Stratospheric Aerosol And Gas Experiment Research Articles

Isentropic Cross-Tropopause Ozone Transport in the Northern Hemisphere

This paper investigates isentropic ozone exchange between the extratropical lower stratosphere and the subtropical upper troposphere in the Northern Hemisphere. The quantification method is based on the potential vorticity (PV) mapping of Stratospheric Aerosol and Gas Experiment (SAGE)-II ozone measurements and contour advection calculations using the NASA Goddard Space Center Data Assimilation Office (DAO) analysis for the year 1990. The magnitude of the annual isentropic stratosphere-to-troposphere ozone flux is calculated to be approximately twice the flux that is directed from the troposphere into the stratosphere. The net effect is that approx.46 x 10(exp 9) kg/yr of ozone are transferred quasi horizontally from the extratropical lower stratosphere into the subtropical upper troposphere between the isentropic surfaces of 330 and 370 K. The estimated monthly ozone fluxes show that the isentropic cross-tropopause ozone transport is stronger in summer/fall than in winter/ spring, and this seasonality is more obvious at the upper three levels (i.e., 345, 355, and 365 K) than at 335 K. The distributions of the estimated monthly ozone fluxes indicate that the isentropic stratosphere-to-troposphere ozone exchange is associated with wave breaking and occurs preferentially over the eastern Atlantic Ocean and northwest Africa in winter and over the Atlantic and Pacific Oceans in summer.

A revised water vapor product for the Stratospheric Aerosol and Gas Experiment (SAGE) II version 6.2 data set

The Stratospheric Aerosol and Gas Experiment (SAGE) II water vapor retrieval process has been updated to reflect a new understanding of the instrument performance. Primarily, this is reflected in a shifted spectral response for the primary water channel near 935 nm for the period after January 1986. In addition, the water vapor and ozone spectroscopy, aerosol clearing process, and error estimation have been updated. The end result is that the measurement bias observed in version 6.1 and earlier versions has been effectively eliminated. The sensitivity to aerosol has been reduced, so that the recommended upper limit for usable water vapor is now 3 × 10−4 km−1 in 1020‐nm aerosol extinction. The comparable value for version 6.1 was ∼2 × 10−5 km−1, or 10 times more sensitive than in the new version. The impact of the channel drift on the retrieved water vapor relative to the simple model employed in this version will be difficult to separate from geophysical change, and therefore caution is recommended in evaluating trends derived from this data set.

A global climatology of stratospheric aerosol size distribution parameters derived from SAGE II data over the period 1984–2000: 2. Reference data

In a companion paper, a global climatology was presented for different parameters characterizing the size distribution of stratospheric aerosols; the particle number density, median radius, and mode width of the particle size distribution were assumed to be lognormal. Those aerosol parameters were retrieved by optical inversion of Stratospheric Aerosol and Gas Experiment (SAGE) II aerosol extinction profiles. The data cover the time period 1984–2000. In this second part of the study, we present the climatology in a more quantitative way, and starting from the case of the Mount Pinatubo eruption in June 1991, we try to derive reference data for the three considered aerosol parameters. Using four time intervals representative of the different phases of the aerosol relaxation after a major eruption, we investigate the data set spanning the period June 1991 to mid‐1999 and present a spatiotemporal description of the aerosol distribution for the whole relaxation period. The range of the data dispersion due to seasonal effects and random variations is also estimated. Finally, a study of the error on the three parameters is presented for the whole data set, as well as an overview of our validation efforts.

