Weather radar data quality in Northern Europe: beam propagation issues

In the 3rd Baltic Weather Radar Workshop in 2004, the identification of uncertainty related to radar beam propagation and the adjustment of vertical profiles of reflectivity (VPR) have been assigned highest priority in Northern Europe. Two projects have been formulated to tackle these challenges and provide solutions applicable to the operational Nordic Weather Radar Network (NORDRAD).

This paper addresses the beam propagation project focusing on beam blockage and anomalous propagation. At high latitudes, where precipitation frequently falls as snow with low echo top heights, even partial beam blockage can lead to total information loss. When deriving precipitation information from radar observations, radar beam propagation is usually assumed to be ideal. For a given elevation and opening angle, the volume and altitude of each radar bin is regarded to be a function of distance from the radar. For the atmospheric variables affecting the index of refraction, standard conditions are assumed. In reality, radar beam propagation is affected by blockages from buildings, trees, mountains etc. and refractivity variations in the atmosphere.

The Beam Propagation Model (BPM) has been developed at met.no. It simulates the radar's field of view using information about the scan geometry, a digital terrain model, and the atmospheric conditions (i.e. temperature, humidity and air pressure). Additionally, a polar quality index field is computed for each NORDRAD radar. We will present some BPM simulations for Norwegian and Swedish radar sites and a comparison with long-time radar derived precipitation accumulations. Moreover, we will show preliminary results from anomalous propagation cases modeled with both radiosonde and numerical weather prediction (NWP) model data as input. The BPM output is compared to echo pattern in corresponding radar data.

The quality index field can be used to optimize the compositing algorithm. It is also valuable as input for SMHI's Mesoscale Analysis (MESAN) system to improve precipitation analysis. Finally, data quality information is very useful for NWP models in terms of data assimilation, and for hydrological models. Recently, a new COST action (731) has been initiated dealing with the propagation of uncertainty in advanced meteo-hydrological forecast systems. The operational application of quality information for radars and radar data is addressed in the EUMETNET OPERA program.