Meteorology and Atmospheric Physics

, Volume 91, Issue 1–4, pp 149–164

Assimilation of COST 716 Near-Real Time GPS data in the nonhydrostatic limited area model used at MeteoSwiss

  • G. Guerova
  • J.-M. Bettems
  • E. Brockmann
  • Ch. Matzler

DOI: 10.1007/s00703-005-0110-6

Cite this article as:
Guerova, G., Bettems, JM., Brockmann, E. et al. Meteorol. Atmos. Phys. (2006) 91: 149. doi:10.1007/s00703-005-0110-6


Application of the GPS derived water vapor into Numerical Weather Prediction (NWP) models is one of the focuses of the COST Action 716 “Exploitation of Ground based GPS for climate and numerical weather prediction applications”. For this purpose the GPS data covering Europe have been collected within the Near-Real Time (NRT) demonstration project and provided for Observing System Experiments (OSE). For the experiments presented in this manuscript the operational NWP system of MeteoSwiss is used. The limited area nonhydrostatic aLpine Model (aLMo) of MeteoSwiss covers most of western Europe, has a horizontal resolution of 7 km, 45 layers in the vertical, and uses a data assimilation scheme based on the Newtonian relaxation (nudging) method. In total 17 days analyses and two 30 hours daily forecasts have been computed, with 100 GPS sites assimilated for three selected periods in autumn 2001, winter and summer 2002. It is to be noted that only in the last period data from 10 french sites, i.e. west of Switzerland are assimilated.

The GPS NRT data quality has been compared with the Post-Processed data. Agreement within 3 mm level Zenith Total Delay bias and 8 mm standard deviation was found, corresponding to an Integrated Water Vapor (IWV) bias below 0.5 kg/m2. Most of the NRT data over aLMo domain are available within a prescribed time window of 1 h 45 min. In the nudging process the NRT data are successfully used by the model to correct the IWV deficiencies present in the reference analysis; stronger forcing with a shorter time scale could be however recommended. Comparing the GPS derived IWV with radiosonde observations, a dry radiosonde bias has been found over northern Italy. Through GPS data assimilation the aLMo analysis bias and standard deviation in the diurnal cycle has been reduced. The negative bias of –0.64 kg/m2 in the reference analysis has been reduced to 0.34 kg/m2 in GPS analysis. However, the diurnal cycle statistic from the forecast does show the characteristic negative bias only slightly reduced starting with the GPS analysis.

The GPS IWV impact on aLMo is large in June 2002 and moderate in September 2001 OSE. January OSE is inconclusive due to inconsistent use of humidity data below the freezing point. In June 2002 OSE, a substantial IWV impact is seen up to the end of the forecast. Over Switzerland the dry bias in the reference analysis has been successfully corrected and the 2 m temperature and dew point have been slightly improved over the whole aLMo domain. The subjective verification of precipitation against radar data in autumn 2001 and summer 2002 gives mixed results. In the forecast the impact is limited to the first six hours and to strong precipitation events. A missing precipitation pattern has been recovered via GPS assimilation in June 20 2002 forecast. A negative impact on precipitation analysis on June 23 has been observed.

The future operational use of GPS will depend on data availability; European GPS networks belong mainly to the geodetic community. A further increase of GPS network density in southern Europe is welcome. The GPS derived gradient and Slant Path estimates could possibly improve efficiency of IWV assimilation via the nudging technique.

Copyright information

© Springer-Verlag/Wien 2005

Authors and Affiliations

  • G. Guerova
    • 1
  • J.-M. Bettems
    • 2
  • E. Brockmann
    • 3
  • Ch. Matzler
    • 4
  1. 1.Institute of Applied Physics, University of BernSwitzerland
  2. 2.Federal Office of Meteorology and ClimatologyZurichSwitzerland
  3. 3.Swiss Federal Office of TopographyWabernSwitzerland
  4. 4.Institute of Applied PhysicsUniversity of BernSwitzerland

Personalised recommendations