Articles | Volume 5, issue 5
10 Jun 2005
10 Jun 2005

Evaluation of SHADOZ sondes, HALOE and SAGE II ozone profiles at the tropics from SAOZ UV-Vis remote measurements onboard long duration balloons

F. Borchi, J.-P. Pommereau, A. Garnier, and M. Pinharanda

Abstract. Ozone profiles from 10 to 26km have been obtained at almost constant latitude (20 ± 5° S) in the tropics using SAOZ UV-vis spectrometers flown onboard long duration balloons in 2001 and 2003. The precision of the measurements is estimate to be better than 2% in the stratosphere (3.5% accuracy) and 5-6% in the troposphere (12% and 25% accuracy at 15km and 10km respectively) with an altitude uncertainty of -30 ± 25m. The variability of ozone concentration along a latitudinal circle at 20° S in the SH summer is found smaller than 3-4% above 20km, but increasing rapidly below in the Tropical Tropopause Layer (TTL). The high correlation between PV and ozone suggests that most of this variability can be attributed to quasi-horizontal exchange with the mid-latitude stratosphere.

The performances of the SHADOZ ozonesonde network, HALOE and SAGE II in the tropics have been studied by comparison with SAOZ measurements. In the stratosphere, the main discrepancies arise from differences in altitude registration, particularly sensitive between 20 and 26km in the tropics because of the strong gradient of ozone concentration. In the upper troposphere, the SAOZ measurements are consistent with those of the sondes and the lidar in cloud free conditions, but biased high by 60% on average compared to ozonesondes over the Western Pacific, at American Samoa and Fiji. The likely explanation is the frequent occurrence of near zero ozone layers in the convective clouds of the South Pacific Convergence Zone which cannot be seen by SAOZ as well as all ground-based and space borne remote sensing instruments. Compared to SAOZ, SAGE II displays a 50-60% low bias similar to that already known with the ozonesondes, and a larger zonal variability. However, the significant correlation with PV suggests that useful information on tropospheric ozone could be derived from SAGE II. Finally, the unrealistic large offsets and variability in the HALOE data compared to all others, indicates that the measurements of this instrument are of limited use below 17km.

Final-revised paper