Evaluation of ICESat-2 ATL09 Atmospheric Products Using CALIOP and MODIS Space-Based Observations
1
Earth System Science Interdisciplinary Center, University of Maryland, College Park, MD 20740, USA
2
NASA Goddard Space Flight Center, Greenbelt, MD 20771, USA
3
Science Systems and Applications, Inc., Lanham, MD 20706, USA
*
Author to whom correspondence should be addressed.
Remote Sens. 2025, 17(3), 482; https://doi.org/10.3390/rs17030482
Submission received: 26 November 2024
/
Revised: 10 January 2025
/
Accepted: 20 January 2025
/
Published: 30 January 2025
(This article belongs to the Special Issue Recent Developments in Remote Sensing Instruments, Technologies, and Results for Aerosol and Cloud Measurements)
Abstract
:Since its launch in 2018, the Ice, Cloud, and Land Elevation Satellite-2 (ICESat-2) mission has provided atmospheric products, including calibrated backscatter profiles and cloud and aerosol layer detection. While not the primary focus of the mission, these products garnered more interest after the end of Cloud–Aerosol Lidar with Orthogonal Polarization (CALIOP) data collection in 2023. In comparing the cloud and aerosol detection frequencies from CALIOP and ICESat-2, we find general agreement in the global patterns. The global cloud detection frequencies were similar in June, July, and August of 2019 (64.7% for ICESat-2 and 59.8% for CALIOP), as were the location and altitude of the tropical maximum; however, low daytime signal-to-noise ratios (SNRs) reduced ICESat-2’s detection frequencies compared to those of CALIOP. The ICESat-2 global aerosol detection frequencies were likewise lower. ICESat-2 generally retrieved a higher average global aerosol optical depth compared to the Moderate Resolution Imaging Spectroradiometer (MODIS) over the ocean, but the two were in closer agreement over regions with higher aerosol concentrations such as the Eastern Atlantic Ocean and the Northern Indian Ocean. The ICESat-2 and CALIOP orbital coincidences reveal highly correlated backscatter profiles as well as similar cloud and aerosol layer top altitudes. Future work with machine learning denoising techniques may allow for improved feature detection, especially during daytime.
1. Introduction
Clouds and aerosols play a significant role in the radiative balance of the planet and the broader Earth system. Based on their placement and composition, they can lead to either positive or negative direct radiative forcings [1] and can interact in important but uncertain secondary radiative effects and chemistry [2,3]. Because of these broad range of effects, lidar remote sensing instruments have been developed to capture the vertical profiles of clouds and aerosols as well as retrieve their optical properties. There were two recent NASA space-based missions designed to capture these measurements, called Cloud–Aerosol Lidar with Orthogonal Polarization (CALIOP) on the Cloud–Aerosol Lidar and Infrared Pathfinder Satellite Observations (CALIPSO) satellite (2006–2023) and the Cloud–Aerosol Transport System (CATS) on the International Space Station (2015–2017) [4,5]. Unfortunately, these two missions have stopped collecting data (2023 for CALIOP and 2017 for CATS). While the EarthCare mission [6] and its 355 nm atmospheric lidar (ATLID) have been recently launched in 2024 [7], the next generation of NASA space-based lidars remain on the distant horizon.
Launched in 2018, the Ice, Cloud, and Land Elevation Satellite-2 (ICESat-2) provides an avenue to continue the recording of lidar observations of clouds and aerosols from the CALIPSO and CATS missions to future space-based lidars. While the ICESat-2 mission [8] is primarily tasked with providing high-resolution altimetry data of the Earth surface and cryosphere, atmospheric products have been developed from ICESat-2 data to provide vertical profiles of clouds and aerosols [9]. An initial evaluation of the ICESat-2 atmospheric products found general agreement in the global cloud distributions as well as highly correlated backscatter profiles compared to those of CALIOP and the airborne Cloud Physics Lidar (CPL) [9].
Here, we evaluate the performance of current ICESat-2 atmospheric products in resolving the vertical and global distributions of clouds and aerosols compared to the CALIOP space-based lidar for a season of mission overlap (June, July, and August 2019). The aerosol optical depths during this time period were compared to those measured by the Moderate Resolution Imaging Spectroradiometer (MODIS). Using coincident points where the ICESat-2 and CALIPSO orbits intersected enable a case study evaluation of the ICESat-2 atmospheric algorithms in accurately detecting, typing, and determining the altitude of these features. Evaluating these ICESat-2 products is important as atmospheric data users look to existing and operating space-based missions to continue the 532 nm spaceborne lidar data record.
2. Methods
2.1. ATL09-v6 ICESat-2 Atmospheric Data
ICESat-2 was launched into a 92° inclination orbit in September of 2018 and has been in operation since October of that year [8,10]. Though specifically designed and optimized to obtain high-resolution altimetry measurements of the Earth’s surface, ICESat-2 also has an atmospheric channel to record backscatter from clouds and aerosols from the surface to 14 km above it. ICESat-2 carries only one instrument, the Advanced Topographic Laser Altimeter System (ATLAS), which utilizes a high repetition rate (10 kHz), low per pulse energy (500 J), 532 nm laser, and photon counting detectors. On board the satellite, 400 shots are summed and downlinked, resulting in 25 Hz profiles providing a 280 m horizontal and 30 m vertical resolution. ATLAS employs a diffractive optical element (DOE) to split the laser pulse into 6 individual beams that are simultaneously emitted from the satellite. Three of the beams have nominal energies of about 25 J per pulse (weak beams) and the other three have energies roughly 4 times those of the weak beams (strong beams). The altimetry measurements utilize all 6 laser beams while backscatter data are captured only from the 3 strong beams for the atmospheric measurements (known as profile1, profile2, and profile3 in the ATL04 and ATL09 data products). Each strong/weak beam pair is separated by about 3 km at the surface (across track).
The ICESat-2 atmospheric data products include level 2A (ATL04), 3A (ATL09), and 3B (ATL16/17) [11,12]. ATL04 and ATL09 are along-track products that include normalized relative backscatter profiles (ATL04) and calibrated, attenuated backscatter (CAB) profiles (ATL09) among many other parameters. ATL16 and ATL17 are weekly and monthly gridded fields of various parameters in the ATL09 product. In addition to the CAB profiles, ATL09 contains the top and bottom height of all atmospheric layers detected in the data and a flag to indicate whether each layer is a cloud, aerosol, or unknown. The ATL09 V6 products used in this analysis do not currently contain information on aerosol type or cloud phase. A Density Dimension Algorithm (DDA) is used to detect cloud and aerosol layers in the ATL09 products [9]. The DDA is an auto-adaptive algorithm that automatically adjusts to varying background conditions and uses a window roughly 12 shots (3 km) wide to locate regions of scattering that exceed the local background level. The DDA reports the top and bottom of layers at the native resolution of the atmospheric data (280 m and 25 Hz) [9,11]. Each detected layer is then categorized as cloud, aerosol, or unknown using characteristics of the layer including its height, relative humidity, and average backscatter magnitude. It should be noted that cloud–aerosol discrimination using only one wavelength is a difficult problem and that CALIPSO with its 2 wavelengths and depolarization channel can perform this more accurately. Also included in the ATL09 product are vertical profiles of temperature, pressure, and relative humidity obtained from the Goddard Earth Observing System Version 5 (GEOS-5) analysis or short period (<12 h) forecast.
