Assimilation of GPS radio occultation data at DWD

Abstract

Abstract. We describe the status of the assimilation of bending angles from GPS radio occultations in the 3D-Var for DWD's operational global forecast model GME ("Global Model for Europe"). Experiments show that the assimilation of GPSRO data leads to a significant reduction of biases in the analyses of temperature, humidity and wind in the upper troposphere and the stratosphere, as well as a better r. m. s. fit in the comparison to radiosondes. The impact on forecasts is most prominent in the data sparse Southern Hemisphere, but is also quite notable in the Northern Hemisphere extratropics. The positive results found in the impact experiments lead to the implementation of the assimilation of GPS radio occultations from GRACE-A, FORMOSAT-3/COSMIC and GRAS/MetOp-A into the operational suite on 3 August 2010. We also show some initial results from assimilation experiments using radio occultation data from the German research satellite TerraSAR-X.

Implementation of a GPS-RO data processing system for the KIAPS-LETKF data assimilation system

“…In the meantime, we continue improving our current system (such as tuning parameters within LETKF, replacing the forecast model by a better one, etc.) so that it could give comparable results to those of other centers (e.g., Cucurull et al, 2007;Anlauf et al, 2011) in the near future.…”

“…Because of these benefits, many operational numerical weather prediction centers, such as the Met Office of the United Kingdom, ECMWF, NCEP, Météo-France, Environment Canada, and JMA, have started incorporating GPS-RO soundings into their assimilation systems with clear positive impacts on weather forecasting (e.g., Healy, 2008;Buontempo et al, 2008;Cucurull and Derber, 2008;Aparicio et al, 2009;Rennie, 2010). In particular, GPS-RO data assimilation shows strong sensitivity to upper atmosphere temperature structures, an area that is only weakly constrained by other observations in the analysis and that is prone to large model uncertainties (Anlauf et al, 2011).…”

“…Generally, the ray-tracing operator (Gorbunov and Kornblueh, 2003) for the bending angle is reported to lead to best results in terms of standard deviation of the difference between observation and model background (Pingel and Rhodin, 2009). However, the use of a ray-tracing operator requires a high computational cost, so most numerical weather prediction centers (e.g., ECMWF, UK Met Office, DWD, and Météo-France) currently use one-dimensional observation operators for the bending-angle assimilation and inflate the observation errors to partially compensate for this source of forward model error (Anlauf et al, 2011). The assumption of spherical symmetry enables calculation of the bending angle from a one-dimensional integration of the background refractivity lapse-rate profile located at the vertical level of the observation point.…”

Radio Occultation Observations as Anchor Observations in Numerical Weather Prediction Models and Associated Reduction of Bias Corrections in Microwave and Infrared Satellite Observations

“…Typically, profiles of refractivity or bending angle are used, and their assimilation makes use of forward operators that neglect horizontal gradients of atmospheric refractivity (e.g., Healy and Th epaut 2006;Cucurull and Derber 2008;Poli et al 2009;Rennie 2010;Anlauf et al 2011). In all known cases, the significant positive impact in terms of weather forecast skill has been found, regardless of the type of RO retrieval used.…”

Implementation of a quality control for radio occultation observations in the presence of large gradients of atmospheric refractivity

“…For example, observations located below regions where the refractivity lapse rate is below −50 km −1 , either from the observations or the background field, are rejected at Météo France (Poli et al, 2009). At the German Weather Service (Deutscher Wetterdienst), starting with an impact height of 8 km, a profile section below a non-monotonous bending angle profile is discarded when the bending angle decreases by more than 3 times the assigned observation standard deviation (Anlauf et al, 2011). Bending angle observations are rejected at the European Centre for Medium-Range Forecasts when the refractivity gradient reaches half the critical gradient (S. B. Healy, personal communication, 2011).…”