Verification of the AIRS and MLS ozone algorithms based on retrieved daytime and nighttime ozone

. Ozone (O 3 ) plays a significant role in weather and climate on regional to global spatial scales. Most studies on the variability in the total column of O 3 (TCO) are typically analysed using daytime data. Based on knowledge of the chemistry and transport of O 3 , significant deviations between daytime and nighttime O 3 are only expected either in the planetary boundary 10 layer (PBL) or high in the stratosphere or mesosphere, having little effect on the TCO. Hence, we expect the daytime and nighttime TCO to be very similar. However, a detailed evaluation of satellite measurements of daytime and nighttime TCO is still lacking, despite the existence of long term records of both. Comparing daytime and nighttime TCOs thus provides a novel approach to verify the retrieval algorithms of for example the Atmospheric InfraRed Sounder (AIRS) and the Microwave Limb Sounder (MLS). In addition, such a comparison also helps in assessing the value of nighttime TCO for scientific research. 15 Applying this verification on the AIRS and the MLS data we identified inconsistencies in observations of O 3 from both satellite instruments. For AIRS, daytime-nighttime differences were found over oceans resembling cloud cover patterns, and over land, mostly over dry land areas, likely related to infrared surface emissivity. These differences point to issues with the representation of both processes in the AIRS retrieval algorithm. For MLS, a major issue was identified with the “ascending-descending” orbit flag, used to discriminate nighttime and daytime MLS measurements. Disregarding this issue, MLS day-night differences 20 were significantly smaller than AIRS day-night differences, providing additional support for retrieval method origin of AIRS day-night TCO differences. MLS day-night differences are dominated by the upper stratospheric and mesospheric diurnal O 3 cycle. These results provide useful information for improving infrared O 3 products and at the same time will allow study the day-night differences of stratospheric and mesospheric O 3

In response to this concern and associated environmental policies, during the last two decades a large number of studies 30 have focused on estimating long-term variations and trends in stratospheric column of O3 (SCO). A summary of the state of the science is frequently reported in the quadrennial O3 Assessment Reports issued by the United Nations Environmental Program (UNEP) and the World Meteorological Organization (WMO). These reports are written in response to the global treaties aiming at minimizing emissions of ODSs. The signatories of these treaties ask for regular updates on the state of the science and knowledge. The most recent O3 Assessment reports extensively discuss long-term variations and trends in 35 stratospheric O3 in relation to expected recovery (WMO, 2011(WMO, , 2014(WMO, , 2018. According to WMO (2018),Antarctic stratospheric O3 has started to recover, while outside of the polar regions, upper stratospheric O3 has also increased. On the other hand, no significant trend has been detected in global (60°S-60°N) total column O3 over the 1997-2016 period with average values for the years since the last Assessment remaining roughly 2% below the 1964-1980 average. Moreover, recently a debate has emerged over the question as to whether lower stratospheric O3 between 60°S-60°N has continued to decline 40 despite decreasing O3 depleting substances (Ball et al., 2018;Ball et al., 2019). In addition to the quadrennial O3 Assessments, the Bulletin of the American Meteorological Society annually publishes its "State of the Climate", which since 2015 includes a description of the relevant stratospheric events of the past year, the state of the Antarctic O3 hole, as well as an annual update of global and zonal trends in stratospheric O3. These regularly recurring reports and publications illustrate the continued attention and monitoring of the O3 layer and its recovery, in which the long records of satellite observations play a crucial role. 45 Establishing and maintaining the quality of the satellite observations of stratospheric O3 is therefore highly relevant.
