KLIWAS 
Seite 10 
KLIWAS 
Project 1.03: 
Atmospheric and 
Oceanic Reference 
Data and Climate 
Projections for 
Coastal and Open 
Sea Areas 
The analysis focuses on the following parameters, which are commonly used to cal 
culate the boundary conditions of oceanic climate models: sea level pressure, wind 
speed and direction, air temperature, global radiation, cloud cover and precipitation. 
All the data are daily means/sums, except for the wind data of DMI, which has a reso 
lution of one hour. 
For the comparison of the regional climate models the 30 year period 1971-2000 is used in 
this study. All models arc driven with boundary conditions of the reanalysis ERA-40 (see 
section 2.2). The data of the different models were interpolated on a common grid with a 
horizontal resolution of 25 km x 25 km (see Fig 1). 
2.2 ERA-40 Re analysis 
Global reanalysis are often used for climatological studies and for the validation of 
global and regional climate models. Another application is nesting regional climate 
models in reanalysis. 
Reanalysis is a scientific method for developing a comprehensive record of the state 
of the atmosphere (or the ocean; see http://reamnalyses.org) using assimilation of data 
into models. This is chosen because of the lack of homogeneous observational data 
base spanning an extended historical period with the required spatial resolution. 
Available observations for the atmosphere come from different sources, e.g. classical 
land based observatories, voluntary observing ships, buoys, radiosondes, aircraft ob 
servations and satellites are used as input of an assimilation scheme of a numerical 
weather prediction model (NWP). The reprocessing for a long time period of several 
decades results in a three-dimensional globally gridded meteorological dataset for the 
atmosphere, a tool ideally qualified for subsequent climatological studies. 
The ERA-40 reanalysis is developed by the European Centre for Medium-Range 
Weather Forecasts (ECMWF); see Uppala et al. (2005). It uses a “frozen” version of 
the Integrated Forecasting System (IFS). ERA-40 covers the 45 year period from Sep 
tember 1957 to August 2002. The model data has a horizontal resolution of 1.125° 
(T159, about 125 km) and 60 vertical levels. 
http://www.ecmwf.int/research/era/do/get/era-40). Data is available six-hourly, i.e. 
for the following hours: 00, 06, 12 and 18 UTC. Conventional observations used for 
the assimilation come from different sources; an overview is given in Uppala et al. 
(2005). Satellite data are assimilated since 1972; cloud motion winds are included 
since 1979. 
The quality of ERA 40 data are evaluated by ECMWF itself can be seen under 
http://www.ecmwf.int/research/era/ERA-40/Performance/index.html. A significant 
improvement in comparison with the former ERA 15 reanalysis is found, but some 
deficiencies e.g. in radiation or water cycle (Bengtsson et al. 2004) still occur.