Next Article in Journal
Evaluating Scale Transferability and Observation-Based Calibration in High-Resolution PM2.5 Downscaling
Previous Article in Journal
Process-Informed Satellite-Ground Fusion for Coastal Compound Humid-Heat and Photochemical Oxidant Early Warning
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Remote Sensing
Volume 18
Issue 17
10.3390/rs18172875
remotesensing-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Altitude and Geographic Sensitivity Characteristics of the AIRS Satellite Spectrometer and Drift Correction Using Methane (CH4) Data

by
Eugenia Fedorova
1,
Vadim Rakitin
1,
Andrey Skorokhod
1,2,*,
Natalia Kirillova
1,
Andrey Belov
1,
Natalia Pankratova
1,
Yusheng Shi
3,
Lin Wang
1,4 and
Vladimir Semenov
1,5
1
A.M. Obukhov Institute of Atmospheric Physics, Russian Academy of Sciences, Moscow 119017, Russia
2
Department of Meteorology and Geophysics, University of Vienna, 1090 Vienna, Austria
3
Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, China
4
Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing 100029, China
5
Institute of Geography, Russian Academy of Sciences, Moscow 119017, Russia
*
Author to whom correspondence should be addressed.
Remote Sens. 2026, 18(17), 2875; https://doi.org/10.3390/rs18172875
Submission received: 7 July 2026 / Revised: 18 August 2026 / Accepted: 21 August 2026 / Published: 25 August 2026
(This article belongs to the Special Issue Integrated Remote Sensing for Global and Regional Atmospheric Environment Monitoring and Assessment)
Browse Figures
Versions Notes

Highlights

What are the main findings?
  • The vertical sensitivity of the AIRS satellite spectrometer was assessed for initial CH4 concentration data across diverse geographic regions.
  • Heterogeneity in AIRS spectrometer drift was identified for methane concentration data at different pressure levels in the v6/v7 AIRS CH4 VMR Standard L3 product, and the pressure levels with maximum drift were determined.
  • Based on drift coefficients calculated for multiple pressure levels, a correction method for the AIRS v6/v7 series was developed and successfully applied.
What are the implications of the main findings?
  • When using AIRS v6/v7 CH4 data at all levels of the standard pressure grid, drift correction is required.
  • The proposed averaged AIRS spectrometer drift correction factors for CH4 measurements are effective at all pressure levels.
  • Corrected AIRS CH4 data show significantly improved agreement with high-precision ground-based measurements; the proposed correction methodology can improve the quality of atmospheric analyses, forecasts and parameter estimates.

Abstract

We analyzed AIRS CH4 volume mixing ratio (VMR) Standard L3 v6/v7 IR-Only Daily products and ground-based measurements from 16 stations of the Network for the Detection of Atmospheric Composition Change (NDACC) at 24 pressure levels from 1000 to 1 mbar. We assessed the dependence of maximum AIRS sensitivity on latitude. At high latitudes, the zone of maximum sensitivity is closer to the surface, at 700–500 mbar; in mid-latitudes, it is 500–250 mbar; and in tropical and subtropical regions, good initial agreement between satellite and ground-based data is observed at 400–200 mbar for both AIRS product versions. At the vast majority of pressure levels and all comparison sites, a unidirectional negative drift in the difference between satellite and ground-based measurements (i.e., discrepancy drift) was observed. Drift coefficients were calculated for each statistically supported pressure level. Two regions of maximum drift were identified: one in the lower atmosphere (925–850 mbar) and another near 50 mbar. The smallest drift was observed at 400–200 mbar. As the main result of the study, we developed and applied correction factors for all 23 AIRS v6 and v7 levels. Using these coefficients led to much better agreement between long-term methane trends from ground-based and satellite measurements and to higher correlation coefficients across all comparison sites.

1. Introduction

It is well established that global warming is largely driven by increasing concentrations of greenhouse gases [1], primarily carbon dioxide and methane. Methane CH4 has an atmospheric lifetime of approximately 10 years and is the second most important greenhouse gas after CO2. Although global methane concentrations (1942 ppb in 2024) are roughly 200 times lower than carbon dioxide concentrations (423.9 ppm in 2024) [2], methane is more effective on a per molecule basis: on a 100-year timescale, its global warming potential is approximately 30 times higher than that of CO2, and on a 20-year timescale, about 80 times higher [3,4,5]. In addition to its greenhouse effect, methane participates actively in atmospheric photochemistry [6,7], depleting hydroxyl OH, the most important atmospheric oxidant, and serving as a major natural source of carbon monoxide (CO), a toxic pollutant and an indicator of economic activity [8].
Methane concentrations were relatively stable in the early 21st century; however, after 2007, they began to increase again, with acceleration in subsequent years [2]. The rate of CH4 growth varies by region [9,10,11]. These facts underscore the importance of high-quality monitoring of methane concentrations and their spatial and long-term variability.
Because ground-based observing networks have uneven spatial coverage—sparse over many continents and absent over oceans—satellite observations are a crucial complement.
Space-based methane monitoring began in the late 20th century. Instruments currently in orbit include AIRS (launched in 2002 [12]), CrIS (three instruments, the earliest launched in 2011 [13,14,15]), IASI (three instruments, the earliest launched in 2006 [16,17], TROPOMI (launched in 2017 [18]), and others. AIRS, IASI and CrIS measure methane in similar spectral regions near 7.7 μm; AIRS and CrIS use the CLIMCAPS retrieval framework [19,20]. Among these, AIRS provides the longest continuous record, covering more than 80% of Earth’s surface from 2003 to the present.
Briefly, AIRS is an instrument on the Aqua satellite designed to monitor meteorological parameters, surface characteristics, and trace gases (CO, CH4, O3 and H2O). Its main element is a grating spectrometer that records Earth’s thermal emissions from 3.75 to 15.4 μm. Methane is measured using channels centered near 7.66 μm [21] with a spatial resolution of about 45 km [12]. Multiple data levels (1, 2, and 3) are publicly available. Level-3 provides daily global maps of several dozen parameters on a 1° × 1° grid; the number of days per year with valid methane data in each grid cell is typically about 320–340, implying very few gaps.
Different versions of AIRS products (e.g., v5, v6, and v7) have been repeatedly validated against ground-based spectroscopic measurements [9,11,22,23,24], data from other orbital instruments [15,25], and through comprehensive validation campaigns involving airborne and balloon-based observations of HIPPO [26], NOAA GML [27], and ATom [28]. Results of these validation works are summarized in [29]. Many studies have used AIRS data on total methane content (CH4 total column, TC), column-averaged dry-air mole fraction (X[CH4]), and VMR at specific pressure levels (e.g., 1000; 925; 800; 600; 300 mbar) [9,30,31,32,33,34,35,36]. In addition, AIRS CH4 data have been used for cross-sensor comparisons; assessments of intra-annual variability and its geographic and temporal anomalies [34]; studies of emissions and source characteristics [35]; and for deriving global and regional trend estimates and characterizing spatiotemporal variability and long-term trends in atmospheric composition [9,32,33,37]. AIRS data continue to be in high demand among atmospheric composition and climate researchers; in the last five years alone, several dozen papers have been published (with more than 20 focused on methane).
AIRS product developers regularly assess the stability of radiance calibration and performance in key spectral channels. These studies also note that the Aqua satellite carrying AIRS has highly stable orbit characteristics, suggesting no concern about drift in surface radiance [38,39], and they estimate the stability of AIRS temperature profile measurements to be high. However, the authors also acknowledge that stability in one group of spectral channels does not necessarily imply the same stability in other groups, which can lead to drift in the quality of specific retrieved characteristics.
Recently, a unidirectional and statistically significant drift of the residual (the difference between satellite and ground-based measurements) was detected: first for AIRS v6 (total CH4 content) [24], and then for v7 (X[CH4], the column-averaged concentration) [11]. In these studies, the measurement series of 18 NDACC stations were used as a reference; based on the calculated average drift coefficient, a linear correction of the AIRS CH4 series was performed, and a significant improvement in agreement between satellite and ground-based measurements was obtained (including for an independent data set of 11 TCCON stations [11]). Since this drift was not taken into account in all prior studies using AIRS methane data, the results of most such studies should be interpreted with caution.
The objective of this study is to extend the investigation of the AIRS methane discrepancy drift and its causes. The primary goals are to: (i) estimate the pressure ranges of maximum sensitivity for AIRS v6 and v7 methane data, accounting for geographic differences; (ii) estimate discrepancy drift at different pressure levels; (iii) calculate correction factors for individual altitudes and geographic zones; (iv) apply the corrections to the orbital time series; and (v) evaluate the effectiveness of the proposed correction. Measurements from high-precision spectrometers in the NDACC network were used as reference data.

