Impact of Ers-2 Altimeter wave height data on ECMWF wave analysis and forecast

  Impact of Ers-2 Altimeter wave height data on ECMWF wave analysis and
                             forecast
                             
                                by
                                
           Peter A.E.M. Janssen, Jean Bidlot and Bjoern Hansen              
           

                             Abstract                       
                             
We have performed an extensive investigation into the quality of the
ERS-2 Altimeter wave height data. First of all ERS-2 wave height
data were compared indirectly with the ones obtained from ERS-1.
Since the two satellites were not measuring wave at the same location
and the same time, WAM model results were used as a go-between. As a
result ERS-2 wave heights are about 8% higher than the ones obtained
from ERS-1. This conclusion was confirmed by a comparison of ERS-2 and
ERS-1 data with buoy observations over the one year period of
June 1995 until May 1996. Whilst the ERS-1 altimeter underestimates
the wave height by 15%, the ERS-2 altimeter only underestimates by
8%. Concentrating now on ERS-2 data only, there is also an indication
that the performance of the present Altimeter wave height algorithme 
depends on the sea state, since the comparison with buoy data shows
a better agreement in case of swell than in case of sea states
dominated by windsea. The number of collocations of buoy and Altimeter
observations in case of swell was rather limited so that one may question
the conclusion of the sea state dependence of the Altimeter algorithme.
In an attempt to remove this uncertainty we returned to the comparison
of Altimeter and WAM wave height and by distinguishing between wind sea
and swell the sea state dependence of the Altimeter algorithme was 
confirmed.
In  order to study the impact of the assimilation of ERS-2 data
we ran the WAM model for the period of December 1995 and generated both
analysis and 10 day forecasts. Assimilation of ERS-1 data was switched off.
As control we used the corresponding results from the operational suite,
which assimilates ERS-1 data only. By comparing with buoy data the ERS-2
analysis agrees, as expected, better with the buoy observations than the
ERS-1 analysis. Also, the mean forecast error in wave height is reduced,
suggesting that there is now a better balance between analysed and
forecast wave height. These promising results allowed for a switch of
ERS-1 to ERS-2 data in the ECMWF wave forecasting system by the end of
April 1996. From the buoy verification statistics the benificial impact
of this switched became evident at once.
We ascribe the improved quality of the ERS-2 wave height data to, when 
compared to ERS-1, an increase in pulse power by a factor of two which led to
a better signal to noise ratio. Further improvements are expected when
an Altimeter algorithme is introduced which allows for a dependence on
the sea state.
 

Keywords: ESA European Space Agency - Agence spatiale europeenne, observation de la terre, earth observation, satellite remote sensing, teledetection, geophysique, altimetrie, radar, chimique atmospherique, geophysics, altimetry, radar, atmospheric chemistry