Improving Spatial Coverage of Satellite Aerosol Classification Using a Random Forest Model

The spatial coverage of satellite aerosol classification was improved using a random forest (RF) model trained with observational data including target (aerosol type) and input (satellite measurement) variables. The AErosol RObotic NETwork (AERONET) aerosol-type dataset was used for the target variables. Satellite input variables with many missing data or low mean-decrease accuracy were excluded from the final input variable set, and good performance in aerosol-type classification was achieved. The performance of the RF-based model was evaluated on the basis of the wavelength dependence of single-scattering albedo (SSA) and fine-mode-fraction values from AERONET. Typical SSA wavelength dependence for individual aerosol types was consistent with that obtained for aerosol types by the RF-based model. The spatial coverage of the RF-based model was also compared with that of previously developed models in a global-scale case study. The study demonstrates that the RF-based model allows satellite aerosol classification with improved spatial coverage, with a performance similar to that of previously developed models.

Download Full-text

A First Approach to Aerosol Classification Using Space-Borne Measurement Data: Machine Learning-Based Algorithm and Evaluation

Remote Sensing ◽

10.3390/rs13040609 ◽

2021 ◽

Vol 13 (4) ◽

pp. 609

Author(s):

Wonei Choi ◽

Hanlim Lee ◽

Jeonghyeon Park

Keyword(s):

Machine Learning ◽

Measurement Data ◽

Wavelength Dependence ◽

Single Scattering ◽

New Method ◽

Spherical Particles ◽

Training Dataset ◽

Aerosol Classification ◽

Input Variables ◽

Aerosol Types

A new method was developed for classifying aerosol types involving a machine-learning approach to the use of satellite data. An Aerosol Robotic NETwork (AERONET)-based aerosol-type dataset was used as a target variable in a random forest (RF) model. The contributions of satellite input variables to the RF-based model were quantified to determine an optimal set of input variables. The new method, based on inputs of satellite variables, allows the classification of seven aerosol types: pure dust, dust-dominant mixed, pollution-dominant mixed aerosols, and pollution aerosols (strongly, moderately, weakly, and non-absorbing). The performance of the model was statistically evaluated using AERONET data excluded from the model training dataset. Model accuracy for classifying the seven aerosol types was 59%, improving to 72% for four types (pure dust, dust-dominant mixed, strongly absorbing, and non-absorbing). The performance of the model was evaluated against an earlier aerosol classification method based on the wavelength dependence of single-scattering albedo (SSA) and fine-mode-fraction values from AERONET. Typical wavelength dependences of SSA for individual aerosol types are consistent with those obtained for aerosol types by the new method. This study demonstrates that an RF-based model is capable of satellite aerosol classification with sensitivity to the contribution of non-spherical particles.

Download Full-text

A Short Note on the Potential of Utilization of Spectral AERONET-Derived Depolarization Ratios for Aerosol Classification

Atmosphere ◽

10.3390/atmos10030143 ◽

2019 ◽

Vol 10 (3) ◽

pp. 143 ◽

Cited By ~ 1

Author(s):

Il-Sung Zo ◽

Sung-Kyun Shin

Keyword(s):

Short Note ◽

Single Scattering ◽

Observation Data ◽

Spectral Variation ◽

Aerosol Classification ◽

Fine Mode ◽

Remote Sensing Techniques ◽

Aerosol Types ◽

Aerosol Type ◽

Depolarization Ratios

We herein present the spectral linear particle depolarization ratios (δp) from an Aerosol Robotics NETwork (AERONET) sun/sky radiometer with respect to the aerosol type. AERONET observation sites, which are representative of each aerosol type, were selected for our study. The observation data were filtered using the Ångström exponent (Å), fine-mode fraction (FMF) and single scattering albedo (ω) to ensure that the obtained values of δp were representative of each aerosol condition. We report the spectral δp values provided in the recently released AERONET version 3 inversion product for observation of the following aerosol types: dust, polluted dust, smoke, non-absorbing, moderately-absorbing and high-absorbing pollution. The AERONET-derived δp values were generally within the range of the δp values measured from lidar observations for each aerosol type. In addition, it was found that the spectral variation of δp differed according to the aerosol type. From the obtained results, we concluded that our findings provide potential insight into the identification and classification of aerosol types using remote sensing techniques.

Download Full-text

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

Remote Sensing ◽

10.3390/rs13132464 ◽

2021 ◽

Vol 13 (13) ◽

pp. 2464

Author(s):

Wonei Choi ◽

Hyeongwoo Kang ◽

Dongho Shin ◽

Hanlim Lee

Keyword(s):

Random Forest ◽

Classification Model ◽

Seasonal Effects ◽

Classification Models ◽

Capital Cities ◽

Microphysical Properties ◽

Aerosol Classification ◽

Absorbing Aerosols ◽

Aerosol Types ◽

Aerosol Type

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.