2. Materials and Methods

2.1. NDACC Ground Measurements (Basic Information)

High-precision measurements from the NDACC ground network were selected as reference datasets for CH4 concentrations at different pressure levels to validate AIRS satellite data [40]. The stations use Fourier Transform Infrared (FTIR) spectrometers and provide broad latitude coverage. NDACC data are publicly available at https://ndacc.larc.nasa.gov (accessed 4 June 2026). Among other products, NDACC provides methane concentrations on multiple vertical levels (up to 48), with the number of levels depending on station altitude.
On clear or partly cloudy days, typically 2 to 10 (or more) spectra are recorded between about 09:00 and 15:00 local time, depending on day and station. Because intraday methane variability is small, modest time offsets between ground-based and satellite observations are not critical. For comparison with AIRS, we used daily average methane values from NDACC. The accuracy of CH4 TC measurements is 0.3% [41]. NDACC sites were chosen because they provide longer, more uniform time series than the newer TCCON network [42]. Overlap of NDACC observation periods with AIRS, the availability of vertical methane profile data (e.g., profiles for different pressure levels), and sufficient statistical coverage motivated our choice of ground-based datasets. The measurement sites used in the analysis, their geographic locations, coverage periods, and the number of comparison pairs with satellite data are presented in Table 1 and Figure 1.
We selected all stations with sufficient data and a relatively uniform distribution of measurements over the study period (2003–2022). The Paramaribo station was excluded due to limited coverage (179 days of synchronous measurements with AIRS v6/v7 CH4 VMR). Thus, 16 stations (all except Paramaribo) were used in subsequent analyses.
When validating the vertical structure of satellite measurements, a key challenge is to correctly compare vertical profiles from two different measurement systems on a common grid. Specifically, it is necessary to align the data with the 24 measurement levels of the standard AIRS grid. This requires converting NDACC profiles, originally provided on altitude (km) levels, to pressure (mbar) and mapping from up to 48 NDACC levels to the 24 AIRS levels.
To convert altitude to pressure, several options can be used:
  • Using the barometric formula [43], which requires additional temperature profile data;
  • Applying the International Standard Atmosphere (ISA) table [44] with interpolation between tabulated values;
  • Extracting pressure values from ERA5 reanalysis fields [45], https://climate.copernicus.eu/climate-reanalysis, accessed 17 June 2026;
  • Using NDACC’s own pressure data, provided at the same altitude levels as the methane data and averaged over the study period, which captures local pressure variations.
Because the purpose of this study is to compare vertical structures, we converted from the height grid to the pressure grid using the in situ pressure reported with the NDACC profiles, averaged over 2003–2022, following [11], thereby avoiding model-based conversions.
When comparing two profile types—the AIRS profile smoothed to 24 pressure levels and the higher-resolution (48 levels) NDACC profile—it is not sufficient to compare nearest levels. Vertical interpolation of the CH4 profile by pressure is required to map the NDACC levels to the standard AIRS levels.
The interpolation is performed as follows:
C P = C 1 + C 2 − C 1 l n P 2 − l n P 1 × L n P − l n P 1
where
CP—this is the concentration at the calculated pressure level (ppm);
C1, C2—this is the concentration at the boundary levels (below and above the calculated level) (ppm);
P—this is the calculated pressure level (mbar);
P1 and P2—this are pressures at the lower and upper bounding levels, respectively (mbar).
It should also be noted that the standard AIRS grid includes 1000 mbar, whereas the NDACC baseline pressure is approximately 970 mbar (and even less for high-altitude measurement sites, such as Zugspitze, Jungfraujoch, Izana, Mauna Loa, and Reunion Island). Therefore, the 1000 mbar level would require extrapolation below the lowest NDACC level. Although extrapolation might yield plausible values with a dense grid, its error cannot be verified and is undesirable for validation. We therefore excluded the 1000 mbar level (and, for high-altitude sites, all levels below station elevation).
Interpolating NDACC profiles to AIRS pressure levels yields a common vertical grid and enables a consistent analysis of the pressure-level structure of differences between satellite and ground-based measurements, aligned with our objectives. Similar approaches are used, for example, in [46,47].

2.2. AIRS CH4 Products

The analysis uses methane measurements from the AIRS L3 v6/v7 satellite spectrometer, and the volume mixing ratio (VMR), measured in ppbv. Only daytime data from the ascending node (approximately 13:30 local time) are used. The spatial resolution of the data is 1° × 1°, and the vertical resolution comprises 24 pressure levels, reported in mbar [48,49,50]. For details, see Table 2.
According to Tian et al. (2020), AIRS atmospheric methane (ppbv) is provided as both a gridded pressure profile and a column-averaged concentration (for v6 and v7) [48]. The v6 data [49,50] also include total CH4. Per [21], the v6 and v7 retrieval algorithms are similar, and the peak sensitivity of AIRS instrument to methane is in the range of 150–300 mbar in tropical latitudes and 200–400 mbar in mid-latitudes.
The coordinates of the NDACC stations in Table 1 are taken as the centers of the corresponding 1° × 1° grid cells. Data extraction was performed using the Tropomi Tools software (v8.023) and associated analysis suite developed at the Institute of Atmospheric Physics, Russian Academy of Sciences [51], which supports multiple data formats and provides extraction, analysis, and visualization.
All data were used as is, without additional quality screening or filtering. Missing data were not gap-filled or extrapolated. Only synchronous measurements were used; i.e., days when both satellite and ground-based observations were available. For NDACC, daily average values were used for comparison. We note that NDACC ground-based measurements are generally performed under clear or partly cloudy conditions; accordingly, AIRS observations on these days are less likely to be affected by cloud-related errors.

2.3. Drift Correction Technique

The drift correction method consists of determining the slope coefficient of the linear trend of the AIRS—GR difference in physical terms (ppm/day) and then applying this drift coefficient for a day-by-day correction of the original series:
C CH4 SSD = C CH4 + (N − 1) × SSD,
where
C CH4 SSD is the adjusted CH4 concentration (ppm);
C CH4 is the original CH4 concentration (ppm);
N is the day number, starting from 1 January 2003;
SSD is the drift coefficient (ppm/day).
The methodology and its application to correct the AIRS methane series for the entire atmospheric column in v6 (CH4 TC) and v7 (XCH4) are described in detail in [11,24]. This correction, using average individual drift coefficients (i.e., calculated for each pressure level), was applied to each AIRS version 7 data level. The presented correction factors enable users of AIRS v6 and v7 CH4 products to adjust long-term series for total column (v6), column-average (v7), and VMR at individual pressure levels. All necessary information is provided in [11,24] and in this paper.

3. Results

3.1. AIRS L3 v7 Validation

3.1.1. Determination of the Maximum Sensitivity Zone of the Initial Satellite Measurement Series

In this study, satellite and ground-based measurements were compared at 23 of the 24 pressure levels of the standard AIRS grid (excluding 1000 mbar). For each station and pressure level, we calculated correlation coefficients and identified the zones of maximum agreement between satellite and ground-based measurements (see Table 3).
The stations are divided into three groups based on geographic conditions:
Group 1—high-latitude sites: Eureka, Ny-Alesund, Thule, and Arrival Heights;
Group 2—mid-latitude sites: Kiruna, Harestua, St. Petersburg, Bremen, Zugspitze, Jungfraujoch, Toronto, Rikubetsu, and Lauder. Note: Although NDACC lists Kiruna and Harestua as high-latitude stations, our sensitivity analysis supports classifying them as mid-latitude sites.
Group 3—subtropical/tropical sites: Izana, Mauna Loa, and Reunion Island.
The altitude of maximum AIRS sensitivity varies significantly with latitude. According to the developers (https://airs.jpl.nasa.gov/data/products/v7-L2-L3/, accessed 3 June 2026), “AIRS methane product (ppbv) is reported as a profile on the fixed pressure grid and also as a total column burden. AIRS methane sensitivity peaks at 150–300 hPa in the tropics and 200–400 hPa in mid-latitudes. The algorithm is based on optimal estimation and the retrieval is performed for 10 layers”.
Our results are consistent with this statement and indicate that, for high-latitude sites, the zone of maximum sensitivity is located closer to the surface, in the range 700–400 mbar (see Table 3). For Group 1 stations, correlation coefficients (R) within this zone are modest, typically 0.5–0.6, with the exception of Arrival Heights, where R reaches ~0.7. At other pressure levels, correlations are weak or not statistically significant.
At mid-latitudes, the greatest agreement—and hence the highest R between satellite and ground-based measurements—occurs in the 500–200 mbar range, with R up to 0.83 for Kiruna (at 400 mbar) and 0.85 for Jungfraujoch (at 300 mbar). For Harestua and Toronto, no significant correlation with ground-based data was found at any pressure level. Overall, as seen in Table 3, correlation coefficients between satellite and ground-based measurements at mid-latitudes are higher than at high-latitude sites.
For subtropical/tropical sites, the band of maximum agreement is shifted to 400–200 mbar. The difference from the developer-noted 300–150 mbar zone may reflect the small number of comparison sites (three), which limits a more detailed analysis. Nevertheless, correlations are also high in this band and decrease slightly at 150 mbar; for example, R reaches 0.92 at 300 mbar for Izana.
The finding that the maximum-sensitivity zone lies closer to the surface at high latitudes is physically consistent with infrared sounding. IR retrievals are influenced by many factors, including the vertical temperature profile, surface temperature, water vapor, tropopause height, and the thermal contrast between the surface and the atmosphere. In polar regions the troposphere is shallower, water vapor is lower during cold periods (water vapor absorbs IR and limits penetration depth), and thermal structure differences (e.g., frequent inversions) are pronounced—all of which shift the sensitivity peak downward.
Our results align with AIRS product documentation and prior studies [52]. According to the AIRS science pages (https://airs.jpl.nasa.gov/sounding-science/composition, accessed 16 June 2026), “the peak sensitivity of AIRS methane measurements is observed at approximately 200 mbar, and the data uncertainties are small enough to allow mapping of seasonal variations in methane content in the middle and upper troposphere. The uncertainty estimate for AIRS methane measurements is 20%, including accuracy and precision”.
However, we recently identified a persistent, unidirectional drift in AIRS total methane measurements for v6 [24], as well as a similar drift in column-averaged methane for v7 [11], which leads to an underestimation of long-term CH4 trends by approximately a factor of two relative to ground-based estimates. Further analysis indicates that a similar issue applies to the profile data at multiple pressure levels.