Compared to CATS and CALIOP, ATLAS has some shortcomings [9]. Among these shortcomings, the atmospheric products are most constrained by the laser repetition rate and lack of linear depolarization. The lack of depolarization with the single 532 nm wavelength provides a challenge in distinguishing feature types and subtypes (i.e., cloud phase and aerosol type), as depolarization is a particle shape proxy and one of the primary measurements used in the CALIOP and CATS algorithms to determine cloud phase or aerosol type [13,14]. The high repetition rate (10 kHz) creates a challenge in computing the background and calibrating the ICESat-2 atmospheric data and limits the data frame to 14 km above the surface since the laser pulses are separated vertically by around 30 km [9]. With 30 km of separation, photons reflected from the surface will “meet” the photons from the successive pulse at 15 km and backscatter from that height will be measured by the detectors at the same time as the surface return. This complicates efforts to accurately determine the height of layers that are above 15 km and those in the lower altitudes of the profile. This effect, called folding, is flagged in the ATL09 product and is used to filter out folded clouds from certain analyses. Additionally, this abbreviated data frame poses challenges in the calibration of ICESat-2 atmospheric data since it provides no convenient calibration region where clouds or aerosols are not present, such as the upper atmosphere, and to background computation, which is ideally carried out very high in the atmosphere or below the surface [9].
2.2. Comparison Datasets
2.2.1. CALIOP
To enable comparisons between the ICESat-2 atmospheric products and those from CALIPSO and MODIS, global gridded maps of cloud and aerosol detection frequencies as well as gridded zonal profiles of cloud and aerosol detection frequencies were created. Global cloud and aerosol detection frequency maps and zonal profiles were created from CALIOP 5 km V4.20 L2 layer products and ICESat-2 ATL09 data. These data were aggregated into 1° × 1° global grids and 1° × 30 m zonal mean profiles. For the gridded maps, cloud (or aerosol) detection frequency is defined as the number of cloud (or aerosol) containing profiles divided by the number of profiles observed within the grid box.The zonal profile cloud (or aerosol) detection frequency is similarly defined as the number of cloud (or aerosol) detections within the 1° latitude × 30 m box divided by the number of valid profiles in the grid box. ICESat-2 profiles with a non-zero cloud folding flag are excluded from the mean zonal profiles. For the nighttime and daytime profiles and maps, we define nighttime (daytime) as where the solar elevation angle is less (more) than 0. In all the analyses, we used data from June, July, and August (JJA) of 2019, a period in which both ICESat-2 and CALIOP were operational.
2.2.2. MODIS
Aerosol optical depths (AODs) are a mainstay of MODIS products (e.g., [15,16,17,18]). These products provide global coverage and have been used to evaluate space-based lidar-derived AODs from CALIOP [19,20] and CATS [21] with good agreements. The ICESat-2 ATL09 product contains column optical depth (COD) over both land and ocean in the apparent surface reflectance (ASR) optical depth variable of ATL09. Because column optical depth retrievals over water are more accurate than those over land (the true ocean reflectance can be modeled), we only consider optical depths over ocean for ICESat-2 and MODIS. The COD product is based on the ratio of the measured surface reflectance to the modeled surface reflectance, which, over ocean, is obtained from the known relationship between 10 m wind speed and ocean surface reflectance [11].The ATL09 COD data contain attenuation due to both cloud and aerosol and are present for each lidar shot that contains a surface return. Surface returns can be detected up to a column optical depth of about 3–4.
To create a global grid of ICESat-2-derived AODs for comparison to MODIS AODs, we aggregated all JJA 2019 ICESat-2 ASR optical depths into a 1° × 1° global grid. Since ICESat-2 does not differentiate between cloud and aerosol optical depths, we excluded ICESat-2 profiles where the folding flag was nonzero and those in which clouds or “unknown” features were present according to the ATL09 cloud–aerosol discrimination. Cases in which the ICESat-2 optical depth was greater than 4 were also excluded from this analysis as the ICESat-2 signal would be fully attenuated and not reach the surface. The average optical depth was then found among these valid profiles.
To facilitate comparison between ICESat-2 optical depths to MODIS AOD, we aggregated MODIS MCD19A2 V6.1 data for the JJA 2019 season and found the average of 0.55 m AOD over the 1° × 1° cells. The MCD19A2 data are created at a 1 × 1 km resolution using merged observations from both the Terra and Aqua satellites. The MODIS 0.55 m AOD was chosen in this analysis as it is closest to the 532 nm wavelength of ICESat-2. Both MODIS AOD and the ICESat-2 derived AOD represent the total column optical depth.
2.3. ICESat-2 and CALIOP Orbital Coincident Points
Though ICESat-2 and CALIPSO are in different orbits (the former precessing and the latter sun-synchronous), the two satellite orbits often cross. Sometimes, the orbit crossings occur within only a few minutes or, in some rare cases, seconds of each other. The data from these coincidences can then be used for comparison of CAB profiles to evaluate calibration, layer detection (presence, type, and height), and column optical depth.
In this paper, we analyze orbital coincidences that occurred in 2019 and 2020. Coincidences were identified using latitude, longitude, and time from the ICESat-2 ATL09 product as well as the analogous fields from CALIPSO L1B calibrated backscatter products. All orbit crossings that occurred within 10 min were recorded in a file noting the location, time, and the respective ATL09 and CALIPSO granule names. While most coincidences occur at high latitudes, tropical and midlatitude coincidences were also identified. These cases are listed in Table A1.
3. Results
In the following section, we detail comparisons between the ICESat-2 atmospheric products and those from the CALIOP space-based lidar for clouds (Section 3.1) and aerosols (Section 3.2). The aerosol section also includes optical depth comparisons to the MODIS-derived column aerosol optical depths.
3.1. Clouds
Globally, ICESat-2 cloud detection agrees well with analogous CALIOP products. Figure 1 shows global comparisons for June, July, and August (JJA) 2019, a year of overlap between the ICESat-2 and CALIPSO missions. The general patterns are very similar, with cloud detection maximums in the tropics associated with the inter-tropical convergence zone (ITCZ) and over the mid- to high-latitude oceans. Globally averaged, the differences between the ICESat-2 and CALIOP cloud detection frequencies are small, with ICESat-2 detecting clouds somewhere in the vertical column 64.7% of the time and CALIOP detecting clouds 59.8% of the time in JJA 2019 between 80°N and S, latitudes observed by both lidars. The cloud detection frequency for 2019 JJA is close to the 67% ICESat-2 cloud fraction for May 2019 [9].
