Projected changes in surface solar radiation in CMIP5 global climate models and in EURO-CORDEX regional climate models for Europe

• Published in: Climate dynamics 2017-10, Vol.49 (7-8), p.2665-2683
• Main Authors: Bartók, Blanka, Wild, Martin, Folini, Doris, Lüthi, Daniel, Kotlarski, Sven, Schär, Christoph, Vautard, Robert, Jerez, Sonia, Imecs, Zoltán
• Format: Article
• Language: English
• Subjects: Analysis; Atmospheric absorption; Atmospheric models; Climate; Climate models; Climatology; Cloud cover; Cloudiness; Computer simulation; Continental interfaces, environment; Earth and Environmental Science; Earth Sciences; Geophysics/Geodesy; Global climate; Global climate models; Greenhouse effect; Numerical weather prediction; Ocean, Atmosphere; Oceanography; Radiation (Physics); Regional climate models; Regional climates; Sciences of the Universe; Solar radiation; Water vapor; Water vapor content; Water vapour
• ISSN: 0930-7575; 1432-0894
• DOI: 10.1007/s00382-016-3471-2

The objective of the present work is to compare the projections of surface solar radiation (SSR) simulated by four regional climate models (CCLM, RCA4, WRF, ALADIN) with the respective fields of their ten driving CMIP5 global climate models. First the annual and seasonal SSR changes are examined in the regional and in the global climate models based on the RCP8.5 emission scenarios. The results show significant discrepancies between the projected SSR, the multi-model mean of RCMs indicates a decrease in SSR of −0.60 W/m 2 per decade over Europe, while the multi-model mean of the associated GCMs used to drive the RCMs gives an increase in SSR of +0.39 W/m 2 per decade for the period of 2006–2100 over Europe. At seasonal scale the largest differences appear in spring and summer. The different signs of SSR projected changes can be interpreted as the consequence of the different behavior of cloud cover in global and regional climate models. Cloudiness shows a significant decline in GCMs with −0.24% per decade which explains the extra income in SSR, while in case of the regional models no significant changes in cloudiness can be detected. The reduction of SSR in RCMs can be attributed to increasing atmospheric absorption in line with the increase of water vapor content. Both global and regional models overestimate SSR in absolute terms as compared to surface observations, in line with an underestimation of cloud cover. Regional models further have difficulties to adequately reproduce the observed trends in SSR over the past decades.