3.1.2. Determination of Zones of Maximum Drift Localization

To localize the maximum drift in AIRS CH4 VMR, we computed slope coefficients of the linear trend for the difference between satellite and ground measurements (satellite minus ground measurements, AIRSv7-GR) at all statistically supported pressure levels at sites listed in Table 1.
Tables S1-1 and S1-2 report the linear trend slopes of the difference for each pressure level from 925 to 1 mbar (23 levels), along with confidence intervals. With few exceptions, the difference drift is negative and statistically significant (confidence intervals exclude zero). The slopes vary with pressure, indicating level-dependent sensitivity of the AIRS retrieval.
Analysis of the slopes for various measurement sites on the altitude grid confirms the presence of a unidirectional drift of the difference. At Ny-Alesund, St. Petersburg, Toronto, and Izana, the AIRS − GR slope becomes positive at some levels, but such isolated cases occur above about 400 mbar.
As noted previously, many research groups use AIRS data at selected levels (e.g., 925, 850, 600, 400 mbar) [9,31,33,35] without accounting for the fact that one of the two maxima in AIRS spectrometer drift lies in this region (see Figure 2, Tables S1-1 and S1-2).
In this area, the average slope coefficient of the difference, calculated by averaging the values from all measurement sites, reaches 1.12 × 10−5 ± 0.18 × 10−5 ppm/day, which significantly affects the representativeness of the data.
A second maximum of the drift appears near 50 mbar; while its impact on total content or column-mean concentration is smaller, the slope there is (1.15 ± 0.30) × 10−5 ppm/day.
High-altitude NDACC sites, such as Reunion Island, Izana, Jungfraujoch, Mauna Loa, and Zugspitze, lack data at pressure levels lower than those corresponding to station altitude (e.g., below about 500 mbar), as do some other elevated sites. Consequently, average drift estimates at these levels are based on fewer stations (nine rather than sixteen) and are less certain. As shown in Section 3.2, applying these correction factors improves the consistency of satellite data with ground-based measurements and enhances correlations both within the maximum-sensitivity zones and at other pressure levels.
The overall average slope coefficient of the AIRS v7-GR difference trend was also estimated for all statistically supported levels (23 of the 24 levels of the standard AIRS grid). After averaging stations by pressure level, all levels were averaged into a single coefficient. This is appropriate because the slope coefficient of the difference trend is a linear value in absolute units and is not related to the level weights, and differences in level thicknesses do not affect the estimated value (unlike, for example, trend assessment, where such averaging must also take into account the weight of each level).
The resulting profile-mean coefficient is (7.19 ± 1.97) × 10−6 ppm/day. As shown in [11], when comparing AIRS v7 VMR with ground-based NDACC measurements for concentration in the air column, the average drift coefficient was 7.20 × 10−6 ppm/day. Despite excluding the 1000 mbar level—which would only slightly increase the value of the averaged coefficient—these results agree well; residual differences arise from interpolation across levels, averaging procedures, and the differing numbers of stations and comparison pairs.

3.2. Trends at Different Pressure Levels and AIRS v7 Drift Correction

Long-term trends in AIRS v7 CH4 VMR and their 95% confidence intervals for 2003–2022, computed from the original (uncorrected) series, are provided for all pressure levels from 925 to 1 mbar in Tables S2-1 and S2-2. These calculations use synchronous satellite and ground-based values; trend estimates are independent and computed separately for each series.
Across the 23 levels, the satellite series substantially underestimate the trends—by more than a factor of two at 925 and 850 mbar and at levels above 100 mbar—relative to ground-based estimates (see also Figure 3). The largest discrepancies occur in the lower troposphere (925–850 mbar; likely also 1000 mbar, as suggested indirectly by Figure 2 and Figure 3) and in the upper atmosphere above 100 mbar. Using uncorrected satellite data, only the 400–200 mbar region yields trend estimates close to ground-based results, broadly consistent with the stated AIRS maximum-sensitivity region; the trend confidence intervals intersect only at these levels. However, the pressure range most commonly used in scientific analyses (1000–500 mbar) clearly requires proper treatment of the instrument’s long-term drift to obtain reliable trends.
Thus, after applying the level-specific drift coefficients from Tables S1-1 and S1-2, agreement between satellite and ground-based measurement data improves markedly at all levels (see Figure 4; details in Tables S4-1 and S4-2). As seen in Figure 4, at pressure levels up to about 800 mbar the discrepancy in trend values is somewhat larger than elsewhere (the satellite underestimates the trend by 10–15% compared to estimates based on NDACC data), but the satellite and ground-based trends still overlap within their 95% confidence intervals. A slight underestimation of the correction coefficients in this range may reflect the smaller number of comparison sites contributing at these levels (9 instead of 16), which reduces statistical robustness. Overall, the method performs well.
For pressure levels above 15 mbar, the drift correction coefficients tend to yield slightly higher trend magnitudes (i.e., slight overcorrection). Nevertheless, as with 925 and 850 mbar, the corrected satellite and ground-based trends overlap within their confidence intervals. This behavior may relate to reduced sensitivity of both satellite and ground-based measurements at such altitudes and to the averaging approach for the difference slopes, which includes cases where the confidence interval exceeds the estimated value (statistically insignificant slopes). We note that such statistically insignificant slopes, when present at other levels, do not impede the derivation of stable drift coefficients in those cases.
Applying drift correction factor increases correlations between satellite and ground-based measurements relative to the original series. Complete correlation coefficients for the original AIRS v7 series at all pressure levels from 925 to 1 mbar are given in Table S5; analogous results for the drift-corrected AIRS v7 series are in Table S6.

3.3. Validation of AIRS L3 v6

Similar to the AIRS VMR CH4 L3v7 product, the AIRS VMR CH4 L3 v6 satellite product was validated. The slope coefficients for the satellite–ground difference at all statistically supported pressure levels are given in Tables S7-1 and S7-2. As with v7, these slopes are predominantly negative and vary substantially with pressure level. The overall vertical distribution (e.g., distribution by pressure levels) of drift coefficients closely matches that for v7; however, station-specific and station-averaged values at individual levels differ slightly from v7.
With respect to altitude sensitivity, the consistency and zones of maximum correlation between satellite and ground-based measurements for the v6 source data are also similar to v7 (see Table 4).
Applying level-specific correction factors from Tables S7-1 and S7-2 (as in v7, an individual drift coefficient was computed and applied for each level) also improved data agreement. After correction across all pressure levels, satellite and ground-based trend estimates became substantially closer. Complete AIRS v6 residual trend estimates and 95% confidence intervals before correction are provided in Tables S8-1 and S8-2; ground-based trends are given in Tables S3-1 and S3-2; and AIRS v6 results after correction are presented in Tables S9-1 and S9-2.

