In August 2006 exceptionally high precipitation amounts were recorded in The Netherlands, in particular in areas close to the sea. August 2006 was preceded by a long period of very warm weather, and consequently the temperature of the North Sea was very high at the end of July. When the atmospheric circulation changed to a very strong North Westerly flow on the 1st of August, this resulted in strong shower activity and lots of precipitation. The question was “how much did the warm North Sea water contribute to the high precipitation amounts in August 2006?” This question is answered by performing simulations with the KNMI regional climate model RACMO2. Figure 1 shows the observed precipitation amounts in August of 300 stations in the Netherlands (left panel). The middle panel shows results of RACMO2, when forced with realistic atmospheric conditions at the lateral boundaries and observed sea surface temperatures. The right panel shows results of an identical simulation with RACMO2, with the exception of the sea surface temperature which is now prescribed to be approximately 2 degrees colder. Such colder sea surface temperatures are representative of the normal sea surface temperature during August. Comparing both RACMO2 simulations it is clear that the high sea surface temperature contributed considerably to the high precipitation amounts in the coastal area. A more in depth analysis showed that about 30 % of the rain in the coastal area was due to the high sea surface temperature.

Lenderink, G., E. van Meijgaard and F. Selten, Intense coastal rainfall in the Netherlands in response to high sea surface temperatures: analysis of the event of August 2006 from the perspective of a changing climate. Clim. Dyn., 2009, 32, 19-33, doi:10.1007/s00382-008-0366-x (link to own preprint).

The above case study clearly shows the potential influence of the North Sea on the Dutch climate. The influence of the North Sea is also visible in the climatological mean. In spring with relatively cold sea water the coastal area is relatively dry compared to the interior of the Netherlands, whereas in autumn the situation in reversed. The difference between coastal precipitation (less than ~40 km from the coastline) and precipitation in the interior is now called “the coastal effect”.

It is not clear how this coastal effect responds to climate change. The main factor causing the coastal effect is the temperature difference between land and sea, and changes herein could be small. The coastal effect also depends strongly on the atmospheric flow with, for instance, north-westerly flows resulting in much precipitation in the coastal area in autumn. Therefore, changes in the general circulation could have a large impact. On the other hand, rising seawater temperatures and therefore higher moisture content near the coast could increase the effect. Also projected drying over the continent in summer could enhance the effect. Potentially important is also the influence of urbanization and land-use changes, which could also impact on the precipitation distribution within The Netherlands. Besides mean precipitation, extremes are of major importance. Rain radar data from recent years show a remarkable peak of daily precipitation extremes near the coast in the province “Zuid-Holland”.

Scenarios issued by KNMI in 2006 did not discriminate within The Netherlands. Projected changes are the same over the Netherlands, and spatial differences in the climate of 2050 are therefore similar to the differences already existing in the present climate. Two reasons (amongst others) for this are: the relatively coarse resolution of the regional climate models at that time (50 km), and the absence of a good description of the sea water temperatures of the North Sea.

Presently, we are running our regional climate model RACMO at a resolution of 10-25 km. Regional climate models usually take sea water temperatures from coarse resolution general circulation models (GCM) , which have a typical resolution of 200 km and therefore lack small scale features close to the coast. We are therefore implementing a simple ocean slab model in our regional climate model to restore these small scale features to the SST from the GCM, and are investigating their effect on the precipitation distribution over the Netherlands.