in northern Europe. However, in the transition zone from strong to weak coupling covering large parts of central Europe the majority of the RCMs tend to overestimate the coupling strength in comparison to both observations. The RCM ensemble spread is caused by the different land surface models applied, and by the model-specific weather conditions resulting from different atmospheric parameterizations. Investigation of land-atmosphere coupling strength in ERA-Interim driven WRF simulations in both 3 km and 12 km resolution for central Europe reveals large year-to-year variability related to the individual soil moisture conditions. Coupling strength largely differs for individual land use types. Forest compared to crop type reacts slower to drought conditions. Coupling is overall slightly stronger in the 3 km simulation, attributed to overall drier soils due to less precipitation. The projected climate change based on a WRF 0.44° simulation downscaling GCM MPI-ESM-LR (RCP4.5) data alters the European land-atmosphere coupling regimes in summer. Due to increasingly drier soils, stronger coupling is simulated for large parts of western, central and southern eastern Europe for the period 2071-2100 compared to 1971-2000. Areas of strongest future increase of extreme temperature coincide with strong coupling areas.
In order to analyse the added value of convection-permitting 3 km climate simulations, nine years of ERA-Interim driven simulations with the WRF RCM at 12 km and 3 km grid resolution over central Europe are evaluated against observations with a focus on sub-daily precipitation statistics and the relation between extreme precipitation and air temperature. A clear added value of the higher resolution simulation is found especially in the reproduction of the diurnal cycle and the hourly intensity distribution of precipitation. Too much light precipitation in the 12 km simulation results in a positive precipitation bias. Largest differences between both resolutions occur in mountainous regions and during the summer months with high convective activity. Moreover, the observed increase of the temperature–extreme precipitation scaling from the Clausius-Clapeyron (C-C) scaling rate of ~7% K-1 to a super-adiabatic scaling rate is reproduced only by the 3 km simulation.
The effect of land surface heterogeneity on the differences between 3 km and 12 km simulations is analysed based on five WRF simulations for JJA 2003, each with the same atmospheric setup in 3 km resolution but different combinations of 12 km resolution land use and soil type, initial soil moisture and orography. A coarser resolved orography significantly alters the flow over and around extensive mountain ridges like the Alps and impact the large-scale flow pattern. The smoothed mountain ridges result in weaker Föhn effects and in enhanced locally generated convective precipitation pattern peaking earlier in the afternoon. The effect of a coarser-resolved land use distribution is overall smaller and mainly related to changes in overall percentages of different land use types, rather than to the loss of heterogeneity in the surface pattern on the scale analysed here. Even small changes in soil moisture have a higher potential to affect the overall simulation results.
WRF climate simulations downscaling the MPI-ESM-LR data at 12 km and 3 km resolution for central Europe are analysed for three 12-year periods: a control, a mid-of-century and an end-of-century projection to quantify future changes in precipitation statistics based on both convection-permitting and convection-parameterized simulations. For both future scenarios both simulations suggest a slight decrease in mean summer precipitation and an increase in hourly heavy and extreme precipitation in large parts of central Europe. This increase is stronger in the 3 km runs. Temperature–extreme precipitation scaling curves in the future climate are projected to shift along the 7% K-1 trajectory to higher peak extreme precipitation values at higher temperatures while keeping their typical shape.
The results of this thesis clearly confirm the added value of convection-permitting climate simulations, provide further insights into land-atmosphere interaction processes and highlight the relevance of the RCMs ability to properly simulate coupling strength.},

url = {https://hdl.handle.net/20.500.11811/7666}
}

The following license files are associated with this item:

InCopyright