4. Discussion

This work extends our research on identifying and correcting data-quality drift in AIRS v6 and v7 methane measurements. In earlier studies [11,24], such drift was detected for total column and column-mean CH4, and a straightforward correction method was proposed and validated against an independent dataset (TCCON). Those studies, however, did not resolve the drift as a function of altitude.
Here, we assessed the vertical sensitivity of AIRS CH4 VMR (v6/v7) and to characterize the vertical distribution of drift across pressure levels from 925 to 1 mbar. Our principal result is the derivation of level-specific drift estimates for all AIRS layers, along with corresponding correction factors. Researchers using AIRS data at individual levels can apply these results directly (see Table S1 for v7 and Table S7 for v6).
One of our primary objectives was to identify the region of maximum drift in the satellite spectrometer and to evaluate the effectiveness of the correction method across all pressure levels in the standard AIRS L3 vertical grid. A second objective was to investigate the causes of drift in satellite CH4 products in general, as discussed in [11].
It is important to distinguish between “observed AIRS–NDACC drift” and “true AIRS instrumental drift.” Strictly speaking, our methodology determines the temporal change in the difference between two observing systems. Therefore, we view the calculated coefficients primarily as empirical correction factors for the observed long-term AIRS–NDACC discrepancy, rather than as a direct assessment of purely physical degradation of the AIRS instrument.
We note that a more representative analysis of the vertical characteristics of this drift would require long-term aircraft or balloon observations across diverse geographic zones; such datasets are presently lacking. Consequently, our study relies on comparisons with long-term records from the most statistically robust NDACC stations.
Our analysis confirms that drift in AIRS data occurs not only in total-column and column-mean CH4 products [11,24], but also in vertical VMR profiles, where it is non-uniform with altitude, varies by level, and shows latitudinal dependence.
Zones of maximum drift are found in the lower troposphere (925–850 mbar) and near 50 mbar. The region identified by developers as having maximum sensitivity (400–200 mbar) is less affected, although a smaller but statistically significant drift is still detected and can be quantified with our method.
Differences in the vertical sensitivity functions of AIRS and NDACC spectrometers likely contribute; nevertheless, even after accounting for such differences, the drift remains unidirectional, with only the vertical structure subject to refinement.
Differences in drift coefficients across pressure levels are undoubtedly linked to the instrument’s sensitivity to CH4 in different atmospheric layers. Nevertheless, in our view, data at all levels exhibit some degree of drift and benefit from correction to improve representativeness.
The coefficients we provide represent a globally averaged estimate of the long-term AIRS systematic error and do not fully account for possible latitude- and region-dependent differences in drift behavior in layers above 100–50 mbar. However, we consider the correction factors for layers below 100 mbar to be representative and suitable for calculating trends and characteristics of methane variability in any geographic zone.
According to the developers, the AIRS v7 VMR product is largely consistent with AIRS v6 VMR, and our analysis found no substantive differences in series quality. Consequently, it is more practical to use the newer v7 product for research, applying the level-specific correction factors provided in the Supplementary Material. Furthermore, this version is recommended for scientific research.
The proposed correction method improves agreement between satellite and ground-based measurements across all levels of the standard AIRS vertical grid, enhancing the quality of trend estimates. Despite the good agreement between the drift coefficients calculated at individual pressure levels for v6 and v7, we recommend using individual averaged coefficients when applying the correction:
  • For v6 total methane content (CH4 TC), use SSD = 1.64 × 1014 molecules/cm2/day (7.62 × 10−6 ppm/day) from [24];
  • For v7 column-average X[CH4], use SSD = 7.20 × 10−6 ppm/day (1.55 × 1014 molecules/cm2/day);
  • For individual levels in v6 and v7, use the level-specific factors listed in Tables S1-1 and S1-2 (v7) and Tables S7-1 and S7-2 (v6).
We also found that agreement at high latitudes is somewhat weaker than in mid-latitude, subtropical, and tropical regions. Moreover, when analyzing the seasonal variability in the AIRS−GR difference for total-column products—AIRS v6 Tot L3 IR-Only CH4 and AIRS v6 Tot L3 IR-Only H2O—we identified seasonal variations in the methane residual that are in strict anti-phase with the H2O total column (see Figure 5). At high latitudes, the amplitude of seasonal water vapor variation is largest, since the cold late-winter/early-spring atmosphere (March–April) contains very little water vapor, while summer values are several times higher. As seen in Figure 5, both the H2O total column and the AIRS − GR CH4 residual reach their minima in March–April, the coldest period for the Arctic, when temperatures can reach −20–30 °C.
The correlation between the two compared datasets indicates a moderate relationship (R = 0.4 for Eureka; 0.6 for Ny Alesund; 0.3 for Thule; 0.4 for ArHeights), which indirectly supports the hypothesis advanced in [11]: the persistent, unidirectional negative drift observed for all AIRS methane products may be linked not only to possible instrument degradation near 7.66 μm spectral region and to direct H2O interference in infrared retrievals but also with an underestimation of water vapor influence, whose absorption lines intersect with the methane extraction lines [50].

5. Limitations and Future Objectives

As noted in our previous papers [11,24], the proposed correction method has several limitations. First, the drift of the AIRS-NDACC difference cannot be assumed to be strictly linear over 2003–2022. However, our earlier works also indicate that some nonlinearity, manifested in discrepancies in the drift magnitude over different time periods, has little effect on the correction’s effectiveness [11].
Second, quantifying seasonal variations in the drift—likely linked to H2O absorption near 7.66 μm—requires more comparison sites with uniform, statistically robust time series matching the AIRS record length. At present, only 16 such series are available. While the high-latitude examples provide indirect evidence of water vapor’s influence on AIRS CH4 retrieval quality, the current dataset is insufficient for a robust geographic characterization of the seasonal residual.
Third, drift estimates at higher altitudes (above ~100 mbar) are more uncertain. The vertical sensitivity functions (averaging kernels) of ground-based and satellite spectrometers differ substantially: NDACC FTIR instruments have higher sensitivity in the troposphere, with decreasing sensitivity aloft. Strictly speaking, NDACC data cannot serve as a reference at these higher altitudes. Nevertheless, most atmospheric CH4 resides in the troposphere, where NDACC sensitivity is strongest, and most users employ AIRS data for 1000–400 mbar levels.
Insufficient consideration of water vapor absorption and its trend may be not only a source of seasonal variability but also a cause of increasing discrepancies [24]. Additional targeted analyses are needed for an unambiguous conclusion.
Airborne or balloon measurements could provide useful information on satellite profile discrepancies above the tropopause. However, aircraft soundings typically reach only ~12–13 km above sea level [26,27,28,53], which is insufficient for sounding in the stratosphere. Balloon soundings can reach 37–40 km [54,55], but both methods have limited spatial and temporal coverage, precluding their use for robust long-term drift assessment.
Thus, at present, one of the few practical options for investigating AIRS vertical profiles over the long term is the use of NDACC FTIR data.
For future work, it would be valuable to assess the presence or absence of drift in the difference between satellite and ground-based data from other orbital instruments that observe CH4 in the same spectral region as AIRS (around 7.7 μm), notably IASI and CrIS—especially CrIS, which offers finer spatial resolution but uses retrieval approaches similar to AIRS (CLIMCAPS [19,20]). It would also be useful to compare vertical profiles across different satellite instruments. In addition, acquiring long-term, multi-regional balloon observations reaching the stratosphere would be an important step toward independent validation of the vertical structure of satellite measurements and their drift at high altitudes.

6. Conclusions

  • We performed a sensitivity analysis of the CH4 AIRS Standard L3 IR AIRS Only Daily satellite products, Versions 7 and 6, across pressure levels from 925 to 1 mbar and across different geographic regions.
  • The results obtained for both products are similar. At high-latitude sites, the zone of maximum sensitivity lies at 700–400 mbar, with modest maximum correlations (R ≈ 0.5–0.6) between original satellite and ground-based series. In mid-latitudes, the greatest agreement occurs at 500–200 mbar, with R up to 0.85 at 300 mbar (Jungfraujoch). In tropical/subtropical regions, maximum sensitivity is at 400–200 mbar, with R up to 0.92 at 300 mbar (Izana).
  • Slope coefficients of the linear trend of the AIRS-GR difference were estimated for all statistically supported pressure levels from 925 to 1 mbar. We found a non-uniform but unidirectional negative trend whose magnitude varies by level, confirming drift in the AIRS CH4 VMR v6/v7 products across the full pressure grid.
  • Zones of maximum drift were identified for the AIRS CH4 VMR v6/v7 products. The first lies at 925–850 mbar, where the drift coefficient reaches (1.12 ± 0.18) × 10−5 ppm/day. The second is near 50 mbar, with a slope of (1.15 ± 0.30) × 10−5 ppm/day. Accordingly, AIRS methane data should be adjusted at all pressure levels before use.
  • Level-specific correction factors were computed and successfully applied for all pressure levels from 925 to 1 mbar. After correction, agreement between CH4 VMR trend estimates improved markedly. Prior to correction, trends derived from the original satellite and ground-based data differed by more than a factor of two, with the satellite underestimating the magnitude. After correction, discrepancies decreased at all levels, and station-averaged level trends overlapped within their confidence intervals, including for levels above 100 mbar.
  • Application of correction factors increased correlations in all data series across all pressure levels.

