|
WSI’s Energy group recently issued its European Seasonal Outlook for the March-June 2004 period. The group utilises a variety of high-tech tools to create the seasonal forecasts.
For the period of March-May 2004, lower-than-normal temperatures will be generally confined to northern Scandinavia and regions near the Mediterranean Sea. As spring progresses, temperatures will cool down, relative to normal, in Scandinavia, and will warm up in the rest of Europe. This will result in lower-than-normal temperatures in the April-June period in Scandinavia and higher-than-normal temperatures in the UK and mainland . |
|
The WSI precipitation forecast shows that the March-May period is expected to be drier-than-normal in all of Scandinavia and mainland Europe, with wetter-than-normal weather in the UK. As spring progresses, precipitation will increase, relative to normal, in Scandinavia. This will result in a wet April-June period in Scandinavia and the UK, with dry conditions continuing in mainland Europe.
 
|
|
Detailed Forecast Discussion
The past winter was relatively warm in much of Europe (below, right), with the exceptions of northern Scandinavia and parts of eastern mainland Europe. This pattern was well predicted by our November forecast (below, left), which depicted the warm weather over most of mainland Europe and the cold temperatures in northern Scandinavia.
 
|
 |
WSI Energycast forecast for winter (December-February), made November 19 (left) and actual winter temperature anomaly (right)
Our objective forecast models are, to a large extent, driven by ocean
temperatures. During February, SST anomalies had increased slightly in
the eastern Atlantic Ocean, but had cooled a bit in the Mediterranean.
In general, SSTs are still slightly warmer-than-normal in and around Europe,
which should lead to a generally warm spring.
As part of the seasonal forecasting process, WSI runs the Community Climate Model (CCM) version 3.6 in-house twice a month on a high-speed suite of computers. The CCM, like all numerical models, is initialized with atmospheric and oceanic data before the run begins. Mathematical equations representing physical processes in the atmosphere, in the ocean, and on the land are then used to predict the future state of atmosphere for the following six months. The CCM is run with two different land-surface modules (LSMs). An LSM controls the transfer of heat and moisture between the vegetation/soil and the overlying atmosphere. In order to quantify the skill of the CCM, we have run the model in “hind cast” mode for 20 years for both LSMs. Based on the results of this long model run, we know where and when the model is or is not skillful for each of the two LSMs. The two LSMs each have their strengths and weaknesses, with one of the LSMs being more skillful in the warm season and the other in the cool season. Using this knowledge, WSI intelligently combines the output from both LSMs to optimize the skill of the seasonal forecast.
For this forecast period, our climate models are still predicting widespread warmer-than-normal conditions (below, right), with lower-than-normal temperatures confined to areas along the Mediterranean and extreme northern Scandinavia. The February run, though still warm, is still much cooler in most locales than the January run (below, left). This may be due to the much colder (relative to normal) Arctic waters.
|
|
|
Climate model output for the March-May period from our January run (left) and our February run (right)
WSI also uses sophisticated statistical techniques to predict seasonal weather patterns. The February output from one of these techniques (below, right) suggests a cool spring for much of Europe. The February run is significantly colder than the January run (below, left), matching the climate model trend.
|
|
Statistical model temperature output for the March-May period from our January run (left) and our February run (right)
The other statistical technique is used to predict precipitation only. The forecast for the spring is shown below, from both our January run (below, left) and our February run (below, right). Both runs are consistent in indicating a dry spring in general, but the February run is dry in all locations except for the UK.
|
|
Statistical
model precipitation output for the March-May period from our January run
(left) and our February run (right).