uu.seUppsala University Publications
Change search
ReferencesLink to record
Permanent link

Direct link
Propagation of biases in climate models from the synoptic to the regional scale: Implications for bias adjustment
Univ Zurich, Dept Geog, Zurich, Switzerland.;Natl Ctr Atmospher Res, Appl Res Lab, POB 3000, Boulder, CO 80307 USA..
Univ Bern, Oeschger Ctr Climate Change Res, Bern, Switzerland.;Univ Bern, Inst Geog, Bern, Switzerland..
Univ Zurich, Dept Math, Zurich, Switzerland..
Uppsala University, Disciplinary Domain of Science and Technology, Earth Sciences, Department of Earth Sciences, LUVAL. Univ Zurich, Dept Geog, Zurich, Switzerland..
2016 (English)In: Journal of Geophysical Research - Atmospheres, ISSN 2169-897X, E-ISSN 2169-8996, Vol. 121, no 5, 2075-2089 p.Article in journal (Refereed) PublishedText
Abstract [en]

Bias adjustment methods usually do not account for the origins of biases in climate models and instead perform empirical adjustments. Biases in the synoptic circulation are for instance often overlooked when postprocessing regional climate model (RCM) simulations driven by general circulation models (GCMs). Yet considering atmospheric circulation helps to establish links between the synoptic and the regional scale, and thereby provides insights into the physical processes leading to RCM biases. Here we investigate how synoptic circulation biases impact regional climate simulations and influence our ability to mitigate biases in precipitation and temperature using quantile mapping. We considered 20 GCM-RCM combinations from the ENSEMBLES project and characterized the dominant atmospheric flow over the Alpine domain using circulation types. We report in particular a systematic overestimation of the frequency of westerly flow in winter. We show that it contributes to the generalized overestimation of winter precipitation over Switzerland, and this wet regional bias can be reduced by improving the simulation of synoptic circulation. We also demonstrate that statistical bias adjustment relying on quantile mapping is sensitive to circulation biases, which leads to residual errors in the postprocessed time series. Overall, decomposing GCM-RCM time series using circulation types reveals connections missed by analyses relying on monthly or seasonal values. Our results underscore the necessity to better diagnose process misrepresentation in climate models to progress with bias adjustment and impact modeling.

Place, publisher, year, edition, pages
2016. Vol. 121, no 5, 2075-2089 p.
Keyword [en]
bias adjustment, regional climate simulation, impact modeling, synoptic circulation, downscaling, process misrepresentation
National Category
Meteorology and Atmospheric Sciences
Identifiers
URN: urn:nbn:se:uu:diva-295574DOI: 10.1002/2015JD024040ISI: 000373598700003OAI: oai:DiVA.org:uu-295574DiVA: diva2:941194
Available from: 2016-06-22 Created: 2016-06-08 Last updated: 2016-06-22Bibliographically approved

Open Access in DiVA

No full text

Other links

Publisher's full text

Search in DiVA

By author/editor
Seibert, Jan
By organisation
LUVAL
In the same journal
Journal of Geophysical Research - Atmospheres
Meteorology and Atmospheric Sciences

Search outside of DiVA

GoogleGoogle Scholar
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Altmetric score

Total: 103 hits
ReferencesLink to record
Permanent link

Direct link