Aerosol Robotic Network Observations Research Articles

Comparative analysis of CAMS aerosol optical depth data and AERONET observations in the Eastern Mediterranean over 19 years

Aerosol optical depth (AOD) is an essential metric for evaluating the atmospheric aerosol load and its impacts on climate, air quality, and public health. In this study, the AOD data from the Copernicus Atmosphere Monitoring Service (CAMS) were validated against ground-based measurements from the Aerosol Robotic Network (AERONET) throughout the Eastern Mediterranean, a region characterized by diverse aerosol types and sources. A comparative analysis was performed on 3-hourly CAMS AOD values at 550 nm against observations from 20 AERONET stations across Cyprus, Greece, Israel, Egypt, and Turkey from 2003 to 2021. The CAMS AOD data exhibited a good overall agreement with AERONET AOD data, demonstrated by a Pearson correlation coefficient of 0.77, a mean absolute error (MAE) of 0.08, and a root mean square error (RMSE) of 0.11. Nonetheless, spatial and temporal variations were observed in the CAMS AOD data performance, with site-specific correlation coefficients ranging from 0.57 to 0.85, the lowest correlations occurring in Egypt and the highest in Greece. An underestimation of CAMS AOD was noted at inland sites with high AOD levels, while a better agreement was observed at coastal sites with lower AOD levels. The diurnal variation analysis indicated improved CAMS reanalysis performance during the afternoon and evening hours. Seasonally, CAMS reanalysis showed better agreement with AERONET AODs in spring and autumn, with lower correlation coefficients noted in summer and winter. This study marks the first comprehensive validation of CAMS AOD performance in the Eastern Mediterranean, offering significant enhancements for regional air quality and climate modeling, and underscores the essential role of consistent validation in refining aerosol estimations within this complex and dynamic geographic setting.

Estimation of pan-European, daily total, fine-mode and coarse-mode Aerosol Optical Depth at 0.1° resolution to facilitate air quality assessments

Aerosol Optical Depth (AOD) data derived from satellites is crucial for estimating spatially-resolved PM concentrations, but existing AOD data over land remain affected by several limitations (e.g., data gaps, coarser resolution, higher uncertainty or lack of size fraction data), which weakens the AOD-PM relationship. We developed a 0.1° resolution daily AOD data set over Europe over the period 2003–2020, based on two-stage Quantile Machine Learning (QML) frameworks. Our approach first fills gaps in satellite AOD data and then constructs three components' models to obtain reliable full-coverage AOD along with Fine-mode AOD (fAOD) and Coarse-mode AOD (cAOD). These models are based on AERONET (AErosol RObotic NETwork) observations, Gap-filled satellite AOD, climate and atmospheric composition reanalyses. Our QML AOD products exhibit better quality with an out-of-sample R2 equal to 0.68 for AOD, 0.66 for fAOD and 0.65 for cAOD, which is 23–92 %, 11–13 % and 115–132 % higher than the corresponding satellite or reanalysis products, respectively. Over 91.6 %, 81.6 %, and 88.9 % of QML AOD, fAOD and cAOD predictions fall within ±20 % Expected Error (EE) envelopes, respectively. Previous studies reported that a weak satellite AOD-PM correlation across Europe (Pearson correlation coefficient (PCC) around 0.1). Our QML products exhibit higher correlations with ground-level PMs, particularly when broadly matched by size: AOD with PM10, fAOD with PM2.5, cAOD with PM coarse (R = 0.41, 0.45 and 0.26, respectively). Different AOD fractions more effectively distinct PM size fractions, than total AOD. Our QML aerosol dataset and models pioneer full-coverage, daily high-resolution monitoring of fine-mode and coarse-mode aerosols, effectively addressing existing AOD challenges for further PMs exposures' estimations. This dataset opens avenues for more in-depth exploration of the impacts of aerosols on human health, climate, visibility, and biogeochemical processes, offering valuable insights for air quality management and environmental health risk assessment.