A variety of techniques exist to measure the O3 column and stratospheric O3. The UV absorption spectroscopy with the sun or stars as sources of UV light is the most used to derive O3 (Weeks et al., 1978;Fussen et al., 2000). In addition to the UV occultation, the absorption of infrared radiation has also been used to detect O3 profiles throughout the column (Gunson et al., 1990;Brühl et al., 1996). Another technique is the detection of the molecular oxygen dayglow emissions (Mlynczak and 50 Drayson, 1990;Marsh et al., 2002). Some ground-based instruments use O3 emissions in the microwave region to infer the O3 density in the mesosphere (Zommerfelds et al., 1989;Connor et al., 1994). Infrared emission measurements overcome the limitations in the local time coverage of solar occultation and dayglow technique and their altitude resolution is significantly higher compared with microwave measurements (Kaufmann et al., 2003). The strongest O3 infrared absorption centers near 9.6 um. 55 Based on knowledge of chemistry and transport of O3, significant deviations between daytime and nighttime O3 are only expected either in the planetary boundary layer (PBL) and high in the stratosphere or mesosphere, having little effect on the total column of O3 (TCO). Hence, we expect that the daytime and nighttime TCO to be very similar. Day-night intercomparisons present a unique opportunity to assess the internal consistency of infrared O3 instruments (Brühl et al., 1996;Parrish et al., 2014). Temperature effects within satellite instruments, calibration procedures between day and night or 60 inversion algorithms could potentially result in systematic differences between TCO measurements from different satellites.
The Stratosphere Aerosol and Gas Experiment (SAGE) applied day-night differences to validate O3 profiles and found daytime data have a low bias due to the retrieval method since the magnitude of the difference was much less in a photochemical model https: //doi.org/10.5194/amt-2020-194 Preprint. Discussion started: 2 July 2020 c Author(s) 2020. CC BY 4.0 License. (Cunnold et al., 1989). There are infrared satellite instruments, like Atmospheric InfraRed Sounder (AIRS), and The Microwave Limb Sounder (MLS), that provide global daytime and nighttime TCO/SCO and O3 profile. Although their daytime 65 O3 retrievals have been validated (Livesey et al., 2008;Sitnov and Mokhov, 2016), day-night differences in TCO and SCO are still largely unexplored.
The O3 diurnal cycle depends on latitude, weather and time. The variations of the diurnal cycle are less than 5% in the tropics and subtropics and increasing to more than 15% near the polar day terminator in the upper stratosphere (Frith et al., 2020). There exist diurnal variations in atmospheric O3 at certain altitudes. There are two distinct O3 maxima in the typical 70 vertical profile of the O3 volume mixing ratio, one in the stratosphere and one in the mesosphere. The secondary maximum in the mesosphere is present during both day and night (Evans and Llewellyn, 1972;Hays and Roble, 1973). Chapman (1930) revealed the photochemical scheme in the mesosphere. The reactions of Chapman cycle are important for us to understand diurnal O3 variation.
The anticorrelation of O3 and temperature is mainly due to the temperature dependence of the chemical rate coefficients (Craig 80 and Ohring, 1958;Barnett et al., 1975). Huang et al. (2008) and Huang et al. (1997) found midnight O3 increases in the mesosphere, based on SABER and MLS data, respectively. Zommerfelds et al. (1989) surmised that eddy transport may explain this increase, while Connor et al. (1994) stated that atmospheric tides are expected to cause systematic day-night variations.
During daytime, photolysis is the major loss process. The main nighttime O3 source in the mesosphere is atomic oxygen, 85 while its sinks are atomic hydrogen and atomic oxygen (Smith and Marsh, 2005). In addition to chemical reactions with active hydrogen and molecular, the turbulent mass transport also plays an important role in the explanation of the secondary O3 maximum (Sakazaki et al., 2013;Schanz et al., 2014).
Tropospheric O3 is mainly produced during chemical reactions when mixtures of organic precursors (CH4 and nonmethane volatile organic carbon, NMVOC), CO, and nitrogen oxides (or NOx), are exposed to the UV radiation in the 90 troposphere (Simpson et al., 2014). At night, in the absence of the sunlight, there is no O3 production, but surface O3 deposition and dark reactions transform the NOx-VOC mixture and remove O3. The dark chemistry affects O3 and its key ingredients mainly depend on the reactions of two nocturnal nitrogen oxides, NO3 (the nitrate radical) and N2O5 (dinitrogen pentoxide).
