Swiss Federal Institute for Forest, Snow and Landscape Research WSL

Avalanche bulletin and snow situation
Edition: Sat 21.12., 17:00
Avalanche Bulletin

CHELSA

CHELSA (Climatologies at high resolution for the earth’s land surface areas) is a high-resolution climate data set for the global land surface area. It is hosted by the Swiss Federal Institute for Forest, Snow and Landscape Research WSL and enjoys great demand among ecologists interested in the distribution of species. The temperature and precipitation data help to model living conditions of species.

Die graphische Abbildung zeigt die mit CHELSA errechneten Niederschlagsmuster in der komplizierten Topographie von Bhutan (Landesgrenzen in Schwarz). Der meiste Niederschlag fällt während des Südwestmonsuns im Sommer, wenn feuchte Luftmassen am Himalaya-Gebirge aufsteigen. Während andere Modelle die grossräumigen Muster auch gut abbilden, zeigt das CHELSA-Modell auf kleinerer Skala seine Stärke. Es sagt die real beobachtete Wolkenverteilung, nämlich wenig Wolken in den abgeschotteten Bergtälern, viele an den windexponierten Südflanken, korrekt vorher. (Grafik: D. Karger et al., 2017)
Figure 1: Precipitation patterns calculated with CHELSA in the complicated topography of Bhutan (national borders in black). Most precipitation falls during the southwest monsoon in summer, when humid air masses rise on the Himalayas. While other models show the large scale patterns equally well, the CHELSA model demonstrates its strength on a smaller scale. It correctly predicts the observed precipitation distribution, namely fewer precipitation in the isolated mountain valleys and many on the wind-exposed southern flanks. (graphic: D. N. Karger et al., 2017)
Die vier Karten zeigen Verbreitungsmodelle der trockenliebenden Pflanze Französischer Tragant (Astragalus monspessulanus), errechnet mit Klimadaten von WorldClim (oben) und CHELSA (unten). Die grünen Punkte zeigen die tatsächlichen Vorkommen. Der Kreis zeigt das trockene Rhonetal, wo WorldClim die Niederschläge überschätzt und damit die Verbreitung der Pflanze nicht korrekt vorhersagt. (Grafik: D. Karger et al., 2017)
Figure 2: Distribution models of the drought-loving plant Astragalus monspessulanus, calculated with climate data from WorldClim (a,b) and CHELSA (c,d). The green dots show the actual occurrences of the plant. The circle shows the dry Rhine valley, where WorldClim overestimates precipitation and thus does not correctly predict the distribution of the plant. (graphic: D. N. Karger et al., 2017)

In order to be able to study living organisms, ecologists need the most precise data possible on the climatic conditions of the respective ecosystems. From the temporal and spatial distribution of temperature and precipitation, they can calculate local habitats (niches), vegetation periods, or the distribution of animal and plant species. Small errors in the climate models can quickly lead to large deviations in such models, which makes it difficult to investigate ecological changes.

An improvement of climate information is therefore desirable. Especially for regions with a low density of meteorological stations as mountains or remote rainforests. With CHELSA, WSL researchers and colleagues now offer a solution that has quickly become popular among ecologists, as is demonstrated by a steadily increasing number of downloads and publications using the data.

CHELSA solves the problem of unequal distribution of meteorological stations by mechanistically downscaling a worldwide, constantly updated climate data set (ERA-interim = global re-analysis model data set) from coarse to fine resolution (to approx. 1 km2 grid size). Thus, the model provides worldwide, monthly time series for temperatures and precipitation for the period from 1979 to 2013. The model is now also applied to future climate change scenarios and paleo climate models, providing data for the next 100, and the last 21,000 years.

Tests show that CHELSA generates more reliable patterns of climate data than the previously used interpolations based on climate station data - especially in mountainous regions and in areas with low station density (see publication). In particular, precipitation patterns are better predicted compared to other models.
Dr. Dirk N. Karger and Prof. Dr. Niklaus E. Zimmermann from the WSL are responsible for hosting the database. The method was developed in cooperation with the Universities of Hamburg (Jürgen Böhner, O. Conrad, T. Kawohl), Zurich (M. Kessler, H.P. Linder) and Göttingen (H. Kreft).

Number of web visitis to chelsa-climate.org since 2016.

Contact