Vertical structure of the lower-stratospheric moist bias in the ERA5 reanalysis and its connection to mixing processes

Abstract. Numerical weather prediction (NWP) models are known to possess a distinct moist bias in the mid-latitude lower stratosphere, which is expected affect ability accurately predict and climate. This paper investigates vertical structure of European Centre for Medium-Range Weather Forecasts (ECMWF) latest global reanalysis ERA5 using unique multi-campaign data set highly resolved water vapour profiles observed with differential absorption lidar (DIAL) on board High Altitude LOng range research aircraft (HALO). In total, 41 flights mid-latitudes from six field campaigns provide roughly 33 000 humidity varying by 4 orders magnitude. The observations cover different synoptic situations seasons thus suitable characterize strong gradients moisture upper troposphere stratosphere (UTLS). comparison indicates high positive negative deviations UT, average lead slightly (15 %–20 %). LS, rapidly increases up maximum 55 % at 1.3 km altitude above thermal tropopause (tTP) decreases again 15 altitude. Such frequently observed, although magnitude varies flight flight. layer depth increased smaller altitudes larger when low. Our results also suggest seasonality bias, summer exceeding autumn factor 3. During one campaign, collocated ozone profile enable classification tropospheric, stratospheric, mixed air vapour–ozone correlations. It revealed that while being small tropospheric stratospheric air, highlights excessive transport into LS plays decisive role formation bias. better representation mixing processes NWP could reduced that, turn, may more accurate climate forecasts. lower-stratospheric should be borne mind climatological studies data.