Download Full-text

Aerosol classification from airborne HSRL and comparisons with the CALIPSO vertical feature mask

Atmospheric Measurement Techniques Discussions ◽

10.5194/amtd-6-1815-2013 ◽

2013 ◽

Vol 6 (1) ◽

pp. 1815-1858 ◽

Cited By ~ 10

Author(s):

S. P. Burton ◽

R. A. Ferrare ◽

M. A. Vaughan ◽

A. H. Omar ◽

R. R. Rogers ◽

...

Keyword(s):

High Spectral Resolution ◽

Dependent Observations ◽

Aerosol Retrieval ◽

Direct Measurements ◽

Aerosol Classification ◽

Lidar Ratio ◽

Langley Research Center ◽

High Spectral Resolution Lidar ◽

Aerosol Types ◽

Aerosol Type

Abstract. Aerosol classification products from the NASA Langley Research Center (LaRC) airborne High Spectral Resolution Lidar (HSRL-1) on the NASA B200 aircraft are compared with coincident V3.01 aerosol classification products from the CALIOP instrument on the CALIPSO satellite. For CALIOP, aerosol classification is a key input to the aerosol retrieval, and must be inferred using aerosol loading-dependent observations and location information. In contrast, HSRL-1 makes direct measurements of aerosol intensive properties, including the lidar ratio, that provide information on aerosol type. In this study, comparisons are made for 109 underflights of the CALIOP orbit track. We find that 62% of the CALIOP marine layers and 54% of the polluted continental layers agree with HSRL-1 classification results. In addition, 80% of the CALIOP desert dust layers are classified as either dust or dusty mix by HSRL-1. However, agreement is less for CALIOP smoke (13%) and polluted dust (35%) layers. Specific case studies are examined, giving insight into the performance of the CALIOP aerosol type algorithm. In particular, we find that the CALIOP polluted dust type is overused due to an attenuation-related depolarization bias. Furthermore, the polluted dust type frequently includes mixtures of dust plus marine aerosol. Finally, we find that CALIOP's identification of internal boundaries between different aerosol types in contact with each other frequently do not reflect the actual transitions between aerosol types accurately. Based on these findings, we give recommendations which may help to improve the CALIOP aerosol type algorithms.

Download Full-text

Aerosol classification using airborne High Spectral Resolution Lidar measurements – methodology and examples

Atmospheric Measurement Techniques ◽

10.5194/amt-5-73-2012 ◽

2012 ◽

Vol 5 (1) ◽

pp. 73-98 ◽

Cited By ~ 252

Author(s):

S. P. Burton ◽

R. A. Ferrare ◽

C. A. Hostetler ◽

J. W. Hair ◽

R. R. Rogers ◽

...

Keyword(s):

Spectral Resolution ◽

High Spectral Resolution ◽

Color Ratio ◽

Lidar Measurements ◽

Aerosol Classification ◽

Langley Research Center ◽

High Spectral Resolution Lidar ◽

Aerosol Types ◽

532 Nm ◽

Aerosol Type

Abstract. The NASA Langley Research Center (LaRC) airborne High Spectral Resolution Lidar (HSRL) on the NASA B200 aircraft has acquired extensive datasets of aerosol extinction (532 nm), aerosol optical depth (AOD) (532 nm), backscatter (532 and 1064 nm), and depolarization (532 and 1064 nm) profiles during 18 field missions that have been conducted over North America since 2006. The lidar measurements of aerosol intensive parameters (lidar ratio, depolarization, backscatter color ratio, and spectral depolarization ratio) are shown to vary with location and aerosol type. A methodology based on observations of known aerosol types is used to qualitatively classify the extensive set of HSRL aerosol measurements into eight separate types. Several examples are presented showing how the aerosol intensive parameters vary with aerosol type and how these aerosols are classified according to this new methodology. The HSRL-based classification reveals vertical variability of aerosol types during the NASA ARCTAS field experiment conducted over Alaska and northwest Canada during 2008. In two examples derived from flights conducted during ARCTAS, the HSRL classification of biomass burning smoke is shown to be consistent with aerosol types derived from coincident airborne in situ measurements of particle size and composition. The HSRL retrievals of AOD and inferences of aerosol types are used to apportion AOD to aerosol type; results of this analysis are shown for several experiments.

Download Full-text

A global aerosol classification algorithm incorporating multiple satellite data sets of aerosol and trace gas abundances

Atmospheric Chemistry and Physics ◽

10.5194/acp-15-10597-2015 ◽

2015 ◽

Vol 15 (18) ◽

pp. 10597-10618 ◽

Cited By ~ 16

Author(s):

M. J. M. Penning de Vries ◽

S. Beirle ◽

C. Hörmann ◽

J. W. Kaiser ◽

P. Stammes ◽

...

Keyword(s):

Classification Algorithm ◽

Atmospheric Composition ◽

Trace Gas ◽

Aerosol Composition ◽

Ångström Exponent ◽

Aerosol Classification ◽

Aerosol Index ◽

Aerosol Types ◽

Good Agreement ◽

Aerosol Type

Abstract. Detecting the optical properties of aerosols using passive satellite-borne measurements alone is a difficult task due to the broadband effect of aerosols on the measured spectra and the influences of surface and cloud reflection. We present another approach to determine aerosol type, namely by studying the relationship of aerosol optical depth (AOD) with trace gas abundance, aerosol absorption, and mean aerosol size. Our new Global Aerosol Classification Algorithm, GACA, examines relationships between aerosol properties (AOD and extinction Ångström exponent from the Moderate Resolution Imaging Spectroradiometer (MODIS), UV Aerosol Index from the second Global Ozone Monitoring Experiment, GOME-2) and trace gas column densities (NO2, HCHO, SO2 from GOME-2, and CO from MOPITT, the Measurements of Pollution in the Troposphere instrument) on a monthly mean basis. First, aerosol types are separated based on size (Ångström exponent) and absorption (UV Aerosol Index), then the dominating sources are identified based on mean trace gas columns and their correlation with AOD. In this way, global maps of dominant aerosol type and main source type are constructed for each season and compared with maps of aerosol composition from the global MACC (Monitoring Atmospheric Composition and Climate) model. Although GACA cannot correctly characterize transported or mixed aerosols, GACA and MACC show good agreement regarding the global seasonal cycle, particularly for urban/industrial aerosols. The seasonal cycles of both aerosol type and source are also studied in more detail for selected 5° × 5° regions. Again, good agreement between GACA and MACC is found for all regions, but some systematic differences become apparent: the variability of aerosol composition (yearly and/or seasonal) is often not well captured by MACC, the amount of mineral dust outside of the dust belt appears to be overestimated, and the abundance of secondary organic aerosols is underestimated in comparison with GACA. Whereas the presented study is of exploratory nature, we show that the developed algorithm is well suited to evaluate climate and atmospheric composition models by including aerosol type and source obtained from measurements into the comparison, instead of focusing on a single parameter, e.g., AOD. The approach could be adapted to constrain the mix of aerosol types during the process of a combined data assimilation of aerosol and trace gas observations.

Download Full-text

Aerosol-type classification based on AERONET version 3 inversion products

Atmospheric Measurement Techniques ◽

10.5194/amt-12-3789-2019 ◽

2019 ◽

Vol 12 (7) ◽

pp. 3789-3803 ◽

Cited By ~ 12

Author(s):

Sung-Kyun Shin ◽

Matthias Tesche ◽

Youngmin Noh ◽

Detlef Müller

Keyword(s):

East Asia ◽

Biomass Burning ◽

Mineral Dust ◽

Anthropogenic Pollution ◽

Single Scattering ◽

Asian Dust ◽

Dust Particles ◽

Aerosol Types ◽

Type Classification ◽

Aerosol Type

Abstract. This study proposes an aerosol-type classification based on the particle linear depolarization ratio (PLDR) and single-scattering albedo (SSA) provided in the AErosol RObotic NETwork (AERONET) version 3 level 2.0 inversion product. We compare our aerosol-type classification with an earlier method that uses fine-mode fraction (FMF) and SSA. Our new method allows for a refined classification of mineral dust that occurs as a mixture with other absorbing aerosols: pure dust (PD), dust-dominated mixed plume (DDM), and pollutant-dominated mixed plume (PDM). We test the aerosol classification at AERONET sites in East Asia that are frequently affected by mixtures of Asian dust and biomass-burning smoke or anthropogenic pollution. We find that East Asia is strongly affected by pollution particles with high occurrence frequencies of 50 % to 67 %. The distribution and types of pollution particles vary with location and season. The frequency of PD and dusty aerosol mixture (DDM+PDM) is slightly lower (34 % to 49 %) than pollution-dominated mixtures. Pure dust particles have been detected in only 1 % of observations. This suggests that East Asian dust plumes generally exist in a mixture with pollution aerosols rather than in pure form. In this study, we have also considered data from selected AERONET sites that are representative of anthropogenic pollution, biomass-burning smoke, and mineral dust. We find that average aerosol properties obtained for aerosol types in our PLDR–SSA-based classification agree reasonably well with those obtained at AERONET sites representative for different aerosol types.

Download Full-text

A NOVEL INDEX FOR ATMOSPHERIC AEROSOL TYPES CATEGORIZATION WITH SPECTRAL OPTICAL DEPTHS FROM SATELLITE RETRIEVAL

ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences ◽

10.5194/isprs-archives-xli-b8-277-2016 ◽

2016 ◽

Vol XLI-B8 ◽

pp. 277-279 ◽

Cited By ~ 1

Author(s):

Tang-Huang Lin ◽

Gin-Rong Liu ◽

Chian-Yi Liu

Keyword(s):

Atmospheric Aerosols ◽

Single Scattering ◽

Anthropogenic Pollutants ◽

Satellite Retrieval ◽

Modis Aod ◽

Mineral Dusts ◽

Normalized Gradient ◽

Aerosol Index ◽

Aerosol Types ◽

Aerosol Type

In general, the type of atmospheric aerosols can be efficiently identified with the characteristics of optical properties, such as Ångström exponent (AE) and single scattering albedo (SSA). However, the retrieval of SSA is not frequently available to global area which may cause the difficulty in the identification of aerosol type. Since aerosol optical depth (AOD) can be easily requested, a novel index in terms of AOD, Normalized Gradient Aerosol Index (NGAI), is proposed to get over the constraint on SSA providing. With the NGAI derived from MODIS AOD products, the type of atmospheric aerosols can be clearly categorized between mineral dusts, biomass burning and anthropogenic pollutants. The results of aerosol type categorization show the well agreement with the ground-based observations (AERONET) in AE and SSA properties, implying that the proposed index equips highly practical for the application of aerosols type categorization by means of remote sensing. In addition, the fraction of AOD compositions can be potentially determined according to the value of index after compared with the products of CALIPSO Aerosol Subtype.

Download Full-text

A global aerosol classification algorithm incorporating multiple satellite data sets of aerosol and trace gas abundances

Atmospheric Chemistry and Physics Discussions ◽

10.5194/acpd-15-13551-2015 ◽

2015 ◽

Vol 15 (9) ◽

pp. 13551-13605

Author(s):

M. J. M. Penning de Vries ◽

S. Beirle ◽

C. Hörmann ◽

J. W. Kaiser ◽

P. Stammes ◽

...

Keyword(s):

Classification Algorithm ◽

Atmospheric Composition ◽

Trace Gas ◽

Aerosol Composition ◽

Ångström Exponent ◽

Aerosol Classification ◽

Aerosol Index ◽

Aerosol Types ◽

Good Agreement ◽

Aerosol Type

Abstract. Detecting the optical properties of aerosols using passive satellite-borne measurements alone is a difficult task due to the broad-band effect of aerosols on the measured spectra and the influences of surface and cloud reflection. We present another approach to determine aerosol type, namely by studying the relationship of aerosol optical depth (AOD) with trace gas abundance, aerosol absorption, and mean aerosol size. Our new Global Aerosol Classification Algorithm, GACA, examines relationships between aerosol properties (AOD and extinction Ångström exponent from the Moderate Resolution Imaging Spectroradiometer (MODIS), UV Aerosol Index from the second Global Ozone Monitoring Experiment, GOME-2) and trace gas column densities (NO2, HCHO, SO2 from GOME-2, and CO from MOPITT, the Measurements of Pollution in the Troposphere instrument) on a monthly mean basis. First, aerosol types are separated based on size (Ångström exponent) and absorption (UV Aerosol Index), then the dominating sources are identified based on mean trace gas columns and their correlation with AOD. In this way, global maps of dominant aerosol type and main source type are constructed for each season and compared with maps of aerosol composition from the global MACC (Monitoring Atmospheric Composition and Climate) model. Although GACA cannot correctly characterize transported or mixed aerosols, GACA and MACC show good agreement regarding the global seasonal cycle, particularly for urban/industrial aerosols. The seasonal cycles of both aerosol type and source are also studied in more detail for selected 5° × 5° regions. Again, good agreement between GACA and MACC is found for all regions, but some systematic differences become apparent: the variability of aerosol composition (yearly and/or seasonal) is often not well captured by MACC, the amount of mineral dust outside of the dust belt appears to be overestimated, and the abundance of secondary organic aerosols is underestimated in comparison with GACA. Whereas the presented study is of exploratory nature, we show that the developed algorithm is well suited to evaluate climate and atmospheric composition models by including aerosol type and source obtained from measurements into the comparison, instead of focusing on a single parameter, e.g. AOD. The approach could be adapted to constrain the mix of aerosol types during the process of a combined data assimilation of aerosol and trace gas observations.

Download Full-text