Abstract
AbstractThe majority of regional climate change assessments for the Euro-CORDEX region is based on high resolution atmosphere models. These models use prescribed lower boundary conditions, such as sea surface temperatures (SST) from global ocean General Circulation Models (GCMs), that do not respond to changes simulated by the regional atmosphere model, thus lacking an important feedback to the atmosphere. However, research during the past decade indicated that the use of coupled atmosphere–ocean models can lead to significantly altered model solutions compared to standalone atmosphere models for the present day climate imposing some uncertainty on the widely used uncoupled future scenarios. We here present the first multi-model and multi scenario (RCP2.6, RCP4.5, RCP8.5) ensemble of future climate change scenarios downscaled with a coupled atmosphere—ocean model in which sea surface temperature and sea ice fields are explicitly simulated by a coupled state-of-the-art high resolution ocean model and communicated to the atmosphere at 3-hourly time steps. Our ensemble generally confirms results of previous uncoupled ensembles over land areas implying that the coupling effect is restricted mainly to the coupled area and the adjacent coastal zone. By contrast, over the North Sea and Baltic Sea small scale processes point to important coupling effects that mediate the response to climate change and that can not be simulated by uncoupled models. Our results therefore impose general uncertainty on the usage of regional climate change data from uncoupled ensembles over marine areas such as for purposes of offshore wind or mussel farming, the planing of marine protected areas, and marine recreation along the coastal zone. It further sets in question the usage of uncoupled scenario data (such as Euro-CORDEX) to force high resolution ocean models. Comparing coupled and uncoupled hindcast simulations reveals that the coupling effect over land is most pronounced during the warm season when prescribed and modelled sea surface temperatures (SST) differ strongest. In addition, a generally weaker wind regime in summer damps the heat dispersion in the atmosphere so that air temperature anomalies can extent further over land compared to winter. Future projections are discussed under consideration of land-sea warming characteristics for selected climate indices as well as mean seasonal climate change. At the end of the century a clear land-sea pattern is seen in all scenarios with stronger warming over land than over open sea areas. On average land areas warm at a rate 1.5 times faster than areas over the open ocean. Over the coupled area, i.e. the North Sea and Baltic Sea tropical nights are impacted strongest and the Baltic Sea turns out to be a hot spot in future climate. This has been unrecognized in previous studies using high resolution atmosphere models with prescribed SSTs from global models which do not represent small scale ocean processes in the Baltic Sea adequately.
Funder
mercator CMEMS
Leibniz-Institut für Ostseeforschung Warnemünde (IOW)
Publisher
Springer Science and Business Media LLC
Reference106 articles.
1. Adloff F, Jordà G, Somot S, Sevault F, Meyssignac B, Arsouze T, Li L, Planton S (2017) Improving sea level simulation in Mediterranean regional climate models. Clim Dyn, Med-CORDEX special issue 1–12
2. Akhtar N, Brauch J, Ahrens B (2017) Climate modeling over the Mediterranean Sea: impact of resolution and ocean coupling. Clim Dyn, Med-CORDEX special issue 1–16
3. Bartók B, Wild M, Folini D, Lüthi D, Kotlarski S, Schär C, Vautard R, Jerez S, Imecs Z (2017) Projected changes in surface solar radiation in CMIP5 global climate models and in EURO-CORDEX regional climate models for Europe. Clim Dyn 49(7–8):2665–2683
4. Bersch M, Gouretski V, Sadikni R, Hinrichs I (2013) KLIWAS North Sea climatology of hydrographic data (version 1.0). World Data Center Clim (WDCC). https://doi.org/10.1594/WDCC/KNSC_hyd_v1.0
5. Boé J, Somot S, Corre L, Nabat P (2020) Large differences in Summer climate change over Europe as projected by global and regional climate models: causes and consequences. Clim Dyn. https://doi.org/10.1007/s00382-020-05153-1
Cited by
15 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献