Réf. Heimann & Sept 2000 -A

Référence bibliographique complète
HEIMANN D., SEPT V., 2000: Climatic change estimates of summer temperature and precipitation in the Alpine region. Theor. Appl. Climatol., Vol 66, p 1-12.

Abstract: Climatic changes of summer temperature and precipitation in the greater Alpine region are assessed by using statistical-dynamical downscaling. The downscaling procedure is applied to two 30-year periods (1971±2000 and 2071±2100, summer months only) taken from the results of a transient coupled ocean/atmosphere climate scenario simulation with increasing greenhouse gas concentrations. The downscaling results for the present-day climate are compared with observations. The estimated regional climate change during the next 100 years shows a general warming. The mean summer temperatures increase by 3 to 5 Kelvin. The most intense climatic warming is predicted in the western parts of the Alps. The amount of summer precipitation decreases in most parts of central Europe by more than 20 percent. Increasing precipitation is simulated only over the Adriatic area and parts of eastern central Europe. The results are compared with observed climate trends for the last decades and results of other regional climate change estimations. The observed trends and the majority of the simulated trends (including ours) have a number of common features. However, there are also climate change estimates of other groups which completely contradict our results.

Greater Alpine Region, downscaling methods, 1971-2000/2071-2100

Organismes / Contacts

Institut für Physic der Atmosphäre, Potsdam Institute für Klimatologie

(1) - Paramètre(s) atmosphérique(s) modifié(s)
(2) - Elément(s) du milieu impacté(s)
(3) - Type(s) d'aléa impacté(s)
(3) - Sous-type(s) d'aléa
Temperature and precipitation      

Pays / Zone
Massif / Secteur
Site(s) d'étude
Période(s) d'observation
Greater Alpine Region
4 to 19 deg E, 43 to 49 deg N, 725 000 km²
For the precipitations: 5000 rain gauge stations in Germany, Austria, France, Italy, Slovenia and Croatia.
1971-2000 / 2071- 2100

(1) - Modifications des paramètres atmosphériques

As a consequence of the prescribed increase of greenhouse gas concentrations (period 2071- 2100 vs. 1971-2000, summer months only) the coupled ocean/atmosphere GCM simulates altered circulation patterns and a general rise in air temperature and absolute humidity. The striking increase of the geopotential height over western and central Europe corresponds with an increased number of days with easterly and northeasterly flow over the Alpine area.

The maximum increase of the temperature is simulated over the western parts of the Alps (above 4.5°C) for the 2071-2100 period. This value provided by a statistical-dynamical downscalng method is under the prediction of the GCM (around 6°C for the same period and same scenario ISC92a). The warming gradually diminishes toward the north.

The downscaled temperature change (period 2071-2100 vs. period 1971-2000) ranges between + 2.8K (Bolzano) and 4.9K (Nice). Nevertheless, the average temperature change (3.90 K) exceeds the mean absolute error in the present-day temperatures by a factor of 6.

Major discrepancies appear in the Ticino area (southern Switzerland) where the model strongly overestimates the mean precipitation. The same applies in the southwesternmost parts of the Alps and along the crest of the Apennines. Underestimations appear along the northern and south-eastern boundary of the Alps.

The simulated change of summer precipitation shows a significant decrease by 20 to 50 percent in most parts of western central Europe. An even stronger decrease (50 to 100 %) is simulated along the Saone-Rhône valley in France.

The strongest absolute decline (-7mm/day), however, appears over the western Alps (northwest of Milano) just where the result for the present-day climate shows the strongest positive bias. Increasing precipitation is simulated around the Adriatic Sea, the eastern parts of the Alps and parts of eastern central Europe. Again, the strongest relative change (>100 percent increase) appears over the eastern Po valley where the present-day result has the strongest (negative) bias. At many Alpine and northern Italian stations the absolute bias exceeds the simulated absolute value of the climatic change. This fact certainly limits the reliability of the climate change results in these areas.


Informations complémentaires (données utilisées, méthode, scénarios, etc.)

In this study statistical-dynamical downscaling is applied to two 30-year periods of a transient climate scenario simulation with a coupled ocean-atmosphere GCM. The first period represents the present-day climate (1971-2000) while the second period represents a possible largescale climate 100 years later (2071-2100). The downscaling was performed for the greater Alpine region. The study is restricted to the summer months, i.e., June, July, and August.

The data used are the results from the coupled atmosphere-ocean modelling system ECHAM4/OPYC3 with ISC92a scenario. In order to assess a possible climate change, the results of two 30-year periods 1971-2000 and 2071-2100 were selected.

The precipitation grid-based data (mesh size: 25x25 km2) were derived from the daily measurements of about 5000 rain gauge stations in southern Germany, Switzerland, Austria, south-eastern France, northern Italy, Slovenia, and Croatia.

(2) - Impacts du changement climatique sur le milieu naturel



Sensibilité du milieu à des paramètres climatiques
Informations complémentaires (données utilisées, méthode, scénarios, etc.)

(3) - Impacts du changement climatique sur l'aléa

Paramètre de l'aléa
Sensibilité du paramètres de l'aléa à des paramètres climatiques
Informations complémentaires (données utilisées, méthode, scénarios, etc.)

(4) - Remarques générales


(5) - Syntèses et préconisations
The estimation of regional climate change in the Greater Alpine Region is unertain for the following reasons: uncertainities in the assumptions underlaying the GCM simulations, error in the GCM results due to physical approximations and numerical deficiencies in the global model, inaccuracies in the statistical-dynamical downscaling method, errors in the regional model results due to physical approximations, numerical resoution, forcing mechanism and other numerical deficiencies in the regional model.