GROUND BASED MICROWAVE RADIOMETER – CLOUD DETECTION, SKY-CLEARING AND O-B STATISTICS

M. Löffler^1,2, C. Knist³, B. Pospichal², A. Claußnitzer¹, U. Löhnert²

¹German Weather Service (DWD), Technical Infrastructure and Operations, Potsdam, Germany, ²Institute for Geophysics and Meteorology, University of Cologne, Germany, ³German Weather Service (DWD), Meteorological Observatory Lindenberg – Richard Aßmann Observatory, Lindenberg Tauche, Germany

In assimilation experiments at the German Weather Service (DWD), data from ground-based microwave radiometers (MWR) show a positive impact on the numerical weather prediction (NWP) under clear-sky conditions. However, liquid clouds involve large random differences between model and observation and therefore, the most frequent reason for rejecting data in the assimilation process is the suspected presence of clouds. With a balanced detection of clouds, an artificial cloud-clearing can be performed. Hence, it is possible to obtain cloud-free brightness temperatures (TB) which represent the current thermodynamic conditions of the atmosphere.

First, we present the resulting effect of a neural network (NN) based liquid water cloud detection on observation minus background (O-B, using the ICON-D2 model) statistics and compare it to an established cloud detection scheme. The NN relies solely on the observed TB and is trained with TB computed with a line-by-line radiative transfer model (Rosenkranz 2022, non-scattering) from ERA5 reanalysis data.

We will also present O-B statistics with artificially cleared TB observations. The spectral signature of the liquid water is subtracted from the observed TB spectrum using a NN retrieval trained on ERA5. The cloud detection and sky clearing allow a thorough discussion of 2-years O-B statistics, which provides insights on the model performance and allows monitoring instrument errors. The presented progress on sky clearing of TB can be a way forward to increase the benefits of assimilating MWR TB in NWP models.