Over certain regions, there are more significant differences between the ICESat-2 and CALIOP cloud detection frequencies. Clouds are observed by ICESat-2 more frequently over the tropical ocean. In these regions, differences between the detection frequencies can be up to 50 percentage points. In contrast, clouds are more frequently observed by CALIOP in polar regions with CALIOP and ICESat-2 observing clouds around 50–60% and 10–20% of the time over Antarctica, respectively. This discrepancy, however, is due to the inclusion of “diamond dust” as a feature type in ATL09 [9]. During this time period, ICESat-2 detected diamond dust in upwards of 75% of Antarctic profiles (Figure A6). Over South America, ICESat-2 detects clouds less frequently than CALIOP. These differences are due to the degraded signal quality over this region as a consequence of the South Atlantic Anomaly (SAA). In the SAA region, which includes most of South America and extends eastward to roughly longitude 10°W, charged particles at the altitude of the satellite cause detector noise that adversely affects signal quality and calibration. This, in turn, can make the backscatter associated with a cloud look more like that of an aerosol. As a consequence, ICESat-2 often mistakes clouds for aerosols, resulting in higher aerosol frequencies than CALIOP in this region.
The zonal average profiles for the same time period reveal similar patterns (Figure 2). ICESat-2 and CALIOP both observe maximum cloud fractions in the high-altitude tropics, which corresponds to the ITCZ in June, July, and August. Other noteworthy agreements are with the altitudes of marine clouds, especially over the Southern Ocean (from ∼65°S to 50°S) and 70°N north. Overall, ICESat-2 observes clouds less frequently throughout the vertical column compared to CALIOP, which tends to have thicker cloud layers (Figure A1). For example, at 10°N, CALIOP observes clouds at 13 km in altitude around 34% of the time compared to 20% with ICESat-2. The lack of cloud detection above 14 km in Figure 2a is due to the limited ICESat-2 vertical data frame.
Part of the reason for the higher cloud detection frequencies in CALIOP compared to ICESat-2 is the higher daytime SNR of CALIOP and differences in the horizontal averaging used to detect layers (CALIOP uses 5, 20, and 80 km while ICESat-2 uses 3 km). Tenuous layers can be missed in these noisy daytime conditions. When considering only nighttime scenes, cloud detection frequencies in the peak region of 10°N and 13 km in altitude increase to 30% in ICESat-2 and to 39% in CALIOP (Figure A2). In the daytime, ICESat-2 observes clouds here less than 11% of the time compared to ∼29% with CALIOP (Figure A3). It should also be noted that ICESat-2 and CALIPSO’s orbits provide different diurnal sampling, with CALIOP having consistent equatorial crossings of ∼1:30 a.m. and p.m. and ICESat-2 having variable crossing times. Diurnal cycles in cloud and aerosol detection have been noted and explored with CATS (e.g., [22,23]) and the temporal sampling facilitated by the CALIPSO orbit does not necessarily coincide with the observed daily maximums/minimums [23].
In contrast, one region where ICESat-2 observes more clouds than CALIOP is in the lower tropical troposphere (1–4 km). However, these low-altitude clouds are likely either miscategorized aerosol layers or unflagged cases of the folding effect discussed earlier. CALIOP’s high-altitude layer detection frequency (>15 km) is close to a factor of 2 times greater at 15°N than the ICESat-2 folding detection frequency (Figure 3). Even considering the differences in diurnal sampling, ICESat-2 folding detection is likely under-reported globally, which is an important consideration in regions with frequent high-altitude features, such as the tropics and wintertime polar regions.
3.2. Aerosols
In contrast to the general agreement between ICESat-2 and CALIOP in global patterns of cloud detection frequency, there is more disagreement between the two lidar datasets in the aerosol detection frequencies and their global patterns. Overall, ICESat-2 detects aerosols less frequently than CALIOP. In JJA 2019, the global average aerosol detection frequency is 28.9% for ICESat-2 and 53.2% for CALIOP for 80°N and S (Figure 3). This difference is especially apparent over the main dust source regions of the Middle East and North Africa, where ICESat-2 observes aerosols only ∼30–60% of the time compared to >80% in CALIOP for JJA 2019. Part of this disagreement in magnitude is due to the lower ICESat-2 daytime SNR preventing aerosol detection. Over bright, reflective regions, such as deserts, this SNR can be reduced even further. With a lower SNR, more tenuous aerosol layers are not detected. In many marine domains, ICESat-2 and CALIOP see similar aerosol detection patterns but with a disagreement in magnitude.
The largest disagreement between CALIOP and ICESat-2 aerosol frequencies are observed over South America, where ICESat-2 detects aerosols 20–40% more frequently (Figure 4). As discussed in Section 3.1, this region coincides with a region of lower ICESat-2 cloud detection compared to CALIOP. Thus, many of the “missing” ICESat-2 cloud layers in Figure 1 are likely miscategorized aerosol layers. As discussed in Section 2, the high frequency of aerosol detection over South America is due to the signal degradation caused by the SAA. Furthermore, CALIOP aerosol detections are less accurate in this region due to the SAA. The added detector noise in this region for both lidar systems causes a large calibration error, which, in turn, affects the cloud–aerosol discrimination [24]. Even outside the SAA, differences between ICESat-2 and CALIOP in cloud or aerosol detection and typing is not unexpected considering the hardware limitations, specifically the lack of depolarization, with ICESat-2.
The lower ICESat-2 aerosol detection frequencies are also apparent in the zonal mean profiles when compared to their CALIOP analog for JJA 2019 (Figure 5). ICESat-2 and CALIOP both identify similar maxima (around 10°S and 20°N). The 10°S maximum corresponds to the latitudes of smoke outflow from Southern Africa over the Atlantic and marine aerosols over the remote Pacific and Indian Oceans. The 20°N maximum corresponds to marine aerosols and to Middle Eastern/Saharan dust and its transport over the North Atlantic.
Differences in the magnitude in aerosol detection frequencies arise from a few instrument/algorithm differences. As previously discussed, ICESat-2 has a much lower daytime SNR than CALIOP, so more tenuous layers, such as many aerosol layers, may remain undetected. Evidence of this can be seen in Figure 5, with the difference in the aerosol detection frequency over the latitudes associated with dust transport ∼20°N, with ICESat-2 observing much lower aerosol detection frequencies than CALIOP. When only considering nighttime ICESat-2 observations, the aerosol detection frequencies increase to ∼30% in the aerosol detection maxima ∼10°S and 20°N, which is closer but still less than those of CALIOP (Figure 5 and Figure A4). Differences in the higher-altitude aerosol detection frequency (above 8 km) arise due to algorithm differences with the ICESat-2 cloud–aerosol discrimination algorithm favoring clouds above this altitude. The elevated aerosol detection frequency in the high-altitude, northern latitudes in Figure 4b is associated with the Raikoke (48.3°N, 153.2°E) volcanic eruption in June 2019 (e.g., [25,26]).
