|Paper title||Progress in preparations towards monitoring and assimilation of EarthCARE observations at ECMWF|
|Form of presentation||Poster|
In recent years at ECMWF, a series of projects were carried out focusing on developments towards direct assimilation and monitoring to exploit space-borne cloud radar and lidar data for Numerical Weather Prediction (NWP) models. Although active observations from such profiling instruments contain a wealth of information on the structure of clouds and precipitation, they have never been assimilated directly in any global NWP model.
To prepare the data assimilation system for the new observations of cloud radar reflectivity and lidar backscatter, several important developments were required. This included the specification of sufficiently accurate observation operators (i.e. models providing equivalent model fields to observations), as well as defining flow-dependent observation errors, and appropriate quality control strategy and bias correction scheme. The feasibility of assimilating CloudSat and CALIPSO data, currently the only available data from space-borne radar and lidar with global coverage, into the Four-Dimensional Variational (4D-Var) data assimilation system used at ECMWF has been investigated. Including cloud radar reflectivity and lidar backscatter in the assimilation system had a positive impact on both the analysis and the subsequent short-term forecast. By running experiments for different seasons and combining them to increase statistical significance lead to promising results; improvements to the zonal mean forecast skill score in the short- and medium-ranges for large-scale variables were found almost anywhere, with the largest impact on storm-tracks and in the tropics.
The performed studies using CloudSat and CALIPSO observations prepared grounds for assimilation of such observation types from the future EarthCARE mission. Additionally, the system developments will facilitate the monitoring of observations both in an operational sense and for model evaluation as soon as observations become available after the mission launch. By using a monitoring system that combines information from observations and model, a statistically significant drift in the measurements can be detected faster than monitoring observations alone. Also the monitoring system allows validation of the observations along the whole EarthCARE track.