Supplementary Materials

The following supporting information can be downloaded at https://www.mdpi.com/article/10.3390/rs18172875/s1: Table S1-1: AIRS v7-GR difference slope (NDACC) ± 95% confidence interval for pressure levels from 1000 mbar to 100 mbar; Table S1-2: AIRS v7-GR difference slope (NDACC) ± 95% confidence interval for pressure levels from 70 mbar to 1 mbar; Table S2-1: CH4 trend ± 95% confidence interval, AIRS v7 initial data, % per year (for 1000–100 mbar pressure levels); Table S2-2: CH4 trend ± 95% confidence interval, AIRS v7 raw data (for 70–1 mbar pressure levels); Table S3-1: CH4 trend ± 95% confidence interval, NDACC initial data (for 1000–100 mbar pressure levels); Table S3-2: CH4 trend ± 95% confidence interval, NDACC initial data (for pressure levels 70–1 mbar); Table S4-1: CH4 trend ± 95% confidence interval, AIRS v7 corrected data (for 1000–100 mbar pressure levels); Table S4-2. CH4 trend ± 95% confidence interval, AIRS v7 corrected data (for 70–1 mbar pressure levels); Table S5: Correlation coefficient R of original satellite AIRS v7 and ground data for different pressure levels; Table S6: Correlation coefficient R of corrected AIRS v7 satellite and ground-based data for different pressure levels; Table S7-1. AIRS v6-GR(NDACC) difference slope ± 95% confidence interval for pressure levels from 1000 mbar to 100 mbar; Table S7-2. AIRS v6-GR(NDACC) difference slope ± 95% confidence interval for pressure levels from 70 mbar to 1 mbar; Table S8-1. CH4 trend ± 95% confidence interval, AIRS v6 initial data for pressure levels from 1000 mbar to 100 mbar; Table S8-2. CH4 trend ± 95% confidence interval, AIRS v6 initial data for pressure levels from 70 mbar to 1 mbar; Table S9-1. CH4 trend ± 95% confidence interval, AIRS v6 corrected data for pressure levels from 1000 mbar to 100 mbar; Table S9-2. CH4 trend ± 95% confidence interval, AIRS v6 corrected data for pressure levels from 70 mbar to 1 mbar.

Author Contributions

Conceptualization, E.F., V.R. and A.S.; methodology, V.R.; software, E.F. and A.B.; validation, E.F., A.B. and N.K.; formal analysis, E.F., V.R., N.P. and N.K.; investigation, E.F., V.R. and Y.S.; resources, N.P., A.B. and N.K.; data curation, E.F. and N.P.; writing—original draft preparation, E.F., V.R. and A.S.; writing—review and editing, V.R., V.S., A.S., Y.S. and L.W.; visualization, V.R., E.F., A.B. and N.K.; supervision, A.S., V.S., Y.S. and L.W.; project administration, L.W. and V.S.; funding acquisition, V.S. All authors have read and agreed to the published version of the manuscript.

Funding

The research was supported by the Russian Science Foundation, within project No 25-77-31009.

Data Availability Statement

Data underlying the results presented in this paper are available on request from the authors. Publicly available datasets were analyzed in this study.

Acknowledgments

The authors are grateful to all principal investigators (PIs) and supporting staff of NDACC and TCCON networks and of AIRS team for deploying, maintaining and making available data from numerous observation sites and AIRS long-term orbital mission used in the paper, as well as to all funding agencies having contributed to these programs over the years.

Conflicts of Interest

The authors declare no conflicts of interest.

Abbreviations

The following abbreviations are used in this manuscript:
AIRS v6/v7Atmospheric InfraRed Sounder version 6/7
CrISCross-Track Infrared Sounder
IASIInfrared Atmospheric Sounding Interferometer
CLIMCAPSCommunity Long-Term Infrared Microwave Combined Atmospheric Product System
TROPOMITROPOspheric Monitoring Instrument
L3Level 3 (3rd data level)
NDACCNetwork for the Detection of Atmospheric Composition Change
TCCONTotal Carbon Column Observing Network
SSDSatellite spectrometer drift
SAT-GR/AIRS-GRSatellite data minus ground-based data (the difference between satellite and ground-based data)
Ppm/PpbParts per million/billion
NASANational Aeronautics and Space Administration
TCTotal column
FTIRFourier-transform infrared spectrometer
X[GAS]Thickness-averaged relative volume concentration of the detected gas in ppm/ppb
VMRVolume mixing ratio
SCSlope coefficient
aslAbove sea level
NDNo data