3.3. Optical Depths
The global patterns of the total column AODs are similar in MODIS and ICESat-2 in JJA 2019 (Figure 6). The main dust and smoke outflow regions over the Eastern Atlantic are similarly captured with elevated AODs in these regions as well as the Northern Indian Ocean. ICESat-2, however, has higher optical depths over the remote ocean, leading to higher average AODs (0.25 for ICESat-2 and 0.16 for MODIS between 50°N and S). Differences between day and night ICESat-2 AODs are small in this domain during this time period (0.26 for day and 0.23 for night). The MODIS AOD is similar to the global average AOD found by Ma et al. [19] (0.148 and 0.140 for Terra and Aqua, respectively).
Part of this disagreement between the ICESat-2 and MODIS optical depths may stem from cloud contamination in the ICESat-2 optical depths. Figure 6 illustrates that many of the regions featuring higher ICESat-2 optical depths, such as the tropical Western Pacific and Northern Pacific Ocean, are some of the cloudiest regions (Figure 1). Because of the hardware limitation of ATLAS, including the lack of depolarization, differentiating between clouds and aerosols is more challenging with ICESat-2 than with CALIOP.
Closer to aerosol source regions, where aerosols are more ubiquitous and optically thick, the ICESat-2 and MODIS AODs tend to agree better. In a domain covering the Eastern Atlantic and Northern Indian Oceans (red box in Figure 6a,b), the mean ICESat-2 and MODIS AODs are very similar (0.34 for ICESat-2 and 0.30 for MODIS) for JJA 2019. This is a closer agreement compared to the global AOD averages (50°N to 50°S) of 0.25 for ICESat-2 and 0.16 for MODIS.
3.4. Coincident ICESat-2/CALIOP Observations
In the period after ICESat-2 started its observations in October 2018 and before CALIOP ended its data collection in July 2023, the orbits of the two instruments crossed many times. These crosses, when sufficiently close in time, provide opportunities to directly compare the ICESat-2 atmospheric products to those of CALIOP.
In the coincident cases we have identified (Table A1), the ICESat-2 and CALIOP profiles agree well. An example of this agreement is shown in Figure 7, where ICESat-2 and CALIOP both observe the same dust aerosol layer over the Middle East. Cases where high-altitude clouds were present and were folded into the data frame are notable exceptions. An illustration of this is presented in Figure 8. Daytime coincidences, when the ICESat-2 and CALIOP SNR is lower, presented much noisier profiles; however, the backscatter profiles of the two instruments remained similar. Among the 38 coincident profiles, the 10 s averaged backscatter profiles around the coincident point were highly correlated (Table A1). The median correlation coefficient of these profiles is 0.80 when excluding portions of the profile below and within two bins (60 m) of the ICESat-2 digital elevation model (DEM) surface altitude and where the total attenuated backscatter is greater than 0.2 km−1 sr−1.
Often, cloud or aerosol layers were present at the orbital coincident point. In the 38 orbital coincidences, we found 26 cases where a cloud or aerosol layer was detected by both ICESat-2 and CALIOP at the orbital coincident point. In most of these cases, the ICESat-2 and CALIOP layer detection algorithms identified nearly identical cloud or aerosol layer top altitudes. The small differences observed are generally within a vertical bin (30 m) or two (Figure 9a). For the example outlined in Figure 7, the dust aerosol layer top was detected at 6.46 km with CALIOP and 5.40 km. The ICESat-2 and CALIOP layer base altitudes were very similar but disagreed somewhat more than the layer tops, with ICESat-2 generally having higher layer base altitudes than CALIOP (Figure 9b). The layers are generally thicker in the CALIOP L2 data products (Figure A1). The thicker layers in CALIOP may occur due to hardware differences—CALIOP features more energetic pulses, allowing for greater penetration of the cloud or aerosol layers. Alternatively, these differences arise from their feature identification algorithms.
4. Conclusions
The current suite of atmospheric cloud products for ICESat-2, especially during nighttime hours, has the potential to fill some of the voids left by the retirement of CALIOP and compliment the new EarthCARE data products. ICESat-2 ATL09 products currently provide CALIOP-like nighttime backscatter profiles for the midlatitudes, polar regions, and other regions without high-altitude clouds or aerosols. The optical depths derived from ICESat-2 data compare favorably to those retrieved by MODIS, especially in regions of higher aerosol loading. Cloud detection frequency maps and mean zonal profiles feature similar global patterns. When cloud and aerosol layers are detected by the current ICESat-2 atmospheric data product algorithms, the ICESat-2 layer top altitudes are nearly identical to those measured by CALIOP.
Users should be aware of certain limitations in the ICESat-2 atmospheric products. Comparisons of the ICESat-2 and CALIOP zonal mean profiles indicate that many higher and optically thinner clouds are detected at lower frequencies compared to CALIOP, especially during the daytime, when ICESat-2 has the lowest signal-to-noise ratio. The aerosol detection frequencies are much lower in ICESat-2 than CALIOP, especially during the daytime over more reflective surface types. While this analysis only considered JJA 2019, we would expect similar issues in other seasons. Cloud and aerosol detection in higher altitudes (>15 km) is not currently possible, with the ATLAS laser repetition rate leading to the folding of high-altitude features into the ICESat-2 data frame. This is not a major consideration in the mid- and high latitudes, where clouds and aerosols are not common at these altitudes, but are an important consideration in the tropics where tropopause altitudes are high and clouds above the ICESat-2 data frame are common. The abbreviated data frame is also a concern after the stratospheric injection of aerosols from volcanoes or pyrocumulonimubus (pyroCb) and in the wintertime polar regions, where polar stratospheric clouds can occur.
There are ample avenues for future improvements in the ICESat-2 atmospheric products. Machine learning techniques can drastically reduce noise and improve feature detection [27,28,29]. These machine learning techniques would be especially valuable in the daytime data where the current signal-to-noise ratios are low and prevent the identification of optically thin layers. Distinguishing between feature subtypes, such as smoke from dust aerosols, using the heritage methods employed by CALIOP and CATS is difficult with ICESat-2 data due, in part, to the lack of depolarization. Additional information provided by model reanalysis, such as that through MERRA-2, may make subtyping possible. Identifying the cloud phase as well as aerosol type of layers would allow for the assignment of lidar ratios and the calculation of layer optical depths, which may improve the ICESat-2 optical depths. Given the dearth of lidar missions currently operating in space with the ability to provide profiles of atmospheric features, these ICESat-2 atmospheric products provide an opportunity to continue the recording of space-based observations of clouds and aerosols before the next generation of atmosphere-focused lidars reach orbit.
