Ozone in the middle atmosphere as measured by the improved stratospheric and mesospheric sounder

Abstract

The improved stratospheric and mesospheric sounder (ISAMS) made ozone measurements in the stratosphere and mesosphere, with dense coverage over a broad range of latitudes, on about 180 days between September 1991 and July 1992. In this paper we are concerned with version 10 of the ISAMS data, which is recommended for use in the pressure range 10-0.1 mbar at night and 10-1 mbar during daytime. We describe the measurements and the retrieval algorithm, present an error analysis, and provide comparisons of the results to both climatological and coincident data. It is demonstrated that biases with respect to other well-validated data sets are less than 10%, which is well within the internal assessment of systemetic errors. aboard the Upper Atmosphere Research Satellite (UARS) in September 1991 and operated, with several interruptions, until July 1992, producing a total of about 180 days of atmospheric measurements. This paper is one of a series on the validation of data products from ISAMS. The introductory paper [Rodgers et al., this issue] contains a great deal of information which is common toall the ISAMS channels and which is necessary to understand the material presented here. This includes a general description of the instrument and its mode of operation, the concepts behind the data analysis and retrieval of ISAMS data, and an error analysis for radiometric calibration and gridding of the measured radiances onto vertical profiles. In addition, a detailed description of the ISAMS hardware can be found in the work of Taylor et al. [1993].The ISAMS ozone measurement is intended to provide a global view of the ozone distribution which would allow direct observation of the relationship of ozone changes to changes in temperature and other constituents. Ozone is somewhat different than most other constituents in that many measurements of its distribution have been made. This implies both that there is a relative wealth of correlative measurements to use for validation and that any given measurement must be Paper number 95JD01714. 0148-0227/96/95 JD-01714505.00 very accurate to be valuable. Equally, the ISAMS ozone measurement must meet a higher standard than most other ISAMS constituent measurements.In this paper we start with brief discussions of the ozone radiance measurement (section 2) and the retrieval method employed (section 3). Section 4 then presents a detailed error analysis and an interpretation of the error estimates in the archived data. Finally, section 5 presents comparisons of ozone measurements and assesses the consistency of the comparisons to the error estimates of section 4. We do not include comparisons to other UARS measurements but only to independently validated data sets. Comparisons among the UARS instruments and SAGE II may be found in the work of Cunhold et al. [this issue]. Throughout, we are concerned with data processed to "level 3" in the UARS designation. Level 3 data are those most readily available and most useful to other investigators and consist of mixing ratios interpolated onto...

Retrieval of vertical profiles of ozone from high‐resolution infrared solar spectra at Rikubetsu, Japan

“…The retrieved 03 VMR at each pressure level represents a kind of average value at that altitude. In order to compare this value with the ozonesonde data with much higher vertical resolution, the ozonesonde data were averaged using the averaging kernel for a weighting function [Rodgers, 1990;Connor et al, 1996]. Since averaging kernels were calculated for only altitude points where retrieval calculations were done, they were interpolated using a cubic spline function (see Figure 1).…”

Trajectory mapping: A tool for validation of trace gas observations

“…A number of strategies have been designed in an attempt to validate satellite data. Previous assessments of instrument accuracy have utilized comparisons of satellite observations with correlative measurements from radiosondes [Rind et al, 1993], balloons [Russell et al, 1996a;Veiga et al, 1995], rockets [Roche et al, 1996], space shuttle instruments [Kumer et al, 1996], ground-based microwave radiometer measurements [Conner et al, 1996], other satellite instruments [Cun-noMet al, 1996], and lidar [Singh et al, 1996;Bailey, et al, 1996]. In each case, correlation criteria are established to define "coincident" observations as those made within a specified distance and time of one another.…”

“…We will refer to these two approaches to data validation as "traditional" approaches. [Roche et al, 1996], space shuttle instruments [Kumer et al, 1996], ground-based microwave radiometer measurements [Conner et al, 1996], other satellite instruments [CunnoMet al, 1996], and lidar [Singh et al, 1996;Bailey, et al, 1996]. In each case, correlation criteria are established to define "coincident" observations as those made within a specified distance and time of one another.…”