References

  1. IIPCC: Summary for Policymakers. In Climate Change 2023: Synthesis Report. Contribution of Working Groups I, II and III to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change; IPCC: Geneva, Switzerland, 2023; pp. 1–34.
  2. World Meteorological Organization. State of the Global Climate 2025. Update for COP30; WMO: Geneva, Switzerland, 2025; Available online: https://library.wmo.int/idurl/4/69674 (accessed on 20 May 2026).
  3. Myhre, G.; Shindell, D.; Bréon, F.-M.; Collins, W.; Fuglestvedt, J.; Huang, J.; Koch, D.; Lamarque, J.-F.; Lee, D.; Mendoza, B.; et al. Anthropogenic and natural radiative forcing. In Climate Change 2013: The Physical Science Basis; Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change; Stocker, T.F., Ed.; Cambridge University Press: Cambridge, UK; New York, NY, USA, 2013; pp. 659–740. [Google Scholar]
  4. Etminan, M.; Myhre, G.; Highwood, E.J.; Shine, K.P. Radiative forcing of carbon dioxide, methane, and nitrous oxide: A significant revision of the methane radiative forcing. Geophys. Res. Lett. 2016, 43, 12614–12623. [Google Scholar] [CrossRef] [Scilit]
  5. Forster, P.; Storelvmo, T.; Armour, K.; Collins, W.; Dufresne, J.-L.; Frame, D.; Lunt, D.J.; Mauritsen, T.; Palmer, M.D.; Watanabe, M.; et al. The Earth’s Energy Budget, Climate Feedbacks, and Climate Sensitivity. In Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change; Masson-Delmotte, V., Zhai, P., Pirani, A., Connors, S.L., Péan, C., Berger, S., Caud, N., Chen, Y., Goldfarb, L., Gomis, M.I., et al., Eds.; Cambridge University Press: Cambridge, UK; New York, NY, USA, 2021; pp. 923–1054. [Google Scholar] [CrossRef] [Scilit]
  6. National Academies of Sciences, Engineering, and Medicine. The Future of Atmospheric Chemistry Research: Remembering Yesterday, Understanding Today, Anticipating Tomorrow; The National Academies Press: Washington, DC, USA, 2016. [Google Scholar] [CrossRef] [Scilit]
  7. Jacob, D.J. Introduction to Atmospheric Chemistry; Princeton University Press: Princeton, NJ, USA, 1999. [Google Scholar]
  8. Wunch, D.; Wennberg, P.O.; Toon, G.C.; Keppel-Aleks, G.; Yavin, Y.G. Emissions of greenhouse gases from a North American megacity. Geophys. Res. Lett. 2009, 36, L15810. [Google Scholar] [CrossRef] [Scilit]
  9. Zhou, L.; Warner, J.; Nalli, N.R.; Wei, Z.; Oh, Y.; Bruhwiler, L.; Liu, X.; Divakarla, M.; Pryor, K.; Kalluri, S.; et al. Spatiotemporal Variability of Global Atmospheric Methane Observed from Two Decades of Satellite Hyperspectral Infrared Sounders. Remote Sens. 2023, 15, 2992. [Google Scholar] [CrossRef] [Scilit]
  10. Wójcik-Gront, E.; Wnuk, A.; Gozdowski, D. Spatio-Temporal Patterns of Methane Emissions from 2019 Onwards: A Satellite-Based Comparison of High and Low Emission 3 Regions. Atmosphere 2025, 16, 670. [Google Scholar] [CrossRef] [Scilit]
  11. Rakitin, V.; Fedorova, E.; Skorokhod, A.; Kirillova, N.; Chernokulsky, A. Validation of Long-Term Methane (CH4) Satellite Measurements from the AIRS Spectrometer and Correction of Their Temporal Drift. Remote Sens. 2026, 18, 1162. [Google Scholar] [CrossRef] [Scilit]
  12. Aumann, H.H.; Chahine, M.T.; Gautier, C.; Goldberg, M.; Kalnay, E.; McMillin, L.; Revercomb, H.; Rosenkranz, P.W.; Smith, W.L.; Staelin, D.; et al. AIRS/AMSU/HSB on the Aqua mission: Design, science objectives, data products and processing systems. IEEE Trans. Geosci. Remote Sens. 2003, 41, 253–264. [Google Scholar] [CrossRef] [Scilit]
  13. Han, Y.; Revercomb, H.; Cromp, M.; Gu, D.; Johnson, D.; Mooney, D.; Scott, D.; Strow, L.; Bingham, G.; Borg, L.; et al. Suomi NPP CrIS measurements, sensor data record algorithm, calibration and validation activities, and record data quality. J. Geophys. Res. Atmos. 2013, 118, 12734–12748. [Google Scholar] [CrossRef] [Scilit]
  14. Han, Y.; Chen, Y. The Cross-track Infrared Sounder (CrIS) Overview and Validation. In Comprehensive Remote Sensing, 2nd ed.; Liang, S., Ed.; Elsevier: Amsterdam, The Netherlands, 2026; pp. 34–103. [Google Scholar] [CrossRef] [Scilit]
  15. Wang, T.; Payne, V.H.; Manning, E.; Pagano, T.S.; Lambrigtsen, B.; Monarrez, R. A Global, Multidecadal Carbon Monoxide (CO) Record from the Sounder AIRS/CrIS System. Remote Sens. 2026, 18, 5. [Google Scholar] [CrossRef] [Scilit]
  16. Clerbaux, C.; Boynard, A.; Clarisse, L.; George, M.; Hadji-Lazaro, J.; Herbin, H.; Hurtmans, D.; Pommier, M.; Razavi, A.; Turquety, S.; et al. Monitoring of atmospheric composition using the thermal infrared IASI/MetOp sounder. Atmos. Chem. Phys. 2009, 9, 6041–6054. [Google Scholar] [CrossRef] [Scilit]
  17. Clerbaux, C.; Coheur, P.-F.; Bauduin, S.; Boynard, A.; Clarisse, L.; Doniki, S.; George, M.; Hadji-Lazaro, J.; Hurtmans, D.; Lacour, J.-L.; et al. Tracking pollutants from space: 10 years of IASI satellite observation. In Proceedings of the IGAC 2016 Science Conference (International Global Atmospheric Chemistry), Breckenridge, CO, USA, 26–30 September 2016. [Google Scholar]
  18. Hu, H.; Landgraf, J.; Detmers, R.; Borsdorff, T.; Aan de Brugh, J.; Aben, I.; Butz, A.; Hasekamp, O. Toward global mapping of methane with TROPOMI: First results and intersatellite comparison to GOSAT. Geophys. Res. Lett. 2018, 45, 3682–3689. [Google Scholar] [CrossRef] [Scilit]
  19. Smith, N.; Barnet, C.D. Uncertainty characterization and propagation in the Community Long-term Infrared Microwave Combined Atmospheric Product System (CLIMCAPS). Remote Sens. 2019, 11, 1227. [Google Scholar] [CrossRef] [Scilit]
  20. Smith, N.; Barnet, C.D. CLIMCAPS observing capability for temperature, moisture, and trace gases from AIRS/AMSU and CrIS/ATMS. Atmos. Meas. Tech. 2020, 13, 4437–4459. [Google Scholar] [CrossRef] [Scilit]
  21. Thrastarson, H.T.; Fetzer, E.J.; Ray, S. Overview of the AIRS Mission: Instruments, Processing Algorithms, Products, and Documentation, 2nd Edition, The AIRS-Team and the CLIMCAPS Algorithms, Jet Propulsion Laboratory, California Institute of Technology, Thomas Hearty, NASA/GES DISC, Nadia Smith, Science and Technology Corporation, 6 May 2021, Document Edition 2.0. Available online: https://docserver.gesdisc.eosdis.nasa.gov/public/project/AIRS/Overview_of_the_AIRS_Mission.pdf (accessed on 8 July 2026).
  22. Yurganov, L.N.; Rakitin, V.; Dzhola, A.; August, T.; Fokeeva, E.; George, M.; Gorchakov, G.; Grechko, E.; Hannon, S.; Karpov, A.; et al. Satellite-and ground-based CO total column observations over 2010 Russian fires: Accuracy of top-down estimates based on thermal IR satellite data. Atmos. Chem. Phys. 2011, 11, 7925–7942. [Google Scholar] [CrossRef] [Scilit]
  23. Rakitin, V.S.; Shtabkin, Y.A.; Elansky, N.F.; Pankratova, N.V.; Skorokhod, A.I.; Grechko, E.I.; Safronov, A.N. Comparison Results of Satellite and Ground_Based Spectroscopic Measurements of CO, CH4, and CO2 Total Contents. Atmos. Ocean. Opt. 2015, 28, 533–542. [Google Scholar] [CrossRef] [Scilit]
  24. Rakitin, V.; Fedorova, E.; Skorokhod, A.; Kirillova, N.; Pankratova, N.; Elansky, N. Principles of Correction for Long-Term Orbital Observations of Atmospheric Composition, Applied to AIRS v.6 CH4 and CO Data. Remote Sens. 2025, 17, 2323. [Google Scholar] [CrossRef] [Scilit]
  25. Zou, M.; Xiong, X.; Saitoh, N.; Warner, J.; Zhang, Y.; Chen, L.; Weng, F.; Fan, M. Satellite observation of atmospheric methane: Intercomparison between AIRS and GOSAT TANSO-FTS retrievals. Atmos. Meas. Tech. 2016, 9, 3567–3576. [Google Scholar] [CrossRef] [Scilit]
  26. Wofsy, S.C.; Daube, B.C.; Jimenez, R.; Kort, E.; Pittman, J.V.; Park, S.; Commane, R.; Xiang, B.; Santoni, G.; Jacob, D.; et al. HIPPO Merged 10-Second Meteorology, Atmospheric Chemistry Aerosol Data (R_20121129); NSF NCAR Earth Observing Laboratory: Boulder, CO, USA, 2017. [Google Scholar] [CrossRef]
  27. Sweeney, C.; Karion, A.; Wolter, S.; Newberger, T.; Guenther, D.; Higgs, J.A.; Andrews, A.E.; Lang, P.M.; Neff, D.; Dlugokencky, E.; et al. Seasonal climatology of CO2 across North America from aircraft measurements in the NOAA/ESRL Global Greenhouse Gas Reference Network. J. Geophys. Res.-Atmos. 2015, 120, 5155–5190. [Google Scholar] [CrossRef] [Scilit]
  28. Wofsy, S.C.; Afshar, S.; Allen, H.M.; Apel, E.C.; Asher, E.C. ATom: Merged Atmospheric Chemistry, Trace Gases, and Aero-Sols; ORNL DAAC: Oak Ridge, TN, USA, 2018. [CrossRef]
  29. Kulawik, S.S.; Worden, J.R.; Payne, V.H.; Fu, D.; Wofsy, S.C.; McKain, K.; Sweeney, C.; Daube, B.C., Jr.; Lipton, A.; Polonsky, I.; et al. Evaluation of single-footprint AIRS CH4 profile retrieval uncertainties using aircraft profile measurements. Atmos. Meas. Tech. 2021, 14, 335–354. [Google Scholar] [CrossRef] [Scilit]
  30. Yurganov, L.N. The relationship between methane transport to the atmosphere and the decay of the Kara Sea ice cover: Satellite data for 2003–2019. Ice Snow 2020, 60, 423–430. [Google Scholar] [CrossRef] [Scilit]
  31. Yurganov, L.; Carroll, D.; Pnyushkov, A.; Polyakov, I.; Zhang, H. Ocean stratification and sea-ice cover in Barents and Kara seas modulate sea-air methane flux: Satellite data. Adv. Polar Sci. 2021, 32, 118–140. [Google Scholar] [CrossRef]
  32. Abed, F.G.; Halos, S.H. The Vertical variations of Atmospheric Methane (CH4) concentrations over selected cities in Iraq based on AIRS data. Baghdad Sci. J. 2021, 18, 3. [Google Scholar] [CrossRef] [Scilit]
  33. Abed, F.; Rajab, J.; Abdulfattah, I.; Lim, H.S. Influence of air temperature and interrelationship with greenhouse gases (CO2 and CH4) over Iraq using AIRS. Atmos. Pollut. Res. 2024, 15, 102293. [Google Scholar] [CrossRef] [Scilit]
  34. Ni, Q.; Zhou, M.; Wang, J.; Wang, T.; Wang, G.; Wang, P. Intercomparison of CH4 Products in China from GOSAT, TROPOMI, IASI, and AIRS Satellites. Remote Sens. 2023, 15, 4499. [Google Scholar] [CrossRef] [Scilit]
  35. Xu, Z.; Chen, Y.; Shan, W.; Deng, C.; Ma, M.; Wu, Y.; Mao, Y.; Ding, X.; Ji, J. Study of Methane Emission and Geological Sources in Northeast China Permafrost Area Related to Engineering Construction and Climate Disturbance Based on Ground Monitoring and AIRS. Atmosphere 2023, 14, 1298. [Google Scholar] [CrossRef] [Scilit]
  36. Shakirov, R.B.; Khazanova, E.S.; Stepochkin, I.E. New Data on the Patterns of Methane Distribution over the Arctic Shelf of Eurasia. Dokl. Earth Sc. 2023, 511, 623–626. [Google Scholar] [CrossRef] [Scilit]
  37. Zou, M.; Xiong, X.; Wu, Z.; Li, S.; Zhang, Y.; Chen, L. Increase of Atmospheric Methane Observed from Space-Borne and Ground-Based Measurements. Remote Sens. 2019, 11, 964. [Google Scholar] [CrossRef] [Scilit]
  38. Strow, L.L.; DeSouza-Machado, S. Establishment of AIRS climate-level radiometric stability using radiance anomaly retrievals of minor gases and sea surface temperature. Atmos. Meas. Tech. 2020, 13, 4619–4644. [Google Scholar] [CrossRef] [Scilit]
  39. DeSouza-Machado, S.; Strow, L.L.; Kramer, R.J. Geophysical trends inferred from 20 years of AIRS infrared global observations. J. Geophys. Res. Atmos. 2025, 130, e2025JD043501. [Google Scholar] [CrossRef] [Scilit]
  40. De Mazière, M.; Thompson, A.M.; Kurylo, M.J.; Wild, J.D.; Bernhard, G.; Blumenstock, T.; Braathen, G.O.; Hannigan, J.W.; Lambert, J.-C.; Leblanc, T.; et al. The Network for the Detection of Atmospheric Composition Change (NDACC): History, status and perspectives. Atmos. Chem. Phys. 2018, 18, 4935–4964. [Google Scholar] [CrossRef] [Scilit]
  41. Sussmann, R.; Forster, F.; Rettinger, M.; Jones, N. Strategy for high-accuracy-and-precisionretrieval of atmospheric methane from the mid-infrared FTIR network. Atmos. Meas. Tech. 2011, 4, 1943–1964. [Google Scholar] [CrossRef] [Scilit]