A global climatology of stratospheric aerosol size distribution parameters derived from SAGE II data over the period 1984–2000: 1. Methodology and climatological observations

A global climatology has been derived for different parameters characterizing the size distribution of stratospheric aerosols; the particle number density, median radius, and mode width of the particle distribution were assumed to be lognormal. Those parameters were retrieved by applying an original optical inversion scheme on Stratospheric Aerosol and Gas Experiment (SAGE) II aerosol extinction profiles spanning from 1984 to 2000. The specificity of the inversion algorithm resides in the use of a continuity constraint that allows one to partly get rid of the ill‐posedness of the inversion problem and to greatly improve the quality and the stability of the solution. The long time period covered by the SAGE II experiment allows us to observe very different situations of volcanic load, from highly volcanic periods following major eruptions (mainly after the Pinatubo eruption of June 1991), to periods of very low aerosol load. Beyond the effects of volcanism on the aerosol parameters, other influences are observed such as seasonal effects, more particularly at midlatitude. Also the quasi‐biennial oscillation (QBO) shows a clear influence on the size distribution. In this paper, we present the general features of the different aerosol profiles, and we investigate the effects of these three main influences affecting the size distribution. We present the climatology in a more quantitative way in a companion paper, by proposing reference data for the three considered aerosol parameters.

SAGE III observations of Arctic polar stratospheric clouds – December 2002

Stratospheric Aerosol and Gas Experiment (SAGE) III aerosol extinction data from the Arctic in December 2002 are presented and used to illustrate the instrument's capability for polar stratospheric cloud (PSC) observations. Numerous PSCs were observed during the month at potential temperatures from about 450–700 K, and Types 1a and 1b PSCs could be discriminated in plots of 1022‐nm extinction versus color ratio (the ratio of 1022‐nm and 449‐nm extinctions). Type 1a (1b) PSCs were observed primarily above (below) a threshold temperature of 4 K below the nitric acid trihydrate (NAT) equilibrium temperature. Mie calculations show that some of the Type 1a PSC observations are consistent with a mixture of background aerosol and very few large NAT particles. This suggests that SAGE III offers the potential for systematic observations of these clouds, which would greatly improve our understanding of how they form and lead to stratospheric denitrification.

Molecular density retrieval and temperature climatology for 40–60 km from SAGE II

The Stratospheric Aerosol and Gas Experiment (SAGE) II is a spaceborne solar occultation instrument that makes long‐term stable measurements of atmospheric transmission at seven wavelengths between the ultraviolet and the near infrared. It provides near‐global coverage (from about 75°S to 75°N latitude) and a data record spanning over 18 years starting in late 1984. Recently, a self‐consistent molecular density retrieval was developed which uses SAGE II version 6.10 transmission data above 40 km altitude. Since at this altitude attenuation of solar radiation due to aerosols can be neglected, there is enough independent information in five SAGE II channels (386–600 nm) to separate the absorption due to ozone and nitrogen dioxide from Rayleigh scattering. Vertical inversion to obtain profiles of these quantities from 40 to 60 km proceeds as in the standard SAGE II algorithm, using a matrix of refracted path lengths. The density gradient information for averaged profiles is converted to temperature. The data are sorted into 20° latitude bands from 70°S to 70°N latitude, and the time series are fitted to provide a mean climatology and long‐term trend estimates. The climatology agrees well with published climatologies for the northern midlatitudes and includes results for the other latitude bands as well. Analysis of long‐term variation finds no statistically significant effect from the solar cycle or quasi‐biennial oscillation and no statistically significant trend. An upper bound on a possible cooling trend of 3 or 4 K/decade is derived for high and middle latitudes between 40 and 45 km altitude.

Post‐Mount Pinatubo eruption ground‐based infrared stratospheric column measurements of HNO3, NO, and NO2 and their comparison with model calculations

