We found a match
Your institution may have rights to this item. Sign in to continue.
- Title
Assimilating cloudy and rainy microwave observations from SAPHIR on board Megha Tropiques within the ARPEGE global model.
- Authors
Duruisseau, Fabrice; Chambon, Philippe; Wattrelot, Eric; Barreyat, Marylis; Mahfouf, Jean‐François
- Abstract
The Megha‐Tropiques satellite was launched in 2011 with a microwave sounder called SAPHIR onboard. This instrument probes the atmosphere with six channels around the 183.31 GHz water vapour absorption band. Its observations are sensitive to water vapour as well as to hydrometeors. This instrument was proven to be useful for data assimilation by different numerical weather prediction centres, in particular for clear‐sky assimilation. At Météo‐France, SAPHIR observations have been routinely assimilated in clear sky since 2015 in the ARPEGE global model. The present article introduces a framework to complement this clear‐sky assimilation route by a new cloudy and rainy assimilation route for satellite microwave brightness temperatures. This framework is based on several steps including a Bayesian inversion of the SAPHIR brightness temperatures into relative humidity retrievals, which are then assimilated within the ARPEGE global model. This study presents the methodology of assimilation, including the development of two error models, one for the Bayesian inversion, and one for the observation errors of relative humidity retrievals within the ARPEGE 4D‐Var data assimilation system. The forecast scores obtained with this methodology over a three‐month period indicate a positive impact of SAPHIR cloudy and rainy observations within the ARPEGE system, in particular on tropical temperature and wind forecasts for which the improvements range from 0.5 to 1.7% on standard deviations with respect to the ECMWF analysis and up to a +60 h lead time. Microwave observations from space‐borne sounders provide precious data for initializing numerical weather prediction models, in particular in cloudy and rainy conditions. Indeed, numerical models often mis‐locate clouds and precipitation as shown in the figure for a convective case along the Thailand coastline for which the observed clouds (left figure) are located eastward of the modelled ones (right figure). The article presents a methodology for taking benefit of microwave observations to better predict clouds and precipitation through better initial conditions.
- Subjects
THAILAND; NUMERICAL weather forecasting; MICROWAVES
- Publication
Quarterly Journal of the Royal Meteorological Society, 2019, Vol 145, Issue 719, p620
- ISSN
0035-9009
- Publication type
Article
- DOI
10.1002/qj.3456