The primary reason why precipitation extremes are expected to increase in a warmer climate follows from the fact that a warmer atmosphere can ‘hold’ more moisture. The increase in the moisture-holding capacity of the atmosphere with temperature occurs at a rate given by the Clausius-Clapeyron relation: approximately 7% per degree temperature rise. If the relative humidity in the future climate remains approximately the same as in the present-day climate - and there are good reasons for this when there is no severe water shortage - the amount of water vapour in the atmosphere will increase at the same rate. Now, the commonly used argument is that in extreme precipitation events all water vapour in the air (or a constant fraction thereof) is converted to rain. Hence, extreme precipitation should scale with the Clausius-Clapeyron relation.

But, it is not that simple. There are more factors that influence the strength of showers, and these obviously could change as well. Such factors are the atmospheric motions on different scales and atmospheric stability. So, let us first look at the observations of the present day climate.

Figure 1 shows the dependency of extremes in hourly precipitation on the dew point temperature in data of 27 stations over the last 15 years. The dew point temperature is actually a measure of atmospheric moisture. But, if the relative humidity does not change, a one degree temperature rise is equivalent with a one degree rise in dew point temperature. For temperatures above 10 ^oC, the most extreme events (denoted by the 99.9% value) reveal a dependency of 14 % per degree – double the dependency predicted by the Clausius-Clapeyron relation. Results are plotted on a logarithmic vertical axis, so that each vertical tic on the axis is doubling of precipitation intensity. The red dotted lines are exponential relations of 14 % per degree. Thus, each 5.3 degree rise in temperature causes a doubling of hourly precipitation intensity.

The reason of the enhanced dependency (compared to the Clausius-Clapeyron relation) is likely related to a positive feedback from the upward motions in the cloud. Heat is released when water vapour condenses into droplets and eventually into precipitation, and this heat fuels the upward motions in the cloud. Higher temperatures lead to more water vapour condensing in the cloud, more heat release, stronger updrafts, and therefore stronger condensation, and precipitation. Yet, the cause is still under debate, and alternative explanations exist. Results appear robust however, and a 14 % per degree dependency is even found in data from Hong Kong. The higher temperatures there lead to intensities approximately three times larger than observed in the Netherlands.

The temperature in the Netherlands has increased in the last decades. Keeping the shown dependency of hourly precipitation extremes on the temperature in mind, do we see an increase in extreme hourly precipitation ? We are presently investigating this, and results will soon appear here.

Our future climate is inherently uncertain since many aspect of regional/local climate change are still poorly understood, and modelled.. Yet, the previous result suggest that hourly precipitation extremes could increase at a rate of 14 % per degree temperature rise. What do models predict? Figure 2 shows the change in hourly precipitation extremes in three different integrations. Two integration with our regional climate model RACMO, yet with different input at the lateral boundaries, and one integration with a different climate model (CLM). A large spread in projected changes is shown. Differences are due to the difference in lateral forcing, with the integration RACMO2-M projecting the largest temperature increase and only small changes in the atmospheric circulation. But also the differences in regional model matter. Despite that RACMO2-E and CLM-H are similar in the large scale response (temperature and circulation change), increases in hourly precipitation extremes are much larger in RACMO2-E. In RACMO2 the response in hourly precipitation extremes is for large parts of Western Europe larger than 10 % per degree, whereas in CLM-H it is below 7 % per degree.

Lenderink, G. and E. van Meijgaard, Increase in hourly precipitation extremes beyond expectations from temperature changes Nature Geoscience, 2008, 1, 8, 511-514, doi:10.1038/ngeo262.

Lenderink, G. and E. van Meijgaard, Reply: Unexpected rise in extreme precipitation caused by a shift in rain type? Nature Geoscience, 2009, 2, 6, 373, doi:10.1038/ngeo524.

Lenderink, G. and E. van Meijgaard, Linking increases in hourly precipitation extremes to atmospheric temperature and moisture changes Environmental Research Letters, 2, 2010, 5, 025208, doi:10.1088/1748-9326/5/2/025208, open access.