Infrared solar spectra recorded between July 1991 to March 1992 and November 2002 with the Fourier transform spectrometer on Kitt Peak (31.9°N latitude, 111.6°W longitude, 2.09 km altitude) have been analyzed to retrieve stratospheric columns of HNO3, NO, and NO2. The measurements cover a decade time span following the June 1991 Mount Pinatubo volcanic eruption and were recorded typically at 0.01 cm−1 spectral resolution. The measured HNO3 stratospheric column shows a 20% decline from 9.16 × 1015 molecules cm−2 from the first observation in March 1992 to 7.40 × 1015 molecules cm−2 at the start of 1996 reaching a broad minimum of 6.95 × 1015 molecules cm−2 thereafter. Normalized daytime NO and NO2 stratospheric column trends for the full post‐Pinatubo eruption time period equal (+1.56 ± 0.45)% yr−1, 1 sigma, and (+0.52 ± 0.32)% yr−1, 1 sigma, respectively. The long‐term trends are superimposed on seasonal cycles with ∼10% relative amplitudes with respect to mean values, winter maxima for HNO3 and summer maxima for NO and NO2. The measurements have been compared with two‐dimensional model calculations utilizing version 6.1 Stratospheric Aerosol and Gas Experiment (SAGE) II sulfate aerosol surface area density measurements through 1999 and extended to the end of the time series by repeating the 1999 values. The model‐calculated HNO3, NO, and NO2 stratospheric column time series agree with the measurements to within ∼8% after taking into account the vertical sensitivity of the ground‐based measurements. The consistency between the measured and model‐calculated stratospheric time series confirms the decreased impact on stratospheric reactive nitrogen chemistry of the key heterogeneous reaction that converts reactive nitrogen to its less active reservoir form as the lower‐stratospheric aerosol surface area density declined by a factor of ∼20 after the eruption maximum.

Problems in separating aerosol and cloud in the Stratospheric Aerosol and Gas Experiment (SAGE) II data set under conditions of lofted dust: Application to the Asian deserts

Stratospheric Aerosol and Gas Experiment II aerosol data obtained at two wavelengths, 525 nm and 1020 nm, have been used for some time to identify the presence of cloud along the optical path from the Sun to the satellite instrument. Examination of data obtained over desert regions in the Northern Hemisphere, particularly the Taklimakan Desert, indicates that this separation method does not always operate correctly. In regions where there is expected to be a large amount of lofted dust, unexpectedly low values of mean extinction are found, combined with higher than expected amounts of cloud. These anomalous data have been analyzed in detail, and the discrepancy is plausibly shown to be due to faulty identification of lofted dust as cloud. Six Northern Hemisphere desert regions, together with three comparison regions, have been identified for study and the anomalies used to develop a description of the seasonal and altitudinal characteristics of the lofted aerosol over these regions.

Global distribution of tropospheric ozone from satellite measurements using the empirically corrected tropospheric ozone residual technique: Identification of the regional aspects of air pollution

Abstract. Using coincident observations of total ozone from the Total Ozone Mapping Spectrometer (TOMS) and stratospheric ozone profiles from the Solar Backscattered Ultraviolet (SBUV) instruments, detailed maps of tropospheric ozone have been derived on a daily basis over a time period spanning more than two decades. The resultant climatological seasonal depictions of the tropospheric ozone residual (TOR) show much more detail than an earlier analysis that had used coincident TOMS and Stratospheric Aerosol and Gas Experiment (SAGE) ozone profiles, although there are many similarities between the TOMS/SAGE TOR and the TOMS/SBUV TOR climatologies. In particular, both TOR seasonal depictions show large enhancements in the southern tropics and subtropics in austral spring and at northern temperate latitudes during the summer. The much greater detail in this new data set clearly defines the regional aspect of tropospheric ozone pollution in northeastern India, eastern United States, eastern China, and west and southern Africa. Being able to define monthly climatologies for each year of the data record provides enough temporal resolution to illustrate significant interannual variability in some of these regions.

Validation of POAM III ozone: Comparisons with ozonesonde and satellite data

This paper describes the validation of ozone profiles from the Polar Ozone and Aerosol Measurement (POAM) III instrument. POAM III O3 is measured with 1‐km vertical resolution throughout most of the stratosphere and random errors of ∼5%. It is shown that sunspots do not significantly affect the POAM O3 retrievals, nor do polar stratospheric clouds, except under rarely encountered, extreme conditions of very low O3 and exceptionally high aerosol extinction. A statistical analysis is presented of comparisons between coincident measurements from POAM III and ozonesondes, the Halogen Occultation Experiment (HALOE) and the Stratospheric Aerosol and Gas Experiment (SAGE) II. On average, POAM III O3 profiles agree to within ±5% with these correlative data from 13 to 60 km. There is a suggestion that from 30 to 60 km POAM III sunrise data might be biased slightly low (<5%) relative to POAM sunset data. There is evidence that POAM III has a high bias of up to ∼0.1 ppmv from 10 to 12 km, and that this bias might stem, in part, from errors in the retrieval of aerosol extinction at 0.6 μm, the primary O3 absorption wavelength in the POAM retrievals. Below 10 km the POAM III data agree with coincident sonde measurements to better than 0.05 ppmv on average, which can correspond to large relative differences of more than +30% at 8–9 km and −100% at 5 km. We conclude that the POAM III profiles are highly accurate and adequate for quantitative scientific studies.

Estimation of Arctic polar vortex ozone loss during the winter of 1999–2000 using vortex‐averaged airborne differential absorption lidar ozone measurements referenced to N2O isopleths

The NASA Langley UV differential absorption lidar (DIAL) system flew on the NASA DC‐8 aircraft during the Stratospheric Aerosol and Gas Experiment (SAGE) III Ozone Loss and Validation Experiment/Third European Stratospheric Experiment on Ozone 2000 (SOLVE/THESEO 2000) mission from 30 November 1999 to 15 March 2000. The UV DIAL system measured ozone (O3) profiles at altitudes from about 1 km above the aircraft up to about 26 km with a vertical resolution of 750 m and a horizontal resolution of 70 km below 19 km and 140 km above 19 km. In comparison with electrochemical concentration cell ozonesonde profiles, the UV DIAL O3 measurements agreed to within 5% up to 20 km and 10% from 20 to 25 km. Ozone loss during the season was determined using the UV DIAL O3 data along with air mass subsidence determined using N2O as a conservative tracer at five levels from 50 to 250 ppbv [Greenblatt et al., 2002]. O3 mixing ratios were determined inside the polar vortex, away from the collar region along these five levels during the mission. The maximum O3 loss determined from 30 November to 12 March was 1.55 ± 0.3 ppmv at the 440–450 K potential temperature (Θ) level, while the loss there between 20 January and 15 March was 1.3 ± 0.3 ppmv. These results are comparable to many of the other reported losses for these periods, but lower than several. Some of the determinations of higher losses used a different method to determine descent during the season. These results indicate that a series of vertical profiles of O3 that sample much of the vortex during the winter, along with determinations of the descent of air masses inside the vortex, can give a reasonable estimate of the O3 changes during the season.

Retrieval analysis of aerosol integral properties from simulated extinction at SAGE II and HALOE wavelengths

Aerosol integral properties inferred from simulated multispectral extinction data at Stratospheric Aerosol and Gas Experiment (SAGE) II and the Halogen Occultation Experiment (HALOE) wavelengths using two different retrieval methods are analyzed and compared. The discrepancy in aerosol integral properties that results from using different retrieval techniques is analyzed, and the differences caused by sampling the same aerosol population at different wavelengths with the same retrieval technique are evaluated. We found that a better agreement was achieved when different methods were applied to the same data set as compared to a case where the same method is applied to data recorded by two different sensors. Our results demonstrate that when comparing aerosol properties derived from different instruments, the particle size information contents from optical instruments have to be taken into consideration together with the propagation of measurement errors and the systematic errors arising from the retrieval methods.

Chlorine budget and partitioning during the Stratospheric Aerosol and Gas Experiment (SAGE) III Ozone Loss and Validation Experiment (SOLVE)

The amount of chlorine in the stratosphere has a direct influence on the magnitude of chlorine‐catalyzed ozone loss. A comprehensive suite of organic source gases of chlorine in the stratosphere was measured during the NASA Stratospheric Aerosol and Gas Experiment (SAGE) III Ozone Loss and Validation Experiment (SOLVE) campaign in the arctic winter of 2000. Measurements included chlorofluorocarbons (CFCs), hydrochlorofluorocarbons (HCFCs), halon 1211, solvents, methyl chloride, N2O, and CH4. Inorganic chlorine contributions from each compound were calculated using the organic chlorine measurements, mean age of air, tropospheric trends, and a method to account for mixing in the stratosphere. Total organic chlorine measured at tropospheric levels of N2O was on the order of 3500 ppt. Total calculated inorganic chlorine at a N2O mixing ratio of 50 ppb (corresponding to a mean age of 5.5 years) was on the order of 3400 ppt. CFCs were the largest contributors to total organic chlorine (55–70%) over the measured N2O range (50–315 ppb), followed by CH3Cl (15%), solvents (5–20%), and HCFCs (5–25%). CH3Cl contribution was consistently about 15% across the organic chlorine range. Contributions to total calculated inorganic chlorine at 50 ppb N2O were 58% from CFCs, 24% from solvents, 16% from CH3Cl, and 2% from HCFCs. Updates to fractional chlorine release values for each compound relative to CFC 11 were calculated from the SOLVE measurements. An average value of 0.58 was calculated for the fractional chlorine release of CFC 11 over the 3–4 year mean age range, which was lower than the previous value of 0.80. The fractional chlorine release values for HCFCs 141b and 142b relative to CFC 11 were significantly lower than previous calculations.

N2O and NOy observations in the 1999/2000 Arctic polar vortex: Implications for transport processes in a CTM

We compare the evolution of simulated NOy and N2O with observations made during the 1999/2000 NASA Stratospheric Aerosol and Gas Experiment (SAGE) III Ozone Loss and Validation Experiment/Third European Stratospheric Experiment on Ozone‐2000 to understand the different effects of transport on these two species and evaluate model transport processes in the Arctic lower stratosphere during mid to late winter. The simulations were made with the Goddard Space Flight Center chemistry and transport model, driven by meteorological analyses from the Goddard Data Assimilation Office. In January, model NOy and N2O distributions agree well with observations, provided a substantial loss of NOy due to polar stratospheric cloud sedimentation is included in the model simulation. In late winter, model NOy is ∼2–3 times larger than observed, despite the continued reasonable agreement of N2O. Anomalously large vertical transport of NOy due to overly vigorous descent in the CTM is primarily responsible for the large late winter NOy mixing ratios. However, the rapid descent is not apparent in the evolution of modeled N2O and presumably other similar species due to compensating horizontal transport of N2O across the vortex edge, which results in the reasonable agreement between modeled and measured N2O at ER‐2 altitudes throughout the winter. The need to properly represent the seasonal evolution of NOy when studying lower stratospheric polar wintertime photochemistry places strong constraints on the meteorological data used to drive global chemistry and transport models.

Characteristics of the 1997/1998 El Niño cloud distributions from SAGE II observations

The present study examines the characteristics of cloud distributions with emphasis on cloud longwave radiative forcing (CLRF) during the peak of the 1997/1998 El Niño in relation to climatological conditions, based on measurements from the Stratospheric Aerosol and Gas Experiment (SAGE) II. The observed distinct cloud occurrence and CLRF during this unusual 1997/1998 El Niño constitutes a unique data set for validating and improving cloud‐radiation‐climate interactions in general circulation and climate models. Using the solar occultation technique, the SAGE II satellite instrument is capable of providing measurements with a 1‐km vertical resolution facilitating the analysis with sufficient vertical as well as near global scale (70°S–70°N) details. The present study indicates (1) above normal high‐altitude opaque cloud occurrence over the eastern tropical Pacific and an opposite situation over the Pacific warm pool, leading to a distribution of the cumulative opaque cloud anomalies above 3 km generally consistent with the pattern of observed tropical sea surface temperature and precipitation anomalies; (2) a similar behavior in the subvisual cloud distributions near the tropical tropopause; (3) a zonally averaged cloud distribution that is characterized by reduced opaque clouds at low latitudes, except in the southern tropics below 10 km, and by enhanced opaque clouds at high latitudes, along with increased subvisual clouds in the southern tropics and decreased subvisual clouds in the northern subtropics in the upper troposphere; and (4) a geographic distribution of model‐calculated CLRF anomalies that resembles closely that inferred from the Earth Radiation Budget Experiment and the Clouds and the Earth's Radiant Energy System. A discussion on the influence of the El Niño on large‐scale mean tropospheric circulations is also provided.