  42. Laughner, J.L.; Toon, G.C.; Mendonca, J.; Petri, C.; Roche, S.; Wunch, D.; Blavier, J.-F.; Griffith, D.W.T.; Heikkinen, P.; Keeling, R.F.; et al. The Total Carbon Column Observing Network’s GGG2020 data version. Earth Syst. Sci. Data 2024, 16, 2197–2260. [Google Scholar] [CrossRef] [Scilit]
  43. Hrgian, A.H. Fizika Atmosfery (Atmospheric Physics); Hydrometeorological Izdatelstvo: Saint-Petersburg, Russia, 1969; p. 644. (In Russian) [Google Scholar]
  44. Government Printing Office. U.S. Standard Atmosphere, 1976; U.S. Government Printing Office: Washington, DC, USA, 1976; p. 241.
  45. Hersbach, H.; Bell, B.; Berrisford, P.; Hirahara, S.; Horányi, A.; Muñoz-Sabater, J.; Nicolas, J.; Carole Peubey, C.; Radu, R.; Schepers, D.; et al. The ERA5 global reanalysis. Q. J. R. Meteorol. Soc. 2020, 146, 1999–2049. [Google Scholar] [CrossRef] [Scilit]
  46. Landgraf, J.; Butz, A.; Hasekamp, O.; Hu, H.; aan de Brugh, J. Sentinel 5 L2 Prototype Processors, Algorithm Theoretical Baseline Document: Methane Retrieval, SRON-ESA-S5L2PP-ATBD-001-v3.1-20190517-CH4; SRON Netherlands Institute for Space Research: Utrecht, The Netherlands, 2019; Available online: https://www.sron.nl/wp-content/uploads/2025/03/SRON-ESA-S5L2PP-ATBD-001.pdf?utm_source=chatgpt.com (accessed on 15 April 2026).
  47. De Wachter, E.; Kumps, N.; Vandaele, A.C.; Langerock, B.; De Mazière, M. Retrieval and validation of MetOp/IASI methane. Atmos. Meas. Tech. 2017, 10, 4623–4638. [Google Scholar] [CrossRef] [Scilit]
  48. Tian, B.; Manning, E.; Roman, J.; Thrastarson, H.; Fetzer, E.; Monarrez, R. AIRS Version 7 Level 3 Product User Guide; Jet Propulsion Laboratory, California Institute of Technology: Pasadena, CA, USA, 2020. Available online: https://docserver.gesdisc.eosdis.nasa.gov/public/project/AIRS/V7_L3_Product_User_Guide.pdf (accessed on 9 September 2025).
  49. Jet Propulsion Laboratory, California Institute of Technology. AIRS/AMSU/HSB. Version 6 Data Release User Guide; Olsen, E.T., Ed.; Jet Propulsion Laboratory, California Institute of Technology: Pasadena, CA, USA, 2017. Available online: https://docserver.gesdisc.eosdis.nasa.gov/repository/Mission/AIRS/3.3_ScienceDataProductDocumentation/3.3.4_ProductGenerationAlgorithms/V6_Data_Release_User_Guide.pdf (accessed on 21 February 2026).
  50. Susskind, J.; Blaisdell, J.M.; Iredell, L. Improved methodology for surface and atmospheric soundings, error estimates, and quality control procedures: The atmospheric infrared sounder science team version-6 retrieval algorithm. J. Appl. Remote Sens. 2014, 8, 084994. [Google Scholar] [CrossRef] [Scilit]
  51. Rakitin, V.S.; Kazakov, A.V.; Elansky, N.F. Multifunctional software of the OIAP RAS for processing and analysis of orbital data on the atmospheric composition: Tasks, possibilities, application results, and ways of development. In Proceedings of the 29th International Symposium on Atmospheric and Ocean Optics: Atmospheric Physics; SPIE: Bellingham, WA, USA, 2023; Volume 12780, p. 127805T. [Google Scholar] [CrossRef] [Scilit]
  52. Xiong, X.; Weng, F.; Liu, Q.; Olsen, E. Space-borne observation of methane from atmospheric infrared sounder version 6: Validation and implications for data analysis. Atmos. Meas. Tech. Discuss. 2015, 8, 8563–8597. [Google Scholar] [CrossRef] [Scilit]
  53. Thompson, C.R.; Wofsy, S.C.; Prather, M.J.; Newman, P.A.; Hanisco, T.F.; Ryerson, T.B.; Fahey, D.W.; Apel, E.C.; Brock, C.A.; Brune, W.H.; et al. The NASA atmospheric tomography (ATom) mission: Imaging the chemistry of the global atmosphere. Bull. Am. Meteorol. Soc. 2022, 103, 761–790. [Google Scholar] [CrossRef] [Scilit]
  54. Qian, S.-E.; Routhier, S.; Wu, T.; Dufour, D.; Lantagne, S.; Grandmont, F.; Degenstein, D.; Bourassa, A.; Bourassa, A.; Zawada, D.; et al. GHG measurements using an imaging FTS on a stratospheric balloon: A tech demo for the Arctic Observing Mission. In Proceedings of the SPIE Environmental Remote Sensing, 2024, Edinburgh, UK, 16–20 September 2024; Volume 13193. Remote Sensing of Clouds and the Atmosphere XXIX, 1319309. [Google Scholar] [CrossRef] [Scilit]
  55. Mélanie, G.-D.; Georges, D.; Emmanuel, R.; Nadir, A. 10—Balloon-borne and airborne measurements. In Advances in Spectroscopic Monitoring of the Atmosphere; Chen, W., Venables, D.S., Sigrist, M.W., Eds.; Elsevier: Amsterdam, The Netherlands, 2021; pp. 521–601. [Google Scholar] [CrossRef] [Scilit]
Remotesensing 18 02875 g001
Figure 1. Map of NDACC measurement stations included in the analysis.
Figure 1. Map of NDACC measurement stations included in the analysis.
Remotesensing 18 02875 g001
Remotesensing 18 02875 g002
Figure 2. AIRS v7-GR difference linear trend slope coefficients (NDACC) ± 95% confidence interval (ppm per day) for 23 pressure levels from 925 mbar to 1 mbar, original synchronous series.
Figure 2. AIRS v7-GR difference linear trend slope coefficients (NDACC) ± 95% confidence interval (ppm per day) for 23 pressure levels from 925 mbar to 1 mbar, original synchronous series.
Remotesensing 18 02875 g002
Remotesensing 18 02875 g003
Figure 3. Average CH4 trends from satellite (AIRS v7) and ground-based measurements at different pressure levels, original synchronous series, 2003–2022, % per year. Vertical bars represent 95% confidence intervals.
Figure 3. Average CH4 trends from satellite (AIRS v7) and ground-based measurements at different pressure levels, original synchronous series, 2003–2022, % per year. Vertical bars represent 95% confidence intervals.
Remotesensing 18 02875 g003
Remotesensing 18 02875 g004
Figure 4. Average CH4 trends from satellite (AIRS v7, drift-corrected) and ground-based measurements at different pressure levels, 2003–2022, % per year. Vertical bars represent 95% confidence intervals.
Figure 4. Average CH4 trends from satellite (AIRS v7, drift-corrected) and ground-based measurements at different pressure levels, 2003–2022, % per year. Vertical bars represent 95% confidence intervals.
Remotesensing 18 02875 g004
Remotesensing 18 02875 g005
Figure 5. Seasonal variability in the AIRSv6 difference Tot CH4-GR NDACC (1017 molecules/cm2) and the water content TotH2OVap_A (kg/m2) at station Eureka, illustrating the anti-phase relationship with H2O.
Figure 5. Seasonal variability in the AIRSv6 difference Tot CH4-GR NDACC (1017 molecules/cm2) and the water content TotH2OVap_A (kg/m2) at station Eureka, illustrating the anti-phase relationship with H2O.
Remotesensing 18 02875 g005
Table 1. Location, altitude, periods considered, and number of comparison pairs with AIRS v6/v7 for NDACC stations included in the analysis.
Table 1. Location, altitude, periods considered, and number of comparison pairs with AIRS v6/v7 for NDACC stations included in the analysis.
NºStationLatitude/Longitude, °Altitude Asl, mTime PeriodNumber of Pairs
v7v6
1Eureka, Canada80.0N/86.4W6102006–2020 895875
2Ny Alesund, Norway78.9N/11.9E152003–2022597610
3Thule, Greenland76.5N/68.7W2202003–202213721419
4Kiruna, Sweden67.8N/20.4E4192003–202213541403
5Harestua, Norway60.2N/10.8E5962009–2020482484
6St. Petersburg, Russia59.9N/29.8E202009–2022887878
7Bremen, Germany53.1N/8.8E272004–2022566535
8Zugspitze, Germany47.4N/11.0E29642003–202219342009
9Jungfraujoch,
Switzerland		46.5N/8.0E35802003–202217501764
10Toronto—TAO, Canada43.7N/79.4W1742003–202216171577
11Rikubetsu,
Japan		43.5N/143.8E3802003–2022 481523
12Izana,
Tenerife, Spain		28.3N/16.5W23672003–202213341335
13Mauna Loa, HI, United States.19.5N/155.9W33972007–202212601302
14Paramaribo, Suriname5.7N/55.2W232004–2022 179179
15Reunion Island, Maido, France21.1S/55.4E21552013–2019584586
16Lauder, New Zealand45.0S/169.7E3702003–202224322353
17Arrival Heights,
Antarctica		77.8S/166.7E1842003–2022912942
Table 2. Information on the version and coding of the AIRS CH4 VMR satellite products.
Table 2. Information on the version and coding of the AIRS CH4 VMR satellite products.
Product NameVariableDescription
AIRS Standard L3
Version 6 IR AIRS Only Daily		CH4_VMRCH4 volume mixing ratio (24 lvl), resolution 1° × 1°, ppbv
AIRS Standard L3
Version 7 IR AIRS Only Daily		CH4_VMRCH4 volume mixing ratio (24 lvl), resolution 1° × 1°, ppbv
Table 3. Correlation coefficient R between the original AIRS v7 series and synchronized ground-based data, and maximum sensitivity zones. Red text indicates weak/insignificant correlation (R < 0.5). Bold text denotes pressure levels with R ≥ 0.5 in maximum sensitivity zones. Fill color groups stations by latitude: gray fill indicates high-latitude sites; light green fill indicates mid-latitude sites; orange fill indicates subtropical/tropical sites.
Table 3. Correlation coefficient R between the original AIRS v7 series and synchronized ground-based data, and maximum sensitivity zones. Red text indicates weak/insignificant correlation (R < 0.5). Bold text denotes pressure levels with R ≥ 0.5 in maximum sensitivity zones. Fill color groups stations by latitude: gray fill indicates high-latitude sites; light green fill indicates mid-latitude sites; orange fill indicates subtropical/tropical sites.
№Pressure, mbar925850700600500400300250200150100
StationR AIRSv7/GR
1Eureka 0.240.360.470.560.50
2Ny Alesund 0.490.530.550.540.46
3Thule 0.500.560.590.590.53
4Kiruna 0.720.780.830.800.760.70
5Harestua 0.280.300.340.390.420.44
6St. Petersburg 0.590.650.680.680.650.62
7Bremen 0.560.600.630.650.650.65
8Zugspitze -0.740.780.820.810.75
9Jungfraujoch -0.790.850.840.80
10Toronto 0.360.390.400.350.300.26
11Rikubetsu 0.550.660.750.760.720.65
12Izana 0.850.900.920.900.860.78
13MaunaLoa -0.580.620.630.630.63
14Reunion Island 0.530.620.680.660.58
15Lauder 0.720.740.750.740.720.69
16Arrival Heights 0.620.680.710.700.55
Table 4. Correlation coefficient R for satellite and ground-based measurements, synchronous series, and maximum sensitivity zones for AIRS v6. By analogy with Table 3, red text indicates R < 0.5; bold indicates R ≥ 0.5 in maximum sensitivity zones. Fill color groups stations by latitude: gray fill indicates high-latitude sites; light green fill indicates mid-latitude sites; orange fill indicates subtropical/tropical sites.
Table 4. Correlation coefficient R for satellite and ground-based measurements, synchronous series, and maximum sensitivity zones for AIRS v6. By analogy with Table 3, red text indicates R < 0.5; bold indicates R ≥ 0.5 in maximum sensitivity zones. Fill color groups stations by latitude: gray fill indicates high-latitude sites; light green fill indicates mid-latitude sites; orange fill indicates subtropical/tropical sites.
Pressure, mbar925850700600500400300250200150100
№StationR AIRSv6/GR
1Eureka 0.080.240.400.550.54
2Ny Alesund 0.560.600.610.600.52
3Thule 0.540.600.620.600.51
4Kiruna 0.690.750.790.730.670.59
5Harestua 0.260.260.270.300.310.33
6St. Peters-burg 0.520.590.630.600.560.49
7Bremen 0.550.590.610.630.630.61
8Zugspitze -0.720.770.800.770.71
9Jungfraujoch -0.810.830.810.75
10Toronto 0.390.410.410.350.310.25
11Rikubetsu 0.710.770.790.720.640.53
12Izana -0.890.920.900.840.75
13Mauna Loa -0.650.660.650.64
14Reunion Island -0.670.690.630.460.17
15Lauder 0.710.740.750.750.740.70
16Arrival Heights 0.620.680.700.680.53
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Fedorova, E.; Rakitin, V.; Skorokhod, A.; Kirillova, N.; Belov, A.; Pankratova, N.; Shi, Y.; Wang, L.; Semenov, V. Altitude and Geographic Sensitivity Characteristics of the AIRS Satellite Spectrometer and Drift Correction Using Methane (CH4) Data. Remote Sens. 2026, 18, 2875. https://doi.org/10.3390/rs18172875