Author Contributions
Conceptualization, K.E.C. and S.P.P.; methodology, K.E.C., S.P.P., J.E.Y. and E.P.N.; formal analysis, K.E.C.; writing—original draft preparation, K.E.C.; writing—review and editing, K.E.C., J.E.Y., E.P.N. and S.P.P.; visualization, K.E.C.; funding acquisition, K.E.C., S.P.P., J.E.Y. and E.P.N. All authors have read and agreed to the published version of the manuscript.
Funding
This work was funded by NASA grant 80NSSC23K0932.
Data Availability Statement
All data used in this project are freely available from NASA: https://search.earthdata.nasa.gov/, accessed on 12 August 2021.
Acknowledgments
We would like to thank the three anonymous reviewers for their thoughtful input and constructive comments.
Conflicts of Interest
Author Stephen P. Palm is employed by Science Systems and Applications, Inc. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Appendix A
Table A1.
Coincident ICESat-2 and CALIOP observations.
| ICESat-2, CALIOP L1, and CALIOP L2 File Names | Correlation Coeff. | Day/Night | Time Offset (seconds) | Coincident Point |
|---|---|---|---|---|
| ATL09_20190104182603_01120201_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2019-01-04T18-45-16ZN.hdf | 0.42 | N | 34 | 77.76°N 116.26°W |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2019-01-04T18-45-16ZN.hdf | ||||
| ATL09_20190109065200_01810201_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2019-01-09T07-08-41ZN.hdf | 0.79 | N | 121 | 78.76°N 53.07°E |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2019-01-09T07-08-41ZN.hdf | ||||
| ATL09_20190119230109_03440201_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2019-01-19T23-20-35ZN.hdf | 0.65 | N | 23 | 80.04°N 161.66°E |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2019-01-19T23-20-35ZN.hdf | ||||
| ATL09_20190201031917_05300201_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2019-02-01T03-40-34ZN.hdf | 0.83 | N | 96 | 80.97°N 87.20°E |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2019-02-01T03-40-34ZN.hdf | ||||
| ATL09_20190205154517_05990201_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2019-02-05T16-04-13ZN.hdf | 0.51 | N | 62 | 81.26°N 102.99°W |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2019-02-05T16-04-13ZN.hdf | ||||
| ATL09_20190211175409_06920201_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2019-02-11T18-14-18ZN.hdf | 0.78 | N | 16 | 81.50°N 140.00°W |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2019-02-11T18-14-18ZN.hdf | ||||
| ATL09_20190216062009_07610201_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2019-02-16T06-37-57ZN.hdf | 0.95 | N | 97 | 81.66°N 29.98°E |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2019-02-16T06-37-57ZN.hdf | ||||
| ATL09_20190527135146_09060301_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2019-05-27T14-17-55ZN.hdf | 0.96 | N | 134 | 76.67°S 19.12°E |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2019-05-27T14-17-55ZN.hdf | ||||
| ATL09_20190720203745_03480401_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2019-07-20T21-20-17ZD.hdf | 0.87 | D | 62 | 28.90°S 110.72°W |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2019-07-20T21-20-17ZD.hdf | ||||
| ATL09_20190720221203_03490401_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2019-07-20T22-12-27ZN.hdf | 0.87 | N | 15 | 30.17°N 57.31°E |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2019-07-20T22-12-27ZN.hdf | ||||
| ATL09_20190722102043_03720401_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2019-07-22T11-06-21ZD.hdf | 0.83 | D | 161 | 28.05°S 42.51°E |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2019-07-22T11-06-21ZD.hdf | ||||
| ATL09_20190723205507_03940401_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2019-07-23T20-49-06ZN.hdf | 0.70 | N | 21 | 19.76°N 75.56°E |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2019-07-23T20-49-06ZN.hdf | ||||
| ATL09_20190725055513_04150401_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2019-07-25T06-26-00ZD.hdf | 0.81 | D | 175 | 11.15°S 108.65°E |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2019-07-25T06-26-00ZD.hdf | ||||
| ATL09_20190725103805_04180401_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2019-07-25T10-35-16ZN.hdf | 0.88 | N | 204 | 18.84°N 131.20°W |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2019-07-25T10-35-16ZN.hdf | ||||
| ATL09_20190726211228_04400401_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2019-07-26T21-04-20ZN.hdf | 0.97 | N | 131 | 11.35°N 69.85°E |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2019-07-26T21-04-20ZN.hdf | ||||
| ATL09_20190728092109_04630401_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2019-07-28T09-12-00ZN.hdf | 0.71 | N | 207 | 7.17°N 112.97°W |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2019-07-28T09-12-00ZN.hdf | ||||
| ATL09_20190729151240_04820401_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2019-07-29T16-23-59ZN.hdf | 0.41 | N | 165 | 6.80°S 136.04°E |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2019-07-29T16-23-59ZN.hdf | ||||
| ATL09_20190801135543_05270401_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2019-08-01T15-00-39ZN.hdf | 0.80 | N | 171 | 18.20°S 154.33°E |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2019-08-01T15-00-39ZN.hdf | ||||
| ATL09_20190915181541_12170401_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2019-09-15T18-47-27ZN.hdf | 0.55 | N | 101 | 73.52°S 66.73°E |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2019-09-15T18-47-27ZN.hdf | ||||
| ATL09_20190918183301_12630401_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2019-09-18T19-02-31ZN.hdf | 0.98 | N | 184 | 74.47°S 60.75°E |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2019-09-18T19-02-31ZN.hdf | ||||
| ATL09_20190929121633_00400501_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2019-09-29T12-50-41ZN.hdf | 0.97 | N | 233 | 76.30°S 148.32°E |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2019-09-29T12-50-41ZN.hdf | ||||
| ATL09_20191005142530_01330501_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2019-10-05T14-59-15ZN.hdf | 0.83 | N | 288 | 77.31°S 112.46°E |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2019-10-05T14-59-15ZN.hdf | ||||
| ATL09_20191020172626_03640501_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2019-10-20T18-38-44ZD.hdf | 0.01 | D | 31 | 79.36°S 58.33°E |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2019-10-20T18-38-44ZD.hdf | ||||
| ATL09_20191216005530_12240501_006_01.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2019-12-16T01-18-44ZN.hdf | 0.45 | N | 112 | 81.82°N 101.01°E |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2019-12-16T01-18-44ZN.hdf | ||||
| ATL09_20200402032743_01010701_006_03.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2020-04-02T04-02-09ZN.hdf | 0.84 | N | 438 | 18.89°N 31.93°W |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2020-04-02T04-02-09ZN.hdf | ||||
| ATL09_20200405034504_01470701_006_03.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2020-04-05T04-17-39ZN.hdf | 0.91 | N | 328 | 4.03°S 40.68°W |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2020-04-05T04-17-39ZN.hdf | ||||
| ATL09_20200406141927_01690701_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2020-04-06T13-54-33ZD.hdf | 0.96 | D | 96 | 28.26°N 11.15°W |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2020-04-06T13-54-33ZD.hdf | ||||
| ATL09_20200406155344_01700701_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2020-04-06T15-33-09ZD.hdf | 0.52 | D | 150 | 20.33°N 33.86°W |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2020-04-06T15-33-09ZD.hdf | ||||
| ATL09_20200406172801_01710701_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2020-04-06T17-11-39ZD.hdf | 0.34 | D | 397 | 12.03°N 56.62°W |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2020-04-06T17-11-39ZD.hdf | ||||
| ATL09_20200409174521_02170701_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2020-04-09T17-27-13ZD.hdf | 0.72 | D | 288 | 32.45°N 65.29°W |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2020-04-09T17-27-13ZD.hdf | ||||
| ATL09_20200411041944_02390701_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2020-04-11T03-56-28ZD.hdf | 0.35 | D | 19 | 45.42°N 133.42°E |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2020-04-11T03-56-28ZD.hdf | ||||
| ATL09_20200412162824_02620701_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2020-04-12T16-56-33ZN.hdf | 0.94 | N | 47 | 48.67°S 117.99°E |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2020-04-12T16-56-33ZN.hdf | ||||
| ATL09_20200414043704_02850701_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2020-04-14T05-04-18ZN.hdf | 0.42 | N | 12 | 53.42°S 65.99°W |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2020-04-14T05-04-18ZN.hdf | ||||
| ATL09_20200415182001_03090701_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2020-04-15T17-58-18ZD.hdf | 0.57 | D | 54 | 55.93°N 81.48°W |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2020-04-15T17-58-18ZD.hdf | ||||
| ATL09_20200417062841_03320701_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2020-04-17T06-06-02ZD.hdf | 0.77 | D | 6 | 59.48°N 94.55°E |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2020-04-17T06-06-02ZD.hdf | ||||
| ATL09_20200418183721_03550701_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2020-04-18T19-06-07ZN.hdf | 0.81 | N | 59 | 61.17°S 79.21°E |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2020-04-18T19-06-07ZN.hdf | ||||
| ATL09_20200420064601_03780701_006_02.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2020-04-20T07-13-52ZN.hdf | 0.93 | N | 3 | 63.68°S 104.61°W |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2020-04-20T07-13-52ZN.hdf | ||||
| ATL09_20200505125531_06110701_006_01.h5 | ||||
| CAL_LID_L1-Standard-V4-10.2020-05-05T13-26-40ZN.hdf | 0.88 | N | 137 | 74.13°S 148.39°E |
| CAL_LID_L2_05kmMLay-Standard-V4-20.2020-05-05T13-26-40ZN.hdf |
Figure A1.
Average cumulative column cloud layer thickness for (a) ICESat-2 and (b) CALIOP (JJA 2019).
Figure A1.
Average cumulative column cloud layer thickness for (a) ICESat-2 and (b) CALIOP (JJA 2019).
Figure A2.
Zonal mean profiles of nighttime JJA 2019 cloud detection frequency for (a) ICESat-2 and (b) CALIOP, and (c) the difference between ICESat-2 and CALIOP (a,b) in percentage points.
Figure A2.
Zonal mean profiles of nighttime JJA 2019 cloud detection frequency for (a) ICESat-2 and (b) CALIOP, and (c) the difference between ICESat-2 and CALIOP (a,b) in percentage points.
Figure A3.
Zonal mean profiles of daytime JJA 2019 cloud detection frequency for (a) ICESat-2 and (b) CALIOP, and (c) the difference between ICESat-2 and CALIOP (a,b) in percentage points.
Figure A3.
Zonal mean profiles of daytime JJA 2019 cloud detection frequency for (a) ICESat-2 and (b) CALIOP, and (c) the difference between ICESat-2 and CALIOP (a,b) in percentage points.
Figure A4.
Zonal mean profiles of nighttime JJA 2019 aerosol detection frequency for (a) ICESat-2 and (b) CALIOP, and (c) the difference between ICESat-2 and CALIOP (a,b) in percentage points.
Figure A4.
Zonal mean profiles of nighttime JJA 2019 aerosol detection frequency for (a) ICESat-2 and (b) CALIOP, and (c) the difference between ICESat-2 and CALIOP (a,b) in percentage points.
Figure A5.
Zonal mean profiles of daytime JJA 2019 aerosol detection frequency for (a) ICESat-2 and (b) CALIOP, and (c) the difference between ICESat-2 and CALIOP (a,b) in percentage points.
Figure A5.
Zonal mean profiles of daytime JJA 2019 aerosol detection frequency for (a) ICESat-2 and (b) CALIOP, and (c) the difference between ICESat-2 and CALIOP (a,b) in percentage points.
Figure A6.
ICESat-2 diamond dust detection frequency (JJA 2019).
References
- Boucher, O.; Randall, D.; Artaxo, P.; Bretherton, C.; Feingold, G.; Forster, P.; Kerminen, V.M.; Kondo, Y.; Liao, H.; Lohmann, U.; et al. Clouds and Aerosols. 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; Cambridge University Press: Cambridge, UK, 2013; pp. 571–657. [Google Scholar]
- Ackerman, A.S.; Toon, O.B.; Stevens, D.E.; Heymsfield, A.J.; Ramanathan, V.; Welton, E.J. Reduction of Tropical Cloudiness by Soot. Science 2000, 288, 1042–1047. [Google Scholar] [CrossRef] [PubMed]
- Crutzen, P.J.; Andreae, M.O. Biomass Burning in the Tropics: Impact on Atmospheric Chemistry and Biogeochemical Cycles. Science 1990, 250, 1669–1678. [Google Scholar] [CrossRef] [PubMed]
- Winker, D.M.; Vaughan, M.A.; Omar, A.; Hu, Y.; Powell, K.A.; Liu, Z.; Hunt, W.H.; Young, S.A. Overview of the CALIPSO Mission and CALIOP Data Processing Algorithms. J. Atmos. Ocean. Technol. 2009, 26, 2310–2323. [Google Scholar] [CrossRef]
- McGill, M.J.; Yorks, J.E.; Scott, V.S.; Kupchock, A.W.; Selmer, P.A. The Cloud-Aerosol Transport System (CATS): A technology demonstration on the International Space Station. In Proceedings of the Lidar Remote Sensing for Environmental Monitoring XV, San Diego, CA, USA, 12–13 August 2015; International Society for Optics and Photonics: Bellingham, WA, USA, 2015; Volume 9612, p. 96120A. [Google Scholar] [CrossRef]
- Illingworth, A.J.; Barker, H.W.; Beljaars, A.; Ceccaldi, M.; Chepfer, H.; Clerbaux, N.; Cole, J.; Delanoe, J.; Domenech, C.; Donovan, D.P.; et al. The EarthCARE Satellite: The Next Step Forward in Global Measurements of Clouds, Aerosols, Precipitation, and Radiation. Bull. Am. Meteorol. Soc. 2015, 96, 1311–1332. [Google Scholar] [CrossRef]
- Donovan, D.P.; van Zadelhoff, G.J.; Wang, P. The EarthCARE lidar cloud and aerosol profile processor (A-PRO): The A-AER, A-EBD, A-TC, and A-ICE products. Atmos. Meas. Tech. 2024, 17, 5301–5340. [Google Scholar] [CrossRef]
- Markus, T.; Neumann, T.; Martino, A.; Abdalati, W.; Brunt, K.; Csatho, B.; Farrell, S.; Fricker, H.; Gardner, A.; Harding, D.; et al. The Ice, Cloud, and land Elevation Satellite-2 (ICESat-2): Science requirements, concept, and implementation. Remote Sens. Environ. 2017, 190, 260–273. [Google Scholar] [CrossRef]
- Palm, S.P.; Yang, Y.; Herzfeld, U.; Hancock, D.; Hayes, A.; Selmer, P.; Hart, W.; Hlavka, D. ICESat-2 Atmospheric Channel Description, Data Processing and First Results. Earth Space Sci. 2021, 8, e2020EA001470. [Google Scholar] [CrossRef]
- Abdalati, W.; Zwally, H.J.; Bindschadler, R.; Csatho, B.; Farrell, S.L.; Fricker, H.A.; Harding, D.; Kwok, R.; Lefsky, M.; Markus, T.; et al. The ICESat-2 Laser Altimetry Mission. Proc. IEEE 2010, 98, 735–751. [Google Scholar] [CrossRef]
- Palm, S.; Yang, Y.; Herzfeld, U.; Hancock, D. ICESat-2 Algorithm Theoretical Basis Document for the Atmosphere, Part I: Level 2 and 3 Data Products; Version 6; National Aeronautics and Space Administration, Goddard Space Flight Center: Glenn Dale, MD, USA, 2022.
- Northam, E.T.; Palm, S.P. Ice, Cloud, and land Elevation Satellite-2 (ICESat-2) Project Algorithm Theoretical Basis Document (ATBD) for Atmosphere Gridded Products; Version 5; National Aeronautics and Space Administration, Goddard Space Flight Center: Glenn Dale, MD, USA, 2022.
- Kim, M.H.; Omar, A.H.; Tackett, J.L.; Vaughan, M.A.; Winker, D.M.; Trepte, C.R.; Hu, Y.; Liu, Z.; Poole, L.R.; Pitts, M.C.; et al. The CALIPSO version 4 automated aerosol classification and lidar ratio selection algorithm. Atmos. Meas. Tech. 2018, 11, 6107–6135. [Google Scholar] [CrossRef]
- Yorks, J.E.; McGill, M.J.; Palm, S.P.; Hlavka, D.L.; Selmer, P.A.; Nowottnick, E.P.; Vaughan, M.A.; Rodier, S.D.; Hart, W.D. An overview of the CATS level 1 processing algorithms and data products. Geophys. Res. Lett. 2016, 43, 4632–4639. [Google Scholar] [CrossRef]
- Lyapustin, A.; Wang, Y.; Korkin, S.; Huang, D. MODIS Collection 6 MAIAC algorithm. Atmos. Meas. Tech. 2018, 11, 5741–5765. [Google Scholar] [CrossRef]
- Remer, L.A.; Kaufman, Y.J.; Tanré, D.; Mattoo, S.; Chu, D.A.; Martins, J.V.; Li, R.R.; Ichoku, C.; Levy, R.C.; Kleidman, R.G.; et al. The MODIS Aerosol Algorithm, Products, and Validation. J. Atmos. Sci. 2005, 62, 947–973. [Google Scholar] [CrossRef]
- Levy, R.C.; Remer, L.A.; Mattoo, S.; Vermote, E.F.; Kaufman, Y.J. Second-generation operational algorithm: Retrieval of aerosol properties over land from inversion of Moderate Resolution Imaging Spectroradiometer spectral reflectance. J. Geophys. Res. Atmos. 2007, 112. [Google Scholar] [CrossRef]
- Levy, R.C.; Remer, L.A.; Kleidman, R.G.; Mattoo, S.; Ichoku, C.; Kahn, R.; Eck, T.F. Global evaluation of the Collection 5 MODIS dark-target aerosol products over land. Atmos. Chem. Phys. 2010, 10, 10399–10420. [Google Scholar] [CrossRef]
- Ma, X.; Bartlett, K.; Harmon, K.; Yu, F. Comparison of AOD between CALIPSO and MODIS: Significant differences over major dust and biomass burning regions. Atmos. Meas. Tech. 2013, 6, 2391–2401. [Google Scholar] [CrossRef]
- Ryan, R.A.; Vaughan, M.A.; Rodier, S.D.; Tackett, J.L.; Reagan, J.A.; Ferrare, R.A.; Hair, J.W.; Smith, J.A.; Getzewich, B.J. Total column optical depths retrieved from CALIPSO lidar ocean surface backscatter. Atmos. Meas. Tech. 2024, 17, 6517–6545. [Google Scholar] [CrossRef]
- Lee, L.; Zhang, J.; Reid, J.S.; Yorks, J.E. Investigation of CATS aerosol products and application toward global diurnal variation of aerosols. Atmos. Chem. Phys. 2019, 19, 12687–12707. [Google Scholar] [CrossRef]
- Noel, V.; Chepfer, H.; Chiriaco, M.; Yorks, J. The diurnal cycle of cloud profiles over land and ocean between 51°S and 51°N, seen by the CATS spaceborne lidar from the International Space Station. Atmos. Chem. Phys. 2018, 18, 9457–9473. [Google Scholar] [CrossRef]
- Nowottnick, E.P.; Christian, K.E.; Yorks, J.E.; McGill, M.J.; Midzak, N.; Selmer, P.A.; Lu, Z.; Wang, J.; Salinas, S.V. Aerosol Detection from the Cloud–Aerosol Transport System on the International Space Station: Algorithm Overview and Implications for Diurnal Sampling. Atmosphere 2022, 13, 1439. [Google Scholar] [CrossRef]
- Noel, V.; Chepfer, H.; Hoareau, C.; Reverdy, M.; Cesana, G. Effects of solar activity on noise in CALIOP profiles above the South Atlantic Anomaly. Atmos. Meas. Tech. 2014, 7, 1597–1603. [Google Scholar] [CrossRef]
- Crafford, A.; Venzke, E. Global Volcanism Program Report on Raikoke (Russia). Bull. Glob. Volcanism Netw. Smithson. Inst. 2019, 44. [Google Scholar] [CrossRef]
- De Leeuw, J.; Schmidt, A.; Witham, C.S.; Theys, N.; Taylor, I.A.; Grainger, R.G.; Pope, R.J.; Haywood, J.; Osborne, M.; Kristiansen, N.I. The 2019 Raikoke volcanic eruption—Part 1: Dispersion model simulations and satellite retrievals of volcanic sulfur dioxide. Atmos. Chem. Phys. 2021, 21, 10851–10879. [Google Scholar] [CrossRef]
- Yorks, J.E.; Selmer, P.A.; Kupchock, A.; Nowottnick, E.P.; Christian, K.E.; Rusinek, D.; Dacic, N.; McGill, M.J. Aerosol and Cloud Detection Using Machine Learning Algorithms and Space-Based Lidar Data. Atmosphere 2021, 12, 606. [Google Scholar] [CrossRef]
- Selmer, P.; Yorks, J.E.; Nowottnick, E.P.; Cresanti, A.; Christian, K.E. A Deep Learning Lidar Denoising Approach for Improving Atmospheric Feature Detection. Remote Sens. 2024, 16, 2735. [Google Scholar] [CrossRef]
- Oladipo, B.; Gomes, J.; McGill, M.; Selmer, P. Leveraging Deep Learning as a New Approach to Layer Detection and Cloud–Aerosol Classification Using ICESat-2 Atmospheric Data. Remote Sens. 2024, 16, 2344. [Google Scholar] [CrossRef]
Figure 1.
Global 1° × 1° JJA 2019 cloud detection frequency (CDF) maps for (a) ICESat-2 and (b) CALIOP, and (c) the difference between ICESat-2 and CALIOP (CDF of ICESat-2 − CDF of CALIOP).
Figure 1.
Global 1° × 1° JJA 2019 cloud detection frequency (CDF) maps for (a) ICESat-2 and (b) CALIOP, and (c) the difference between ICESat-2 and CALIOP (CDF of ICESat-2 − CDF of CALIOP).
Figure 2.
Zonal mean profiles of JJA 2019 cloud detection frequency for (a) ICESat-2 and (b) CALIOP, and (c) the difference between ICESat-2 and CALIOP (CDF of ICESat-2 − CDF of CALIOP).
Figure 2.
Zonal mean profiles of JJA 2019 cloud detection frequency for (a) ICESat-2 and (b) CALIOP, and (c) the difference between ICESat-2 and CALIOP (CDF of ICESat-2 − CDF of CALIOP).
Figure 3.
Frequency of positive ICESat-2 folding flag and CALIOP detection of high-altitude cloud or aerosol layers (>15 km) (JJA 2019).
Figure 3.
Frequency of positive ICESat-2 folding flag and CALIOP detection of high-altitude cloud or aerosol layers (>15 km) (JJA 2019).
Figure 4.
Global 1° × 1° JJA 2019 aerosol detection frequency (ADF) maps for (a) ICESat-2, (b) CALIOP, and (c) the difference between ICESat-2 and CALIOP (ADF of ICESat-2 − ADF of CALIOP).
Figure 4.
Global 1° × 1° JJA 2019 aerosol detection frequency (ADF) maps for (a) ICESat-2, (b) CALIOP, and (c) the difference between ICESat-2 and CALIOP (ADF of ICESat-2 − ADF of CALIOP).
Figure 5.
Zonal mean profiles of JJA 2019 aerosol detection frequency (ADF) for (a) ICESat-2 and (b) CALIOP, and (c) the difference between ICESat-2 and CALIOP ((a) and (b)) (ADF of ICESat-2 − ADF of CALIOP).
Figure 5.
Zonal mean profiles of JJA 2019 aerosol detection frequency (ADF) for (a) ICESat-2 and (b) CALIOP, and (c) the difference between ICESat-2 and CALIOP ((a) and (b)) (ADF of ICESat-2 − ADF of CALIOP).
Figure 6.
Global 1° ×1° average aerosol optical depth for JJA 2019 as observed by (a) ICESat-2 and (b) MODIS. Mean values between 50°N and S are highlighted in lower left box. Red boxes outline the region of higher aerosol optical depth analyzed in Section 3.3.
Figure 6.
Global 1° ×1° average aerosol optical depth for JJA 2019 as observed by (a) ICESat-2 and (b) MODIS. Mean values between 50°N and S are highlighted in lower left box. Red boxes outline the region of higher aerosol optical depth analyzed in Section 3.3.
Figure 7.
Coincident ICESat-2 (b) and CALIOP observations (d) (20 July 2019; 22:20 UTC). Ten second average profile of ICESat-2 and CALIOP is shown in (c) around the orbital coincident point (black line; 30.2°N 57.3°E). A map of the ICESat-2 and CALIOP orbits is shown in (a) with colored portion (ICESat-2 in red and CALIOP in blue) corresponding to the portion plotted in (b,d).
Figure 7.
Coincident ICESat-2 (b) and CALIOP observations (d) (20 July 2019; 22:20 UTC). Ten second average profile of ICESat-2 and CALIOP is shown in (c) around the orbital coincident point (black line; 30.2°N 57.3°E). A map of the ICESat-2 and CALIOP orbits is shown in (a) with colored portion (ICESat-2 in red and CALIOP in blue) corresponding to the portion plotted in (b,d).
Figure 8.
Coincident ICESat-2 and CALIOP observations (23 July 2019; 20:59 UTC). The black lines denote the orbital coincident point. Folding of high-altitude clouds is apparent in boxed region in the ICESat-2 profile.
Figure 8.
Coincident ICESat-2 and CALIOP observations (23 July 2019; 20:59 UTC). The black lines denote the orbital coincident point. Folding of high-altitude clouds is apparent in boxed region in the ICESat-2 profile.
Figure 9.
Comparison of ICESat-2 and CALIOP identified layer top (a) and layer base (b) for cloud or aerosol layers at orbital coincident points.
Figure 9.
Comparison of ICESat-2 and CALIOP identified layer top (a) and layer base (b) for cloud or aerosol layers at orbital coincident points.
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. |
© 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
MDPI and ACS Style
Christian, K.E.; Palm, S.P.; Yorks, J.E.; Nowottnick, E.P. Evaluation of ICESat-2 ATL09 Atmospheric Products Using CALIOP and MODIS Space-Based Observations. Remote Sens. 2025, 17, 482. https://doi.org/10.3390/rs17030482
AMA Style
Christian KE, Palm SP, Yorks JE, Nowottnick EP. Evaluation of ICESat-2 ATL09 Atmospheric Products Using CALIOP and MODIS Space-Based Observations. Remote Sensing. 2025; 17(3):482. https://doi.org/10.3390/rs17030482
Chicago/Turabian StyleChristian, Kenneth E., Stephen P. Palm, John E. Yorks, and Edward P. Nowottnick. 2025. "Evaluation of ICESat-2 ATL09 Atmospheric Products Using CALIOP and MODIS Space-Based Observations" Remote Sensing 17, no. 3: 482. https://doi.org/10.3390/rs17030482
APA StyleChristian, K. E., Palm, S. P., Yorks, J. E., & Nowottnick, E. P. (2025). Evaluation of ICESat-2 ATL09 Atmospheric Products Using CALIOP and MODIS Space-Based Observations. Remote Sensing, 17(3), 482. https://doi.org/10.3390/rs17030482
Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.
