Original paper

Shifts of climate zones in multi-model climate change experiments using the Köppen climate classification

Hanf, Franziska; Körper, Janina; Spangehl, Thomas; Cubasch, Ulrich

Meteorologische Zeitschrift Vol. 21 No. 2 (2012), p. 111 - 123

published: Apr 1, 2012

DOI: 10.1127/0941-2948/2012/0344

BibTeX file

ArtNo. ESP025012102001, Price: 29.00 €

Download preview PDF Buy as PDF


This study investigates the future changes in the climate zones' distribution of the Earth's land area due to increasing atmospheric greenhouse gas concentrations in three IPCC SRES emissions scenarios (A1B, A2 and B1). The Köppen climate classification is applied to climate simulations of seven atmosphere-ocean general circulation models (AOGCMs) and their multi-model mean. The evaluation of the skill of the individual climate models compared to an observation-reanalysis-based climate classification provides a first order estimate of relevant model uncertainties and serves as assessment for the confidence in the scenario projections. Uncertainties related to differences in simulation pathways of the future projections are estimated by both, the multi-model ensemble spread of the climate change signals for a given scenario and differences between different scenarios. For the recent climate the individual models fail to capture the exact Köppen climate types in about 24-39 % of the global land area excluding Antarctica due to temperature and precipitation biases, while the multi-model ensemble mean simulates the present day observation-reanalysisbased distribution of the climate types more accurately. For the end of the 21st century compared to the present day climate the patterns of change are similar across the three scenarios, while the magnitude of change is largest for the highest emission scenario. Moreover, the temporal development of the climate shifts from the end of the 20st century and during the 21st century show that changes of the multi-model ensemble mean for the A2 and B1 scenario are generally within the ensemble spread of the individual models for the A1B scenario, illustrating that for the given range of scenarios the model uncertainty is even larger than the spread given by the different GHG concentration pathways. The multi-model ensemble mean's projections show climate shifts to dryer climates in the subtropics (Australia, Mediterranean Basin, southern Africa). This is consistent with an increase of area classified as Tropical Savanna Climate as well as Dry Climates. Furthermore, there is a poleward extension of the warmer climate types in the northern hemisphere causing a retreat of regions with Cold Climate with Moist Winter and Tundra Climate. The European region shows largest changes comparing the shifts in the different continents (37.1 % of the European land area) as a result of a large extension of the Humid Temperate Climate across eastern and north-eastern Europe at the cost of the Cold Climate with Moist Winter.