AMA Style

Fedorova E, Rakitin V, Skorokhod A, Kirillova N, Belov A, Pankratova N, Shi Y, Wang L, Semenov V. Altitude and Geographic Sensitivity Characteristics of the AIRS Satellite Spectrometer and Drift Correction Using Methane (CH4) Data. Remote Sensing. 2026; 18(17):2875. https://doi.org/10.3390/rs18172875

Chicago/Turabian Style

Fedorova, Eugenia, Vadim Rakitin, Andrey Skorokhod, Natalia Kirillova, Andrey Belov, Natalia Pankratova, Yusheng Shi, Lin Wang, and Vladimir Semenov. 2026. "Altitude and Geographic Sensitivity Characteristics of the AIRS Satellite Spectrometer and Drift Correction Using Methane (CH4) Data" Remote Sensing 18, no. 17: 2875. https://doi.org/10.3390/rs18172875

APA Style

Fedorova, E., Rakitin, V., Skorokhod, A., Kirillova, N., Belov, A., Pankratova, N., Shi, Y., Wang, L., & Semenov, V. (2026). Altitude and Geographic Sensitivity Characteristics of the AIRS Satellite Spectrometer and Drift Correction Using Methane (CH4) Data. Remote Sensing, 18(17), 2875. https://doi.org/10.3390/rs18172875

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop