Comparing ECMWF high‐resolution analyses with lidar temperature measurements in the middle atmosphere

• Published in: Quarterly Journal of the Royal Meteorological Society
• Main Authors: Ehard, Benedikt, Malardel, Sylvie, Dörnbrack, Andreas, Kaifler, Bernd, Kaifler, Natalie, Wedi, Nils
• Other Authors: Deutsche Forschungsgemeinschaft
• Format: Article in Journal/Newspaper
• Language: English
• Published: Wiley 2018
• Subjects: Sodankylä
• Online Access: http://dx.doi.org/10.1002/qj.3206; https://api.wiley.com/onlinelibrary/tdm/v1/articles/10.1002%2Fqj.3206; https://onlinelibrary.wiley.com/doi/pdf/10.1002/qj.3206; https://onlinelibrary.wiley.com/doi/full-xml/10.1002/qj.3206; https://rmets.onlinelibrary.wiley.com/doi/pdf/10.1002/qj.3206

Middle atmospheric lidar temperature observations conducted above Sodankylä, Finland (67.4°N, 26.6°E), during December 2015 are compared with two estimates of the atmospheric state computed by the integrated forecast system (IFS) of the European Centre for Medium‐Range Weather Forecasts (ECMWF). The first set corresponds to an hourly sampling of the middle atmosphere by high‐resolution analyses and very short‐range forecasts produced by the operational IFS cycle 41r1 at a horizontal resolution of 16 km. The second set is retrieved from the upgraded IFS cycle 41r2 (horizontal resolution 9 km), which was running in parallel with cycle 41r1 during the validation before it became operational. A remarkable agreement between both IFS datasets and the lidar temperature observations above Sodankylä is found below 45 km altitude. Above 45 km altitude, within the sponge layer of the IFS, both IFS datasets depict lower temperatures than the observations, with the 9 km runs showing the coldest temperatures. Various sensitivity experiments conducted with the IFS are analyzed and compared with the lidar observations to investigate the impact of the different changes implemented in IFS cycle 41r2. It is found that both the scientific changes and the horizontal resolution upgrade contribute to the colder mesosphere above Sodankylä. The data assimilation seems to amplify this effect even further.