NO3 oxidizes VOC at night, while reaction of N2O5 with aerosol particles containing water removes NOx. Both processes remove O3 as well at night (Brown et al., 2006).
The diurnal cycle of O3 in the middle stratosphere had generally been considered small enough to be inconsequential, with known larger variations in the upper stratosphere and mesosphere (Prather, 1981;Pallister and Tuck, 1983). Later studies have highlighted observed and modelled peak-to-peak variations of the order of 5% or more in the middle stratosphere between 30 and 1 hPa (Sakazaki et al., 2013;Parrish et al., 2014;Schanz et al., 2014).
In terms of dynamics, vertical transport due to atmospheric tides is expected to contribute to diurnal O3 variations at 100 altitudes where background O3 levels have a sharp vertical gradient (Sakazaki et al., 2013). The Brewer/Dobson circulation transports air upwards in the tropics, polewards and downwards at high latitudes, with stronger transport towards the winter pole (Chipperfield et al., 2017).
The main objective of this paper is to analyse day-night differences in the AIRS TCO and the MLS SCO, as well as in MLS upper atmospheric O3 profiles. Section 2 discusses the data used. Section 3 presents results for AIRS, MLS, the 105 comparison of AIRS with MLS, and an application of AIRS TCO data over the Pacific low O3 regions to highlight how daynight differences affect use and interpretation of TCO data. Finally, section 4 ends the paper with a brief summary and conclusions.

AIRS TCO retrievals 110
The AIRS satellite instrument was the first in a new generation of high spectral resolution infrared sounder instruments flown aboard the National Aeronautics and Space Administration (NASA) Earth Observing System (EOS) Aqua satellite (Divakarla et al., 2008). The AIRS radiance data at 9.6 µm band are used to retrieve column O3 and O3 profiles during both day and night (including the polar night) (Pittman et al., 2009;Fu et al., 2018). The AIRS V6 level 3 standard TCO products (2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015)(2016)(2017)(2018) comprising daily averaged measurements on the ascending and descending branches of an orbit with the quality 115 indicators 'best' and 'good' and binned into 1°×1° (latitude × longitude) grid cells are used here. Outside of the polar zones (60°N-90°N and 90°S-60°S), ascending and descending correspond respectively to daytime (13:30 in local solar time) and nighttime (01:30 in local solar time). Hereafter we refer to "day" and "night" rather than ascending and descending over 60°S-60°N. In the polar zones, it is inappropriate to use ascending/descending mode to define daytime/nighttime, therefore, we just compare differences between ascending and descending mode. AIRS TCO measurements agree well with the global 120 Brewer/Dobson Network station measurements with a bias of less than 4% and a root mean squared error (RMSE) difference of approximately 8% (Divakarla et al., 2008). Analysis of AIRS TCO monthly maps revealed that its retrievals depict seasonal trends and patterns in concurrence with OMI and SBUV/2 observations (Divakarla et al., 2008) (Sitnov and Mokhov, 2016).

MLS SCO and O3 profile retrievals
The MLS instrument on-board Aura satellite, which was launched on 15 July 2004 and placed into a near-polar Earth orbit at 705 km with an inclination of 98°, uses the microwave limb sounding technique to measure vertical profiles of chemical 130 constituents and dynamical tracers between the upper troposphere and the lower mesosphere (Waters et al., 2006). Its orbital ascending mode is at 13:42 (local solar time) and the orbital descending mode at 01:42 (local solar time) over 60°S-60°N. In this study, we use the MLS v4.2x standard O3 product (2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015)(2016)(2017)(2018). Its retrieval is using 240-GHz radiance, providing nearglobal spatial coverage (82°S to 82°N latitude), with each profile spaced 1.5 degrees or ~165 km along the orbit track. This O3 product includes the O3 profile on 55 pressure surfaces and the recommended useful vertical range is from 261 to 0.02 hPa. In 135 addition, it contains an O3 column, which is the integrated stratospheric column down to the thermal tropopause calculated from MLS measured temperature (Livesey et al., 2015). Jiang et al. (2007) found the MLS stratospheric O3 data between 120 and 3 hPa agreed well with ozonesonde measurements, within 8% for the global daily average. Froidevaux et al. (2008) reported MLS stratospheric O3 uncertainties of the order of 5%, with values closer to 10% (and occasionally 20%) at the lowest stratospheric altitudes, where small positive biases are found. Livesey et al. (2008) found MLS O3 accuracy was estimated at 140 ~40 ppbv or +5% (~20 ppbv or +20% at 215 hPa). Comparisons with expectations and other observations show good agreements for the MLS O3 product, generally consistent with the systematic errors quoted above. large day-night temperature differences. The same phenomenon is observed in Western Australia at summer-time.

AIRS TCO day-night differences
In Figure 1e shows for the annual mean large differences of AIRS TCO retrievals over deserts, the Intertropical Convergence Zone (ITCZ) with persistent clouds and Arctic regions. These are regions with atypical earth surface properties or oceanic regions with persistent cloud cover. The spatial patterns over land mimic regions with low IR surface emissivity and/or regions where IR surface emissivity exhibits large seasonal variations (Feltz et al., 2018). Figure 1f shows significant 155 TCO changes at the land-ocean interface. All these effects are important parameters for the retrieval algorithm but bear no physical relation with total O3. Hence, the differences shown in Figure 1 provide strong indications that the largest AIRS day-https://doi.org/10.5194/amt-2020-194 Preprint. Discussion started: 2 July 2020 c Author(s) 2020. CC BY 4.0 License. night TCO differences are dominated by retrieval artefacts. As such changes are unphysical, it confirms the hypothesis that clouds and the surface type (land/desert/vegetation/snow or ice) affects the AIRS TCO retrievals.
The AIRS emissivity retrieval uses the NOAA regression emissivity product as a first guess over land. The NOAA 160 approach is based on clear radiances simulated from the European Centre for Medium-Range Weather Forecasts (ECMWF) forecast and a surface emissivity training data set (Goldberg et al., 2003). The training data set used for the AIRS V4 algorithm has a limited number of soil, ice, and snow types and very little emissivity variability in the training ensemble. In the AIRS V5 version, the regression coefficient set has been upgraded using a number of published emissivity spectra (12 spectra for ice/snow, 14 for land) blended randomly for land and ice . These improvements generated a better emissivity 165 first guess for use with the AIRS V5, and improved retrievals over the desert regions (Divakarla et al., 2008). In AIRS V6, a surface climatology was constructed from the 2008 monthly MODIS MYD11C3 emissivity product, and extended to the AIRS IR frequency hinge points using the baseline-fit approach described by Seemann et al. (2008). Nevertheless, using of day-night differences for evaluation of the AIRS V6 O3 product suggest that further refinements for better surface emissivity retrievals are required and cloud covers is another problem that needs to be solved. 170

MLS O3 profile day-night differences
In order to better understand day-night differences in TCO, we also study day-night changes in the vertical profile of O3 using MLS O3 profile measurements. Figure 2a shows that the global (60°S-60°N) differences between day and night MLS O3 profile occur in the mesosphere (10 hPa -0.1 hPa). The O3 mixing ratios are about an order of magnitude larger during night 175 in the mesosphere, which was revealed by Huang et al. (2008) previously. Different latitude bands (30 degree) between 60°S and 60°N all display similar results.
We also find an unexpected polar bias at high latitudes in Figure 2d and 2g. On the one hand, the larger differences between ascending and descending MLS O3 profile at high latitude extend from the stratosphere to the mesosphere. On the other hand, ascending O3 is smaller than that at descending at 10 hPa over 60°N-90°N in Figure 2d, which is in contrast with 180 the result of other latitude bands.

MLS O3 retrievals in 90°S-60°S and 60°N-90°N
The MLS O3 profile polar bias mentioned above turns out to be related to an inconsistency in the 'AscDescMode' flag of MLS v4.2x standard O3 product in 90°S-60°S and 60°N-90°N. When this flag has a value of plus one or minus, it means an ascending or descending observation mode. We counted the daily number of pixels at both poles for which the flag has a value 185 of plus one or minus one. Figures 3a and 3c For 60°N-90°N, ascending mode O3 also becomes larger than descending mode O3 at 10 hPa in Figure 4b. This indicates that the MLS 'AscDescMode' flag is correct for 2016-2018. 195 The O3 retrieval algorithm adopted by the MLS v2.2 products has been validated to be highly accurate using multiple correlative measurements and the data have been used widely (Jiang et al., 2007;Froidevaux et al., 2008). The MLS v3.3 and v3.4, O3 profile was reported on a finer vertical grid and the bottom pressure level with scientifically reliable values increases from 215 to 261 hPa (Livesey et al., 2015). The latest MLS v4.2x O3 profile used in this study, released in February 2015, were in general similar to the previous version. One of the major improvements of MLS v4.2x was the handling of contamination 200 from cloud signals in trace gas retrievals that resulted in significant reduction in the number of spurious MLS profile in cloudy regions and a more efficient screening of cloud-contaminated measurements. Furthermore, the MLS O3 products have been improved through additional retrieval phases and reduction in interferences from other species (Livesey et al., 2015). We find no indications that changes in instrument or algorithm are responsible for this 'AscDescMode' flag inconsistency. This flag inconsistency is not present between 60°S and 60°N. 205

Day-night difference of equatorial Pacific low O3 regions
Generally, the Pacific low O3 region (TCO < 220 DU) exist all year round and its size is larger at night than during the day, unlike the seasonal O3 hole which occurs over Antarctica during the Southern Hemisphere polar winter. On the one hand, there are limited direct NOx emissions that is why O3 low over oceans compared to land. On the other hand, the low O3 over 220 https://doi.org/10.5194/amt-2020-194 Preprint. Discussion started: 2 July 2020 c Author(s) 2020. CC BY 4.0 License. the tropical western Pacific can be attributed to tropospheric O3 loss in this area. Its presence is related to a pronounced minimum in the tropospheric column of O3 over the west Pacific, which exists due to efficient loss of photochemical mechanism with higher air temperatures and higher water concentrations for O3. In addition, high sea surface temperatures also favour strong convective activity in the tropical West Pacific, which can lead to low O3 mixing ratios in the convective outflow regions in the upper troposphere in spite of the increased lifetime of odd oxygen (Kley et al., 1996;Rex et al., 2014). 225 A further reduction in the tropospheric O3 burden through bromine and iodine emitted from open-ocean marine sources has been postulated by numerical models (Vogt et al., 1999;von Glasow et al., 2002;von Glasow et al., 2004;Yang et al., 2005) and observations (Read et al., 2008). Figure 6a and 6c show the low O3 region is mainly located over the western Pacific by AIRS. Rajab et al. (2013) investigated similar low TCO in Malaysia using AIRS data. They found the highest O3 concentration occurred in April and 230 May and the lowest O3 concentration occurred during November and December, which is consistent with our result in Figure   6f. They also found that O3 concentrations exhibited an inverse relationship with rainfall, but was positively correlated with temperature. MLS results show that the daytime low O3 region also exists mainly in tropical western Pacific.
However, we find the occurrence frequency and intensity of low O3 regions is higher at night by AIRS TCO and MLS SCO retrievals. Especially for MLS, the low O3 regions appear in large areas at night besides in tropical western Pacific. For 235 AIRS, clouds over oceans may have greater impact on the AIRS TCO retrievals at night. For MLS, more active chemical reactions may occur in these low O3 regions at night.
