In 2018, we reported that radiosonde descent data looked promising (ECMWF Newsletter No. 157). The data after balloon burst can provide an extra profile. Last year, ECMWF began to assimilate descent data from radiosondes launched in Germany, and recently it started to assimilate descent data from radiosondes launched from ships.
In June 2020, following an assimilation experiment, ECMWF started assimilating descent data from Vaisala RS41 radiosondes launched in Germany. The German national meteorological service (DWD) also started using these data at a similar time. The data quality is best from radiosondes with a) a pressure sensor and b) a parachute – the German radiosondes have both. After balloon burst the radiosondes can fall very fast, up to 100 m/s or more, although the descent rate is very variable and reduced with a parachute. Whilst falling fast, the measured temperatures are too high, apparently due to frictional heating. As the air density increases, the descent rate slows. Without a pressure sensor, the temperature biases can cause an offset in the descent pressures in the troposphere. It is possible to use the descent rate to correct most of the error. When this is applied in the Vaisala processing, we expect to use descent data from more stations. Much more detail on these issues is available in a manuscript (‘On the quality of RS41 radiosonde descent data’) under open review at Atmospheric Measurement Techniques.
In May 2021, a number of ships started reporting radiosonde descent data. Most of these are from the European Automated Shipboard Aerological Programme (ASAP) in the North Atlantic (see the map; for more details on ASAP, see ECMWF Newsletter No. 157). Four of these ships are using the RS41 radiosondes, and the descent quality looks good (see the graph). ECMWF started using the data operationally on 8 September 2021. Data use is restricted to pressures greater than 150 hPa, as for the land stations, to avoid very fast fall rates. These radiosondes have pressure sensors but not parachutes. They use smaller balloons than most land stations, and the temperature bias problems at upper levels are slightly less because they do not go as high. Five ships are providing descent data from Graw radiosondes; the quality makes the data unusable for now. Some ships use Modem radiosondes, but these are not providing descent data yet.
On average, radiosonde profiles from ships have more impact than land profiles because they are in datasparse regions. At land stations, the descent reports often stop some kilometres above the surface due to hills blocking the radiosonde signal, but for ships the descent reports can get very close to the surface. We hope to have more usable descent reports from ship and land stations in the future.