The prototype NOAA Aerosol Reanalysis version 1.0: description of the modeling system and its evaluation

Abstract. In this study, we describe the first prototype version of global aerosol reanalysis at the National Oceanic and Atmospheric Administration (NOAA), the prototype NOAA Aerosol Reanalysis version 1.0 (pNARA v1.0) that was produced for the year 2016. In pNARA v1.0, the forecast model is an early version of the operational Global Ensemble Forecast System-Aerosols (GEFS-Aerosols) model. The three-dimensional ensemble-variational (3D-EnVar) data assimilation (DA) system configuration is built using elements of the Joint Effort for Data Assimilation Integration (JEDI) framework being developed at the Joint Center for Satellite Data Assimilation (JCSDA). The Neural Network Retrievals (NNR) of aerosol optical depth (AOD) at 550 nm from the Moderate Resolution Imaging Spectroradiometer (MODIS) instruments are assimilated to provide reanalysis of aerosol mass mixing ratios. We evaluate pNARA v1.0 against a wide variety of Aerosol Robotic Network (AERONET) observations, against the National Aeronautics and Space Administration's Modern-Era Retrospective Analysis for Research and Applications 2 (MERRA-2; Gelaro et al., 2017; Randles et al., 2017; Buchard et al., 2017) and the European Centre for Medium-Range Weather Forecasts' Copernicus Atmosphere Monitoring Service Reanalysis (CAMSRA; Inness et al., 2019), and against measurements of surface concentrations of particulate matter 2.5 (PM2.5) and aerosol species. Overall, the 3D-EnVar DA system significantly improves AOD simulations compared with observations, but the assimilation has limited impact on chemical composition and size distributions of aerosols. We also identify deficiencies in the model's representations of aerosol chemistry and their optical properties elucidated from evaluation of pNARA v1.0 against AERONET observations. A comparison of seasonal profiles of aerosol species from pNARA v1.0 with the other two reanalyses exposes significant differences among datasets. These differences reflect uncertainties in simulating aerosols in general.

Estimation of high-resolution aerosol optical depth (AOD) from Landsat and Sentinel images using SEMARA model over selected locations in South Asia

The study investigates a high-resolution retrieval of AOD over complex surfaces and various aerosol regimes of South Asian countries such as China, Bangladesh and Nepal from Landsat 7/8/9 and Sentinel 2 images. To estimate AOD from these satellite images, unlike the conventional radiative transfer-based look-up-table (LUT) approach, this study uses SEMARA coupled model which combines a simplified and robust surface reflection estimation (SREM) model and a simplified aerosol retrieval algorithm (SARA). The SEMARA model uses top-of-atmospheric reflectance (TOA) and land surface reflectance (LSR), solar and sensors viewing geometries, including the in-situ AOD data from the Aerosol Robotic Network (AERONET) station to simulate AOD at 30 m and 10 m spatial resolutions. The results are compared with AERONET and MODIS-based AOD for the validation sites in the South Asian countries. A high degree of agreement with AERONET AOD is obtained over Beijing in China (R2 > 0.9) for all sensors, while it varies for the validation AERONET stations over Nepal and Bangladesh (R2 = 0.40–0.86). The accuracy metrices suggest a low uncertainty over China Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) for Beijing: 0.07 and 0.13), while it remains higher for stations over Nepal and Bangladesh (MAE and RMSE: 0.33, and 0.41; 0.15–0.27, respectively). >85% of the modelled observations fall within expected error (EE) for China, while 20–75% of observations fall within EE for Nepal and Bangladesh. We have also obtained a proximity effect for the accuracy of the estimated AOD from the SEMARA model. The findings, in general, suggest that the complexity in surface cover and aerosol type produce a higher offset with respect to AERONET observations over Bangladesh and Nepal, and therefore further validation studies are needed for other parts of the globe to measure the consistency of the model.

An AERONET-based methodology to retrieve black carbon light absorption and comparison with MERRA-2 data

The contribution of black carbon in absorption aerosol optical depth (BC AAOD) can be retrieved using depolarization ratio (DPR, δp), single scattering albedo (SSA, ω), and Ångstrom exponent (AE, å). AErosol RObotic NETwork (AERONET) inversion products provide ground-based long-term observations of aerosol optical properties based on the sun photometer's observations. In this study, the AERONET version 3 level 2.0 products are utilized to approximate BC AAOD on a global scale during the recent two decades (2000–2018). We compared our AERONET-retrieved BC AAODs with the ones provided by the Modern-Era Retrospective analysis for Research and Applications, version 2.0 (MERRA-2) data, at the selected AERONET site's location and observation time. We found the highest MERRA-2 and AERONET-derived BC AAOD over densely populated countries such as China and India. Further, Nigeria, one of the largest sources of biomass burning and a leading gas-flaring country, showed increased values of BC AAOD. Both datasets confirmed a drop in BC AAOD over Sao-Paulo, Beijing, and Seoul from 2000 to 2018. AERONET and MERRA-2 data indicated higher BC variations during cold months in Beijing, Kanpur, Seoul, Ilorin, and Sao-Paulo. We concluded that the high BC values over the populous countries mainly owe to the considerable BC emission from transportation, industrial, and residential sectors in response to the high population. A comparison of MERRA-2 and our AERONET-derived retrieval methodology suggested lower values of MERRA-2 BC AAOD compared to AERONET-derived variations in Sao-Paulo.

An Aerosol Optical Module With Observation‐Constrained Black Carbon Properties for Global Climate Models

AbstractAtmospheric black carbon (BC) aerosols have been long‐lasting uncertain components in environmental and climate studies. Global climate models (GCMs) potentially overestimate BC absorption efficiency due to a lack of consideration of complex BC microphysical and mixing properties. We extract multiple BC properties from observations and develop an aerosol optical module known as Advanced Black Carbon (ABC) in the framework of the Modal Aerosol Model version 4 (MAM4). The ABC module is implemented in the Community Atmosphere Model version 6 (CAM6) and evaluated by in situ and remote sensing observations. CAM6‐ABC addresses the shortcomings of CAM6‐MAM4 in terms of BC microphysical and mixing properties, particularly their size, mixing state and optical simulations. Sensitivity simulations show that the global BC absorption aerosol optical depth at 550 nm simulated by CAM6‐ABC is reduced by ∼29% compared with that in CAM6‐MAM4. The BC absorption enhancement simulated by CAM6‐ABC is reduced from ∼2.6 of the default MAM4 to ∼1.4, which is closer to the observed values (mostly less than 1.5). With improved BC absorption estimation, the biases of aerosol single‐scattering coalbedo simulations are reduced by 18%–69% compared with global Aerosol Robotic Network observations. Moreover, the globally averaged BC direct radiative effect is reduced from 0.37 to 0.28 W/m2 at the top of the atmosphere. Our new scheme alleviates the overestimation of BC absorption in GCMs by constraining BC microphysical and mixing properties when assessing aerosol radiative and climate effects, and it can be easily implemented in most modal‐based aerosol modules of climate models.

An effective atmospheric correction method for the wide swath of Chinese GF-1 and GF-6 WFV images on lands

Accurate land surface reflectance plays an important role in the accurate inversion of surface parameters, and atmospheric correction plays a decisive role in obtaining accurate reflectance. For GF-1 WFV and GF-6 WFV images, there are two major issues to be addressed, including the spectral differences between nadir with far off-nadir pixels and the spatial variability of atmospheric components for wide imaging. Therefore, this study focuses on these two issues using the Sensor Invariant Atmospheric Correction (SIAC) method. Our results reveal that the SIAC approach improves the correlation accuracy from 0.8868 to 0.9173 for GF-1 WFV image compared with Sentinel-2 reflectance, from 0.9530 to 0.9620 for GF-6 WFV image compared with the results using FLAASH model. For alleviating wide-swathed anisotropy, the directional imaging angle is calculated with the result ranging from 5.6450° to 33.7497°. Furthermore, the atmospheric components have been inversed pixel by pixel with obvious spatial variation. And the correlation of inversed aerosol optical thickness (AOT) and total column water vapor (TCWV) with a spatial resolution of 500 m TCWV with measured results of AERONET (AErosol RObotic NETwork) observation stations are 0.9175 and 0.4442, respectively. These results reveal that the atmospheric correction method works well, which is effective for the wide swath of Chinese GF-1 WFV and GF-6 WFV images on land.

A global evaluation of daily to seasonal aerosol and water vapor relationships using a combination of AERONET and NAAPS reanalysis data

Abstract. The co-transport of aerosol particles and water vapor has long been noted in the literature, with a myriad of implications such as air mass characterization, radiative transfer, and data assimilation. Here, the relationship between aerosol optical depth (AOD) and precipitable water vapor (PW) is evaluated to our knowledge for the first time globally, at daily to seasonal levels using approximately 20 years of NASA Aerosol Robotic Network (AERONET) observational data and the 16-year Navy Aerosol Analysis Prediction System (NAAPS) reanalysis v1.0 (NAAPS-RA) model fields. The combination of AERONET observations with small uncertainties and the reanalysis fields with global coverage is used to provide a best estimate of the seasonal AOD and PW relationships, including an evaluation of correlations, slope, and PW probability distributions for identification of statistically significant differences in PW for high-AOD events. The relationships produced from the AERONET and NAAPS-RA datasets were compared against each other and showed consistency, indicating that the NAAPS-RA provides a realistic representation of the AOD and PW relationship. The analysis includes layer AOD and PW relationships for proxies of the planetary boundary layer and the lower, middle, and upper free troposphere. The dominant AOD and PW relationship is positive, supported by both AERONET and model evaluation, which varies in strength by season and location. These relationships were found to be statistically significant and present across the globe, observed on an event-by-event level. Evaluations at individual AERONET sites implicate synoptic-scale transport as a contributing factor in these relationships at daily levels. Negative AOD and PW relationships were identified and predominantly associated with regional dry-season timescales in which biomass burning is the predominant aerosol type. This is not an indication of dry-air association with smoke for an individual event but is a reflection of the overall dry conditions leading to more biomass burning and higher associated AOD values. Stronger correlations between AOD and PW are found when evaluating the data by vertical layers, including the boundary layer and the lower, middle, and upper free troposphere (corresponding to typical water vapor channels), with the largest correlations observed in the free troposphere – indicative of aerosol and water vapor transport events. By evaluating the variability between PW and relative humidity in the NAAPS-RA, hygroscopic growth was found to be a dominant term to (1) amplify positive AOD–PW relationships, particularly in the midlatitudes; (2) diminish negative relationships in dominant biomass burning regions; and (3) lead to statistically insignificant changes in PW for high-AOD events for maritime regions. The importance of hygroscopic growth in these relationships indicates that PW is a useful tracer for AOD or light extinction but not necessarily as strongly for aerosol mass. Synoptic-scale African dust events are an exception where PW is a strong tracer for aerosol transport shown by strong relationships even with hygroscopic effects. Given these results, PW can be exploited in coupled aerosol and meteorology data assimilation for AOD, and the collocation of aerosol and water vapor should be carefully taken into account when conducting particulate matter (PM) retrievals from space and in evaluating radiative impacts of aerosol, with the season and location in mind.

TROPOMI/S5P Total Column Water Vapor validation against AERONET ground-based measurements

Abstract. Water vapor plays an important role in the greenhouse effect, rendering it an atmospheric constituent that requires continuous and global monitoring by different types of remote sensing instruments. The TROPOspheric Monitoring Instrument Sentinel-5 Precursor (TROPOMI/S5P) Total Column Water Vapor (TCWV) is a new product retrieved from the visible blue spectral range (435–455 nm), using an algorithm that was originally developed for the GOME-2/MetOp sensors. For the purposes of this work, 2.5 years of continuous satellite observations at high spatial resolution are validated against co-located (in space and in time) precipitable water Level 2.0 (quality-assured) ground-based measurements from the NASA AERONET (AErosol RObotic NETwork). The network uses Cimel Sun photometers located at approximately 1300 stations globally to monitor precipitable water among other products. Based on data availability, 369 of the stations were used in this study. The two datasets, satellite- and ground-based, were co-located, and the relative differences of the comparisons were calculated and statistically analyzed. The Pearson correlation coefficient of the two products is found to be 0.91, and the mean bias of the overall relative percentage differences is of the order of −2.7 %. For the Northern Hemisphere midlatitudes (30–60∘ N), where the density of the ground-based stations is high, the mean relative bias was found to be −1.8 %, while in the tropics (±15∘) the TROPOMI TCWV product has a relative dry bias of up to −10 %. The effect of various algorithm and geophysical parameters, such as air mass factor, solar zenith angle, clouds and albedo, is also presented and discussed. It was found that the cloud properties affect the validation results, leading the TCWV to a dry bias of −20 % for low cloud heights (cloud top pressure (CTP) >800 hPa). Moreover, cloud albedo introduces a wet bias of 15 % when it is below 0.3 and a dry bias up to −25 % when the clouds are more reflective. Overall, the TROPOMI/S5P TCWV product, on a global scale and for moderate albedo and cloudiness, agrees well at -2.7±4.9 % with the AERONET observations but probably within about −8 % to −13 % with respect to the “truth”.

Application of Optimal Interpolation to Spatially and Temporally Sparse Observations of Aerosol Optical Depth

Aerosol optical depth (AOD) is one of the basic characteristics of atmospheric aerosol. A global ground-based network of sun and sky photometers, the Aerosol Robotic Network (AERONET) provides AOD data with low uncertainty. However, AERONET observations are sparse in space and time. To improve data density, we merged AERONET observations with a GEOS-Chem chemical transport model prediction using an optimal interpolation (OI) method. According to OI, we estimated AOD as a linear combination of observational data and a model forecast, with weighting coefficients chosen to minimize a mean-square error in the calculation, assuming a negligible error of AERONET AOD observations. To obtain weight coefficients, we used correlations between model errors in different grid points. In contrast with classical OI, where only spatial correlations are considered, we developed the spatial-temporal optimal interpolation (STOI) technique for atmospheric applications with the use of spatial and temporal correlation functions. Using STOI, we obtained estimates of the daily mean AOD distribution over Europe. To validate the results, we compared daily mean AOD estimated by STOI with independent AERONET observations for two months and three sites. Compared with the GEOS-Chem model results, the averaged reduction of the root-mean-square error of the AOD estimate based on the STOI method is about 25%. The study shows that STOI provides a significant improvement in AOD estimates.

Atmospheric Correction Inter-comparison eXercise, ACIX-II Land: An assessment of atmospheric correction processors for Landsat 8 and Sentinel-2 over land

The correction of the atmospheric effects on optical satellite images is essential for quantitative and multi-temporal remote sensing applications. In order to study the performance of the state-of-the-art methods in an integrated way, a voluntary and open-access benchmark Atmospheric Correction Inter-comparison eXercise (ACIX) was initiated in 2016 in the frame of Committee on Earth Observation Satellites (CEOS) Working Group on Calibration & Validation (WGCV). The first exercise was extended in a second edition wherein twelve atmospheric correction (AC) processors, a substantially larger testing dataset and additional validation metrics were involved. The sites for the inter-comparison analysis were defined by investigating the full catalogue of the Aerosol Robotic Network (AERONET) sites for coincident measurements with satellites' overpass. Although there were more than one hundred sites for Copernicus Sentinel-2 and Landsat 8 acquisitions, the analysis presented in this paper concerns only the common matchups amongst all processors, reducing the number to 79 and 62 sites respectively. Aerosol Optical Depth (AOD) and Water Vapour (WV) retrievals were consequently validated based on the available AERONET observations. The processors mostly succeeded in retrieving AOD for relatively light to medium aerosol loading (AOD < 0.2) with uncertainties <0.08, while the overall uncertainty values were typically 0.23 ± 0.15. Better performances were observed for WV retrievals with >90% of the results falling within the suggested empirical specifications and with the Root Mean Square Error (RMSE) being mostly <0.25 g/cm2. Regarding Surface Reflectance (SR) validation two main approaches were followed. For the first one, a simulated SR reference dataset was computed over all of the test sites by using the 6SV (Second Simulation of the Satellite Signal in the Solar Spectrum vector code) full radiative transfer modelling (RTM) and AERONET measurements for the required aerosol variables and water vapour content. The performance assessment demonstrated that the retrievals were not biased for most of the bands. The uncertainties ranged from approximately 0.003 to 0.01 (excluding B01) for the best performing processors in both sensors' analyses. For the second one, measurements from the radiometric calibration network RadCalNet over La Crau (France) and Gobabeb (Namibia) were involved in the validation. The performance of the processors was in general consistent across all bands for both sensors and with low standard deviations (<0.04) between on-site and estimated surface reflectance. Overall, our study provides a good insight of AC algorithms' performance to developers and users, pointing out similarities and differences for AOD, WV and SR retrievals. Such validation though still lacks of ground-based measurements of known uncertainty to better assess and characterize the uncertainties in SR retrievals.

Aerosol characteristics over Indo-Gangetic Plain from ground-based AERONET and MERRA-2/CAMS model simulations

A comprehensive investigation of physical, optical, and chemical characteristics of columnar aerosols over two locations with distinct environmental settings in the Indo-Gangetic Plain (IGP) region, namely, Kanpur (urban and industrial area) and Gandhi College (rural area), is conducted using high-quality aerosol datasets obtained from ground-based Aerosol Robotic Network (AERONET) observations during the recent five year period (2015–2019). This study utilizes all the crucial columnar aerosol parameters necessary for accurately estimating aerosol radiative forcing. Quantification of contribution by different aerosol species originating from natural and anthropogenic sources to the total aerosol optical depth (AOD) and single scattering albedo (SSA) is important to understand the specific mechanisms that influence the aerosol composition, thereby reducing the uncertainty in aerosol radiative forcing. For the first time, two highly spatially resolved models' (Modern-Era Retrospective Analysis for Research and Applications-2 (MERRA-2) and Copernicus Atmosphere Monitoring Service (CAMS)) simulated absorbing species-wise (black carbon (BC), dust, and brown carbon (BrC)) AOD, and absorption AOD (AAOD) are compared and contrasted against the AERONET observations over the IGP region in a systematic manner. MERRA-2 AODs are mostly lower, whereas CAMS AODs are consistently higher than the AERONET AODs. A comparison of collocated time and space observations with models clearly suggests that improvements in emission inventories on a seasonal scale are essential. MERRA-2 SSA is noted lower than the AERONET SSA during the winter season due to overestimation in BC AOD. During winter in >70% of MERRA-2 simulated SSA the difference is higher than ±0.03 (the uncertainty range of AERONET SSA) whereas during pre-monsoon and monsoon seasons >60% of MERRA-2 SSA lies within the uncertainty range of AERONET SSA. Both models show a gradient in AODDust decreasing from west to east in the IGP. However, observations do not often exhibit the gradient in dust, which is validated by air mass back trajectory analyses as air masses travel through different pathways to IGP and reverse the west to east gradient in AODDust. This quantitative and comparative collocated analysis of observed aerosol characteristics with models on a seasonal scale will enable a better estimation of aerosol radiative forcing, and can help improve aerosol processes and parameterizations in models.

Validation, analysis, and comparison of MISR V23 aerosol optical depth products with MODIS and AERONET observations

The latest Multi-angle Imaging Spectro Radiometer (MISR) Version (V) 23 aerosol optical depth (AOD) products were released, with an improved spatial resolution of 4.4 km, providing an unprecedented opportunity for the refined regional application. To ensure the reliability of their applications and build a scientific reference for the further optimization, it is imperative to conduct a comprehensive evaluation, especially for the unique size-resolved AOD products: small-size AOD (AODS, representing the contribution of fine-mode aerosols), medium-size AOD (AODM), and large-size AOD (AODL), and AODM+L represents the AOD part of coarse-mode aerosols. AErosol RObotic NETwork (AERONET) and MODerate-resolution Imaging Spectroradiometer (MODIS) Collection (C) 6.1 aerosol products from 2001 to 2020 are utilized for the validation, analysis, and inter-comparison, considering three spatial scales and four key factors. In general, MISR V23 aerosol products show a good accuracy compared with AERONET. The best performance for all AOD products appears in forest units (the highest R ~ 0.93, data percentage within Expected Error bounds, %EE > 93), related to the inactive human activity and dark underlying surface. Dependences of retrieval deviations illustrate that the performance of MISR AOD deteriorates as aerosol loading increases. Namely, with the increase of aerosols, total AOD (AODT) and AODS show increasing negative deviations, while increasing positive deviations are observed for AODM+L. This suggests that the Empirical Orthogonal Functions do not perform well in this situation, since numerous aerosol particles can obstruct the underlying reflection and reduce the surface spectral contrast. In addition, AODT and AODS often exhibit anomalous positive deviations in areas with low vegetation cover, such as deserts, revealing that MISR will overestimate aerosol content over bright surfaces and in environments dominated by coarse-mode particles. The above findings not only deepen the understanding of MISR aerosols products from multiple perspectives, but also provide useful information for the product improvement.

Simultaneous Characterization of Wildfire Smoke and Surface Properties With Imaging Spectroscopy During the FIREX-AQ Field Campaign.

We introduce and evaluate an approach for the simultaneous retrieval of aerosol and surface properties from Airborne Visible/Infrared Imaging Spectrometer Classic (AVIRIS‐C) data collected during wildfires. The joint National Aeronautics and Space Administration (NASA) National Oceanic and Atmospheric Administration Fire Influence on Regional to Global Environments and Air Quality field campaign took place in August 2019, and involved two aircraft and coordinated ground‐based observations. The AVIRIS‐C instrument acquired data from onboard NASA's high altitude ER‐2 research aircraft, coincident in space and time with aerosol observations obtained from the Aerosol Robotic Network (AERONET) DRAGON mobile platform in the smoke plume downwind of the Williams Flats Fire in northern Washington in August 2019. Observations in this smoke plume were used to assess the capacity of optimal‐estimation based retrievals to simultaneously estimate aerosol optical depth (AOD) and surface reflectance from Visible Shortwave Infrared (VSWIR) imaging spectroscopy. Radiative transfer modeling of the sensitivities in spectral information collected over smoke reveal the potential capacity of high spectral resolution retrievals to distinguish between sulfate and smoke aerosol models, as well as sensitivity to the aerosol size distribution. Comparison with ground‐based AERONET observations demonstrates that AVIRIS‐C retrievals of AOD compare favorably with direct sun AOD measurements. Our analyses suggest that spectral information collected from the full VSWIR spectral interval, not just the shortest wavelengths, enables accurate retrievals. We use this approach to continuously map both aerosols and surface reflectance at high spatial resolution across heterogeneous terrain, even under relatively high AOD conditions associated with wildfire smoke.

Retrieval of Black Carbon Absorption Aerosol Optical Depth from AERONET Observations over the World during 2000–2018

Black carbon (BC) absorption aerosol optical depth (AAODBC) defines the contribution of BC in light absorption and is retrievable using sun/sky radiometer measurements provided by Aerosol Robotic Network (AERONET) inversion products. In this study, we utilized AERONET-retrieved depolarization ratio (DPR, δp), single scattering albedo (SSA, ω), and Ångström Exponent (AE, å) of version 3 level 2.0 products as indicators to estimate the contribution of BC to the absorbing fractions of AOD. We applied our methodology to the AERONET sites, including North and South America, Europe, East Asia, Africa, India, and the Middle East, during 2000–2018. The long-term AAODBC showed a downward tendency over Sao Paulo (−0.001 year−1), Thessaloniki (−0.0004 year−1), Beijing (−0.001 year−1), Seoul (−0.0015 year−1), and Cape Verde (−0.0009 year−1) with the highest values over the populous sites. This declining tendency in AAODBC can be attributable to the successful emission control policies over these sites, particularly in Europe, America, and China. The AAODBC at the Beijing, Sao Paulo, Mexico City, and the Indian sites showed a clear seasonality indicating the notable role of residential heating in BC emissions over these sites during winter. We found a higher correlation between AAODBC and fine mode AOD at 440 nm at all sites except for Beijing. High pollution episodes, BC emission from different sources, and aggregation properties seem to be the main drivers of higher AAODBC correlation with coarse particles over Beijing.

Estimation of Aerosol Optical Depth at 30 m Resolution Using Landsat Imagery and Machine Learning

Current remote sensing-based aerosol optical depth (AOD) products have coarse spatial resolutions, which are useful for studies at continental and global scales, but unsatisfactory for local scale applications, such as urban air pollution monitoring. In this study, we investigated the possibility of using Landsat imagery to develop high-resolution AOD estimations at 30 m based on machine learning algorithms. We assessed the performance of six machine learning algorithms, including Extreme Gradient Boosting, Random Forest, Cascade Random Forest, Gradient Boosted Decision Trees, Extremely Randomized Trees, and Multiple Linear Regression. To obtain accurate AOD estimations, we used prior knowledge from multiple sources as inputs to the machine learning models, including the Global Land Surface Satellite (GLASS) albedo, the 1-km AOD product from MODIS data using the Multi-Angle Implementation of Atmospheric Correction (MAIAC) algorithm, and meteorological and surface elevation data. A total of 13,624 AOD measurements from Aerosol Robotic Network (AERONET) sites were used for model training and validation. We found that all six algorithms exhibited good performance, with R2 values ranging from 0.73 to 0.78 and AOD root-mean-square errors (RMSE) ranging from 0.089 to 0.098. The extremely randomized trees algorithm, however, demonstrated marginally superior performance as compared to the other algorithms; hence, it was used to produce AOD estimates at a 30 m resolution for one Landsat scene coving Beijing in 2013–2019. Through a comparison with overlapping AERONET observations, a high level of accuracy was achieved, with an R2 = 0.889 and an RMSE = 0.156. Our method can be potentially used to generate a global high-resolution AOD dataset based on Landsat imagery.

Global estimation of clear-sky shortwave aerosol direct radiative effects based on CALIPSO observations

ABSTRACT The total and individual aerosol direct radiative effects (ADREs) were estimated for clear-sky conditions using Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations (CALIPSO) observations and radiative transfer model. In this study, a parametric sensitivity analysis was performed for the Santa Barbara DISORT Atmospheric Radiative Transfer model using a global sensitivity analysis method. Uncertainties in the ADREs due to aerosol optical properties and surface albedo errors were evaluated. The single-scattering albedo and asymmetry factor for marine, dust, pollution, and smoke aerosols required for the model calculations were obtained from the classified AErosol RObotic NETwork observations. The estimated average global ADREs and errors at the top-of-atmosphere (TOA) and the surface were −2.36 ± 0.54 and −4.78 ± 2.2 Wm−2, respectively. In regions with higher dust, pollution, and smoke aerosol loading, the ADREs exhibited significant seasonal variability. In the Sahara and Arabian deserts, during the June-July-August season with higher dust aerosol loading, the seasonal average ADREs in the region were −3.23 and −21.37 Wm−2 at the TOA and surface, respectively. In the Indian region, during the March–April–May season with higher pollution aerosol loading, the ADREs were −10.33 and −28.04 Wm−2 at the TOA and surface, respectively. In Southern Africa, the smoke aerosol with a single-scattering albedo of 0.87 caused negative radiative effects at the TOA, and during the September-October-November season with higher smoke aerosol loading, the seasonal average ADREs were −2.34 and −6.36 Wm−2 at the TOA and surface, respectively.

Satellite-Based Aerosol Classification for Capital Cities in Asia Using a Random Forest Model

Aerosol types in Asian capital cities were classified using a random forest (RF) satellite-based aerosol classification model during 2018–2020 in an investigation of the contributions of aerosol types, with or without Aerosol Robotic Network (AERONET) observations. In this study, we used the recently developed RF aerosol classification model to detect and classify aerosols into four types: pure dust, dust-dominated aerosols, strongly absorbing aerosols, and non-absorbing aerosols. Aerosol optical and microphysical properties for each aerosol type detected by the RF model were found to be reasonably consistent with those for typical aerosol types. In Asian capital cities, pollution-sourced aerosols, especially non-absorbing aerosols, were found to predominate, although Asian cities also tend to be seasonally affected by natural dust aerosols, particularly in East Asia (March–May) and South Asia (March–August). No specific seasonal effects on aerosol type were detected in Southeast Asia, where there was a predominance of non-absorbing aerosols. The aerosol types detected by the RF model were compared with those identified by other aerosol classification models. This study indicates that the satellite-based RF model may be used as an alternative in the absence of AERONET sites or observations.

AEROCOM and AEROSAT AAOD and SSA study – Part 1: Evaluation and intercomparison of satellite measurements

Abstract. Global measurements of absorbing aerosol optical depth (AAOD) are scarce and mostly provided by the ground network AERONET (AErosol RObotic NETwork). In recent years, several satellite products of AAOD have been developed. This study's primary aim is to establish the usefulness of these datasets for AEROCOM (Aerosol Comparisons between Observations and Models) model evaluation with a focus on the years 2006, 2008 and 2010. The satellite products are super-observations consisting of 1∘×1∘×30 min aggregated retrievals. This study consists of two papers, the current one that deals with the assessment of satellite observations and a second paper (Schutgens et al., 2021) that deals with the evaluation of models using those satellite data. In particular, the current paper details an evaluation with AERONET observations from the sparse AERONET network as well as a global intercomparison of satellite datasets, with a focus on how minimum AOD (aerosol optical depth) thresholds and temporal averaging may improve agreement between satellite observations. All satellite datasets are shown to have reasonable skill for AAOD (three out of four datasets show correlations with AERONET in excess of 0.6) but less skill for SSA (single-scattering albedo; only one out of four datasets shows correlations with AERONET in excess of 0.6). In comparison, satellite AOD shows correlations from 0.72 to 0.88 against the same AERONET dataset. However, we show that performance vs. AERONET and inter-satellite agreements for SSA improve significantly at higher AOD. Temporal averaging also improves agreements between satellite datasets. Nevertheless multi-annual averages still show systematic differences, even at high AOD. In particular, we show that two POLDER (Polarization and Directionality of the Earth's Reflectances) products appear to have a systematic SSA difference over land of ∼0.04, independent of AOD. Identifying the cause of this bias offers the possibility of substantially improving current datasets. We also provide evidence that suggests that evaluation with AERONET observations leads to an underestimate of true biases in satellite SSA. In the second part of this study we show that, notwithstanding these biases in satellite AAOD and SSA, the datasets allow meaningful evaluation of AEROCOM models.

Simulating and Evaluating Global Aerosol Distributions With the Online Aerosol‐Coupled CAS‐FGOALS Model

AbstractWe implement an existing aerosol module named Spectral Radiation Transport Model for Aerosol Species (SPRINTARS) in the Chinese Academy of Sciences Flexible Global Ocean–Atmosphere–Land System (CAS‐FGOALS) model and simulate the global aerosol properties over 2002–2014. The simulated surface mass concentrations of individual aerosols generally reproduce the observed ones. The simulated spatial‐temporal distributions of the aerosol optical depths (AODs) are evaluated with multisource satellite retrievals, and the simulated AOD, Ångström Exponent (AE), and single scattering albedo (SSA) are further evaluated with the ground‐based Aerosol Robotic Network (AERONET) measurements. The spatial distribution of the modeled AOD is found to be generally comparable to the satellite retrievals. The interannual and seasonal variations of the modeled AOD over various aerosol regimes are also overall consistent with the AERONET observations. With respect to the AE, our model can reproduce the interannual variations fairly well over the dust dominant regions, and generally capture the seasonal variations over the industrial domain regions. Over the industrial domain regions, however, the systematic underestimation of the simulated AE is found, which is in part due to the biases of wind fields and overestimation of the relative humidity (RH) by the host model. With respect to the SSA, our model can also generally reproduce the observed seasonal variations, while the modeled values are typically lower than the observations, especially over North Africa.