For past, current and future monitoring of atmospheric phenomena like the Pacific tropospheric low O3 area, it is important that observations are sufficient accurate. The evaluation of day-night differences in both MLS and AIRS has revealed the existence of biases in the satellite data that are sufficiently large in comparison to expected variations and changes in 240 atmospheric O3 that they may hamper the use of these satellite data studying them.

Conclusions
Comparison of daytime and nighttime AIRS TCO has revealed small but not insignificant biases in AIRS TCO. The differences are likely related to surface type (land/desert/vegetation/snow or ice) and infrared surface emissivity, especially over regions that exhibit smaller infrared emissivity or large seasonal variability in infrared emissivity. Differences typically 245 were of the order of a few percent, which is significant given that long term changes in TCOs related to anthropogenic emissions of stratospheric O3 depleting substances outside of polar regions are also of the order of a few percent.
There were major changes to the surface emissivity retrieval in AIRS V6 compared to previous versions resulting in a very significant improvement in yield and accuracy for surface temperature and emissivity over land and ice surfaces compared to previous versions. Nevertheless, our results indicate that the AIRS V6 TCO still can be further improved. In addition, AIRS 250 TCO differences over oceans bear a clear cloud cover signature which is likely related to uncertainties in the representation of https://doi.org/10.5194/amt-2020-194 Preprint. Discussion started: 2 July 2020 c Author(s) 2020. CC BY 4.0 License. clouds in the retrieval algorithm. The latter may also impact AIRS TCO retrievals over land, although detection of cloud features in AIRS TCO day-night differences is difficult due to the presence of the land surface emissivity related bias.
The MLS v4.2x was very useful for verification of daytime and nighttime SCO and O3 profile between 60°S-60°N. MLS day-night differences in SCO and O3 profiles show that day-night differences are only small (< 1 DU for the upper atmospheric 255 SCO), expect in the upper stratosphere and mesosphere. However, an inconsistency was found in the 'AscDescMode' flag in 60°N-90°N and in 90°S-60°S, resulting in inconsistent profiles in these regions before 14 May 2015. In processor version v4.22 and later versions this issue has been fixed, but since it is a relatively small issue, the MLS data set before 2016 has not been reprocessed.
Comparison of AIRS TCO and MLS SCO in 60°S-60°N for 2005-2018 showed the values of MLS SCO were lower than 260 AIRS TCO because the MLS SCO was based on the stratosphere only. MLS SCO day-night difference in the stratosphere (0.03 DU) and in the mesosphere (0.79 DU) was much smaller compared with AIRS TCO day-night difference (4.89 DU). As shown in Smith et al. (2015) the lifetime of O3 due to chemistry is strongly altitude dependent. Only in the mesosphere the timescale becomes low enough to see significant differences between average daytime and nighttime concentrations. Figures S1 to S4 indicate that AIRS TCO retrieval artefacts dominate the day/night variability of tropospheric O3 residuals (TOR = 265 AIRS TCO -MLS SCO) and the relatively small day-night differences of tropospheric O3 are hard to discriminate comparing day/night TCO.
We found that the frequency and intensity of low O3 regions between 60°S and 60°N was higher at night by AIRS and MLS. The daytime low O3 in tropical western Pacific was investigated, including its extent and causes. In order to clarify whether the more serious low O3 regions at night are due to the problem of the algorithm itself or the atmospheric physical and 270 chemical factors different from that in the daytime, we compared both MLS and AIRS at day and at night. It is necessary to verify day-night differences by infrared TCO observations for retrieval aspect first. Our results show that maintaining the quality of the satellite observations of stratospheric O3 is therefore highly relevant.

Data availability
Satellite data sets used in this research can be requested from public sources. AIRS total ozone column data are available 275 online (https://giovanni.gsfc.nasa.gov/giovanni/). The MLS Level 2 data can be obtained from the NASA Goddard Space