Assimilation of Dual-Polarization Radar Observations into Météo France Convective Scale Model AROME

The main purpose of this study is to examine how dual-polarization weather radar observations could be used to improve the analyses and the short-term forecasts of convective scale atmospheric models.

First, a polarimetric radar forward operator was added to the Meso-NH mesoscale model (Lafore et al., 1998) by taking advantage of the detailed mixed-phase microphysics of the model. Comparisons between simulated and observed polarimetric variables were conducted at S, C and X-band on two mid-latitude deep-convection cases and demonstrated a rather good agreement between observations and simulations. These comparisons were used to validate the forward operator (Augros et al., 2016) that was subsequently deployed into the Météo France convective scale operational model AROME (Seity et al., 2011) that has the same 1-moment microphysics scheme as Meso-NH.

Next, the 1D+3D-Var assimilation scheme, used operationally to assimilate radar reflectivity in AROME (Caumont et al, 2010; Wattrelot et al, 2014), has been extended to test the assimilation of the differential phase Zdr and the specific differential phase Kdp. In this indirect assimilation scheme, a 1D Bayesian method is used to infer vertical profiles of humidity at the observation location, from a linear combination of background humidity profiles, with weights determined by the differences between observed and simulated Zhh, Zdr and Kdp. The retrieved humidity profiles are then assimilated together with the other observations in AROME using 3D-Var.

The impact of attenuation correction, and of the use of Zdr and/or Kdp on the retrieved humidity profiles will be shown, particularly in case of large Kdp values and/or Path Integrated Attenuation (PIA). A validation of the retrieved humidity profiles by comparison with IWV (Integrated Water Vapor) from ground-based GPS observations will also be presented.

Finally, the results of assimilation experiments with or without polarimetric observations will be discussed. The impact of the assimilation of Kdp is particularly evidenced in a case of strong convection and partial beam blockage. After assimilating Kdp, humidity analysis are found to be higher in a convective region (than when Zhh only is assimilated) if partial beam blockage affects Zhh observations. For the two convective cases studied, the impact of dual-polarization observations on precipitation forecast is slightly positive, during the periods of intense convection. The examination of a larger number of cases should be conducted in the near future to comfort the benefit of DPOL observations on precipitation forecasts.

References:

Augros, C., O. Caumont, V. Ducrocq, N. Gaussiat, and P. Tabary, 2016 : Comparisons between S, C, and X band polarimetric radar observations and convective-scale simulations of HyMeX first special observing period. Quarterly Journal of the Royal Meteorological Society, 142, Issue S1: 347-362

Augros, C., O. Caumont, V. Ducrocq, and N. Gaussiat, 2017 : Assimilation of radar dual-polarization observations in AROME model. Quarterly Journal of the Royal Meteorological Society, in review

Caumont, O., V. Ducrocq, É. Wattrelot, G. Jaubert, et S. Pradier-Vabre, 2010 : 1D+3DVar assimilation of radar reflectivity data : a proof of concept. Tellus A, 62 (2), 173–187

Lafore, J. P., et al., 1998 : The Meso-NH atmospheric simulation system. Part I : adiabatic formulation and control simulations. Annales Geophysicae, 16 (1), 90–109

Seity, Y., P. Brousseau, S. Malardel, G. Hello, P. Bénard, F. Bouttier, C. Lac, et V. Masson, 2011 : The AROME-France convective-scale operational model. Monthly Weather Review, 139 (3), 976–991

Wattrelot, E., O. Caumont, et J.-F. Mahfouf, 2014 : Operational implementation of the 1D+3DVar assimilation method of radar reflectivity data in the AROME model. Monthly Weather Review, 142, 1852–1873