Partner und Internationale Organisationen
(Englisch)
|
A, B, CY, CZ, DK, FIN, F, D, GR, H, IRL, I, NL, N, PL, P, SI, E, S, CH, GB
|
Abstract
(Englisch)
|
Project Title: 'Assimilation of Three-Dimensional Radar Reflectivity into a Nonhydrostatic NWP Model This project, performed in the framework of COST-717 'Use of Radar Observations in hydrological and NWP Models', aims at the development of a method to assimilate three-dimensional radar reflectivity information into the Alpine Model (aLMo), the Swiss version of the Lokal Modell (LM) developed in the Consortium for Small Scale Modelling (COSMO). This is done on the basis of the Latent Heat Nudging (LHN) technique, originally developed for satellite data and adopted for radar by the UK Metoffice. It uses observed surface rain rates as input for the assimilation and adds/removes latent heat related to precipitation processes, where the vertical distribution of the total heat in an atmospheric column is determined by the model. In the first year of the project a LHN code was adopted for the use within the model and observation framework of MeteoSwiss and was tested on a case of severe convection over Switzerland. Here we report on the progress made during the second year. An idealized model environment has been set up in order to investigate the performance and the behaviour of the assimilation scheme. Results obtained from studies using this environment include: · LHN was successful in analysing and initiating an idealized supercell storm. A free forecast starting from the analysis at different times compared very well with the reference run indicating, that the LHN scheme is able to force this idealized system in the right way. · A sensitivity study suggests that the LHN scheme needs radar observations sampled with a high temporal frequency compared to the lifetime of the supercell in order to produce a reasonable precipitation analysis. · The low level humidity of the environment in which the storm is embedded turned out to be a critical factor for the successful storm development. In addition to the real-case study considered during the first year of the project, a second case of strong, long-lived summer convection has been investigated using LHN. The value of the radar data for the model precipitation analysis has been confirmed in this second case. The assimilation algorithm was able to produce rain fields close to those observed by radar. Moreover, the positive impact in the free forecast starting from the LHN analysis lasted longer than in the first case. There is a general trend, both in operational and research NWP towards using high resolution models aiming at a more realistic simulation of convective precipitation without relying on convection parametrization schemes. Therefore the LHN scheme has been tested using a grid size of 2.2km. The overall benefit compared with the 7km resolution has yet to be studied in detail.
|
