Comparison of STOIC 1989 ground‐based lidar, microwave spectrometer, and Dobson spectrophotometer Umkehr ozone profiles with ozone profiles from balloon‐borne electrochemical concentration cell ozonesondes

Abstract

Ground‐based measurements of stratospheric ozone using a Jet Propulsion Laboratory (JPL) lidar, a NASA Goddard Space Flight Center (GSFC) lidar, a Millitech Corporation/NASA Langley Research Center (Millitech/LaRC) microwave spectrometer, and a NOAA Dobson ozone spectrophotometer were compared with in situ measurements made quasi‐simultaneously with balloon‐borne electrochemical concentration cell (ECC) ozonesondes during 10 days of the Stratospheric Ozone Intercomparison Campaign (STOIC). The campaign was conducted at Table Mountain Observatory, California, during the summer of 1989. ECC ozonesondes were flown by NOAA Climate Monitoring and Diagnostics Laboratory (CMDL) personnel as well as by personnel from the NASA Wallops Island Flight Facility (WFF). Within the altitude range of 20–32 km, ozone measurement precisions were estimated to be ±0.6 to ±1.2% for the JPL lidar, ±0.7% for the GSFC lidar, ±4% for the microwave spectrometer, and ±3% for the NOAA ECC ozonesonde instruments. These precisions decreased in the 32 to 38.6‐km altitude range to ±1.3, ±1.5, and ±3% to ±10% for the JPL lidar, GSFC lidar, and the ECC sondes, respectively, but remained at ±4% for the microwave instrument. Ozone measurement accuracies in the 20 to 32 km altitude range were estimated to be ±1.2 to ±2.4% for the JPL lidar, ±1.4% for the GSFC lidar, ±6% for the microwave radiometer, and ±5% for the ECC ozonesondes. The accuracies decreased in the 32 to 38.6‐km altitude range to ±2.6, ±3.0, ±7, and 1 ± 4% to −4 ± 10% for the JPL lidar, the GSFC lidar, the microwave spectrometer, and the ECC ozonesondes, respectively. While accuracy estimates for the ECC sondes were obtained by combining random and estimated bias errors, the accuracies for the lidar instruments were obtained by doubling the measurement precision figures, with the assumption that such doubling accounts for systematic errors. Within the altitude range of 20–36 km the mean ozone profiles produced by the JPL, GSFC, and the Millitech/LaRC groups did not differ from the mean ECC sonde ozone profile by more than about 2, 4, and 5%, respectively. Six morning Dobson instrument Umkehr observations yielded mean ozone amounts in layers 3 and 5–7 that agreed with comparison ECC ozonesonde data to within ±4%. In layer 4 the difference was 7.8%. (Less favorable comparison data were obtained for six afternoon Umkehr observations made in highly polluted near‐surface air.) This good agreement in overall results obtained lends credence to the reliability of the ozone measurements made at Table Mountain Observatory during STOIC 1989.