Assessment of AIRS Version 7 Temperature Profiles and Low-Level Inversions with GRUAN Radiosonde Observations in the Arctic

The extensive global climate observing system (GCOS) reference upper-air network (GRUAN) datasets provide a chance to validate newly released Atmospheric Infrared Sounder (AIRS) version 7 (v7) products over the Arctic. This manuscript reports on analysis performed evaluate errors from AIRS 6 (v6) and v7 temperature profiles characterize derived low-level inversion (LLI) representativeness in Arctic region. averaging kernel, representing measurement sensitivity, is applied reduce vertical resolution of radiosonde for comparison. Due improved retrieval algorithms, produces smaller biases troposphere suppresses cold bias v6. Nevertheless, profile-averaged root mean square error (RMSE) increased by 30% v7, particularly winter half-year when showed larger RMSE below 800 hPa. accuracy primarily sensitive surface type cloud fraction. Compared v6, has less frozen land sea ice different fraction conditions. However, RMSEs are more effective (ECF) highly influenced significant contribution nonfrozen samples. kernel-averaged profiles, than 80% v6 accurately detect LLIs. discreteness AIRS’s predefined pressure level results consistent with radiosondes only 65% time LLI depth calculation. In contrast, can obtain intensity relatively high correlation (>0.9). With boundary layer further improved, it potential be used as an independent detector