Tracer Equivalent Latitude: A Diagnostic Tool for Isentropic Transport Studies

Abstract Area equivalent latitude based on potential vorticity (PV) is a widely used diagnostic for isentropic transport in the stratosphere and upper troposphere. Here, an alternate method for calculating equivalent latitude is explored, namely, a numerical synthesis of a PV-like tracer from a long-term integration of the advection–diffusion equation on isentropic surfaces. It is found that the tracer equivalent latitude (TrEL) behaves much like the traditional PV equivalent latitude (PVEL) despite the simplified governing physics; this is evidenced by examining the kinematics of the Arctic lower stratospheric vortex. Yet in some cases TrEL performs markedly better as a coordinate for long-lived trace species such as ozone. These instances include analysis of lower stratospheric ozone during the Stratospheric Aerosol and Gas Experiment (SAGE) III Ozone Loss and Validation Experiment (SOLVE) campaign and three-dimensional reconstruction of total column ozone during November–December 1999 from fitted ozone...

A semisimultaneous inversion algorithm for SAGE III

The Stratospheric Aerosol and Gas Experiment (SAGE) III instrument was successfully launched into orbit on 10 December 2001. The planned operational species separation inversion algorithm will utilize a stepwise retrieval strategy. This paper presents an alternative, semisimultaneous species separation inversion that simultaneously retrieves all species over user‐specified vertical intervals or blocks. By overlapping these vertical blocks, retrieved species profiles over the entire vertical range of the measurements are obtained. The semisimultaneous retrieval approach provides a more straightforward method for evaluating the error coupling that occurs among the retrieved profiles due to various types of input uncertainty. Simulation results are presented to show how the semisimultaneous inversion can enhance understanding of the SAGE III retrieval process. In the future, the semisimultaneous inversion algorithm will be used to help evaluate the results and performance of the operational inversion. Compared to SAGE II, SAGE III will provide expanded and more precise spectral measurements. This alone is shown to significantly reduce the uncertainties in the retrieved ozone, nitrogen dioxide, and aerosol extinction profiles for SAGE III. Additionally, the well‐documented concern that SAGE II retrievals are biased by the level of volcanic aerosol is greatly alleviated for SAGE III.

Stratospheric Aerosol and Gas Experiment (SAGE) III measurements: A study on the retrieval of ozone, nitrogen dioxide, and aerosol extinction coefficients

This paper presents the inversion algorithms that we propose to apply to the Stratospheric Aerosol and Gas Experiment (SAGE) III measurements for retrieving the aerosol and gas components of the Earth atmosphere. SAGE III is an upgraded version of SAGE II: The main improvements are the larger number of channels, including several resolved channels and a channel centered near 1.55 μm, and a higher vertical resolution. The species retrieval from the transmission measurements can be achieved via the inversion algorithm that we developed at Laboratoire d'Optique Atmosphérique for SAGE II with only minor changes, in which all the channels are handled altogether for species separation by using an iterative method. For SAGE III, ozone and nitrogen dioxide can also be retrieved independently from two specific resolved channels by using their spectral signature over a small wavelength range; this second inversion leads only to the ozone and nitrogen dioxide concentrations and to the aerosols at two wavelengths. Since these later O3 and NO2 retrievals are shown to be of better quality, it is further possible to use them in the first algorithm, in a combined analysis, for aerosol separation in the other channels. The performances of these three inversions have been compared for various atmospheric models, accounting for some probable measurement errors. The three methods perform very well in the case of nominal data. When errors are added to the measurements, the combined analysis leads generally to better retrieval for O3 and NO2, while there is almost no improvement for the aerosols. For random measurement error, the retrieved profiles exhibit no bias with respect to the input profiles, except at high altitude. The accuracy of the results is shown to be dependent on the species and on the atmospheric model.

A discussion of upper stratospheric ozone asymmetries and SAGE trends

A chemical box model constrained by Upper Atmosphere Research Satellite (UARS) measurements of CH4, H2O, NOx, and temperature in 1992–1994 has been used in an attempt to simulate ozone trend measurements from the Stratospheric Aerosol and Gas Experiment (SAGE) at 43 km, 45° latitude and their hemispheric differences. The model is successful in simulating the seasonal cycle in ozone mixing ratios observed by the Microwave Limb Sounder (MLS) and the Solar Backscattered Ultraviolet (SBUV) instrument including the 15% larger values in the Southern Hemisphere at the wintertime ozone maximum. The model indicates that this hemispheric asymmetry is associated with hemispheric temperature differences of approximately 10°K; in contrast, hemispheric differences in methane and the resulting effects on ClOx make only a small contribution to this ozone asymmetry. The model predicts that, based on observed Cly increases, the ozone decrease in the Southern Hemisphere should have exceeded that in the Northern Hemisphere by approximately 1%/decade. This is consistent with an observed ozone hemispheric trend difference of 1.7 ± 2.1%/decade (1σ) for 1979–1997 and −0.8 ± 2.8%/decade for 1985–1997 from SAGE I and SAGE II (version 6.1) measurements. These trend differences are caused by hemispheric differences in temperature in early winter and in CH4 in late winter. The model, based on changes in Cly but not in other specified parameters, calculates a larger downtrend of ozone than is observed by SAGE II in 1985–1997 by approximately 3.1 ± 1.4%/decade (1σ), but it qualitatively simulates the observed ozone trend variations over the period 1979–1997. The amplitude of the model calculated downtrend would be reduced by approximately 1%/decade for a methane increase of 1%/yr. Observed temperature decreases of approximately 0.1°K/yr would have decreased the magnitudes of the calculated ozone downtrends on pressure surfaces by approximately 1%/decade but could have affected SAGE trends by a small amount in the opposite direction because of associated geopotential height changes.

Assessment of SAGE version 6.1 ozone data quality

The Stratospheric Aerosol and Gas Experiment (SAGE) II V6.1 ozone retrievals are shown to be of better precision at all levels and to be much more accurate than previous retrievals in the lower stratosphere below 20 km altitude. A filtering procedure for removing anomalous ozone profiles associated with volcanic aerosol/cloud effects and other identified artifacts in V6.1 ozone is described. The agreement between SAGE and ozonesondes in the mean is shown to be approximately 10% down to the tropopause. Relative to the sondes, SAGE tends to slightly overestimate ozone (less than 5%) between 15 and 20 km altitude and systematically underestimates ozone in the troposphere by approximately 30% in the regions between 8 km altitude and 2 km below the tropopause. The precisions (random errors) of SAGE ozone retrievals above 25 km altitude are estimated to be 4% or better; they are a factor of 10 worse below 16 km altitude. Linear trends in the differences between coincident SAGE and ozonesondes measurement are generally less than 0.3%/yr and not significantly different from zero in 95% confidence intervals. Compared to V5.96 retrievals, ozone trend differences between 20 and 50 km altitude are approximately 0.1%/yr; below 20 km altitude the SAGE II trends are more positive by approximately 0.2%/yr. For the 1984–1999 period, the SAGE II shows a localized ozone loss of −0.4 ± 0.25%/yr (2σ) in the tropics at 20 km altitude. In the lower stratosphere, between 16 and 22 km altitudes, the SAGE shows significant ozone losses in the midlatitudes in both hemispheres during the 1979–1999 periods. The ozone trends range from −0.24 ± 0.18%/yr to −0.77 ± 0.46%/yr (2σ). However, in the 1984–1999 period, the downward trends are smaller (−0.07%/yr to −0.25%/yr) in this altitude range, and the trends in the integrated column from 12 to 17 km altitude in midlatitudes (35°–60°) are not significantly different from zero (0.1 ± 0.6%/yr (2σ)). Averaged over the tropics (20°S–20°N), the ozone column above 15 km altitude exhibit a trend of −0.12 ± 0.08%/yr (2σ).