SPIE Membership Get updates from SPIE Newsroom
  • Newsroom Home
  • Astronomy
  • Biomedical Optics & Medical Imaging
  • Defense & Security
  • Electronic Imaging & Signal Processing
  • Illumination & Displays
  • Lasers & Sources
  • Micro/Nano Lithography
  • Nanotechnology
  • Optical Design & Engineering
  • Optoelectronics & Communications
  • Remote Sensing
  • Sensing & Measurement
  • Solar & Alternative Energy
  • Sign up for Newsroom E-Alerts
  • Information for:
SPIE Photonics West 2018 | Call for Papers




Print PageEmail PageView PDF

Remote Sensing

Aerosol monitoring by the Geostationary Ocean Color Imager

A newly developed algorithm for processing hourly satellite data will provide information on aerosols over East Asia, including aerosol optical depth over turbid water.
16 December 2010, SPIE Newsroom. DOI: 10.1117/2.1201011.003250

Atmospheric aerosols have important physical and chemical effects on Earth's climate and on the radiation reaching its surface. The environment and climate of East Asia, in particular, are affected by the large amounts of aerosols in the region, which are of several types with different effects. Numerous studies have retrieved aerosol optical properties from multiple visible (VIS)/near-IR (NIR) channels, such as the Sea-viewing Wide Field-of-view Sensor (SeaWiFS), Moderate Resolution Imaging Spectroradiometer (MODIS), and Medium-spectral Resolution Imaging Spectrometer (MERIS). However, all of these sensors, aboard low Earth orbit satellites, have provided aerosol information only once a day.

The Geostationary Ocean Color Imager (GOCI) is the first multi-channel VIS/NIR ocean color sensor operating in geostationary orbit. It is located onboard the Communication, Ocean, and Meteorological Satellite (COMS) launched on 27 June 2010 to observe ocean color around the Korean Peninsula. The GOCI has eight spectral channels at 412, 443, 490, 555, 660, 680, 745, and 865nm, with spatial coverage of 2500km×2500km centered at 36°N and 130°E and a resolution of 500m.1 By taking advantage of a geostationary platform, GOCI can provide hourly spectral images that can be used for continuous monitoring of aerosols as well as ocean color over cloud-free areas. Aerosol optical properties can be retrieved from the reflectance measured by the GOCI.

In this article, we introduce an aerosol retrieval algorithm that has been developed for the GOCI.2 The algorithm retrieves the aerosol optical depth (AOD), fine-mode fraction (FMF), and aerosol type at 500m×500m resolution. In particular, we have tried to retrieve the AOD over turbid water, which has been difficult because of high surface reflectance. Since the initial operational test of the GOCI is still under way, MODIS data is used as a proxy.

To develop an optimized algorithm for the GOCI's target area, the optical properties of aerosols are analyzed using extensive observations by AERONET sun photometers to generate a lookup table (LUT). From the AERONET database, three fine-mode and two coarse-mode aerosol models are constructed. Next, the LUT is calculated by combining the fine- and coarse-mode models by changing the FMF to that appropriate for the respective aerosol model.

Figure 1. An example of a lookup table (LUT) computed by using the radiative transfer model rstar5b developed at the University of Tokyo for different aerosol loadings (AOD at 550nm) and FMF, assuming a combination of non-absorbing fine-mode and dust aerosols. TOA: Time of arrival.

Figure 1 shows an example of a LUT for a combination of non-absorbing fine-mode aerosols and dust (coarse-mode aerosols). The spectral response function of MODIS was used so that the algorithm could be applied to a real case. From the LUTs and time of arrival (TOA) reflectances at 640 and 860nm, the AOD and FMF for both fine- and coarse-mode aerosols are retrieved simultaneously. The calculated TOA reflectances of the other channels are then compared with the observed values to select the appropriate aerosol models (see Figure 2 for an example of GOCI products).

Figure 2. An example of products retrieved from the GOCI on 27 January 2006.

Since a large portion of the Yellow Sea, which is an important pathway of aerosols, is covered by persistent turbid water, we needed to develop an aerosol retrieval algorithm that could be applied over turbid water. Figure 3 shows the frequency distributions of the surface reflectance ratio (SRR) between 640 and 860nm over the Yellow Sea, where both turbid and clear water exist, and an area of clear water. Since the SRR of turbid water is higher than that of clear water, the threshold of detection is 2.5. Thus, the surface reflectance of the turbid water area can be determined from a clear-sky composite of the Rayleigh- and gas-corrected TOA reflectance.

Figure 3. Cumulative histogram of surface reflectance ratio (SRR) for 640 and 860nm over (left) the Yellow Sea and (right) clear water in 2006, calculated using MODIS data. Each gray layer from bottom to top represents a month in 2006 from January to December. Surface reflectance (SR) is determined by 30-day second minimum Rayleigh- and gas-corrected time of arrival reflectance.

To estimate the accuracy of the AOD retrieved by the GOCI algorithm, it is compared with those from AERONET sun photometer observations in 2006, as shown in Figure 4. When the turbid water is not corrected for, the coefficients of determination are 0.51, 0.92, and 0.78 for Anmyon, Gosan, and Shirahama, respectively. Compared with other stations, the agreement at Anmyon station, which is located close to the turbid water of the Yellow Sea, is poor, and the current algorithm significantly overestimates the AOD. On the other hand, the algorithm produces a reliable AOD product at Gosan and Shirahama, which are located near clear ocean water.

Figure 4. The AOD retrieved in 2006 with the GOCI algorithm by using MODIS data is compared with those of AERONET observations (left) at Anmyon without turbid water correction, at Gosan, and at Shirahama and (right) at Anmyon after turbid water correction.

After turbid water correction, the coefficient of determination at Anmyon improved significantly from 0.51 to 0.89 with an increase in regression slope (from 0.69 to 1.09) and a decrease in the y-intercept (0.31 to 0.015). This result demonstrates that turbid water correction should be applied over the Yellow Sea for better accuracy as well as better coverage, although this procedure may cause discontinuities at the boundary of turbid water.

The main scientific objective of aerosol products from the GOCI is environmental monitoring. Its capacity for continuous observation allows the detection of diurnal cycles. Moreover, as long-term data from the mission accumulates, it will become possible to assess the role of aerosols in climate change. The information on aerosol type provided by AOD measurements based on continuous observation will be valuable in many fields, in particular for monitoring the environment and climate change and investigating the effect of different aerosol types over East Asia. The algorithm introduced in this study has been updated to improve its accuracy. The results obtained with the developed algorithm will be analyzed over the long term to diagnose various sources of error.

This article is adapted from portions of a paper2 appearing in the journal Remote Sensing of Environment published in 2010 by Elsevier Inc. All rights reserved, with permission from Elsevier.

Jhoon Kim
Yonsei University
Seoul, South Korea

Jhoon Kim developed an algorithm for retrieving aerosol properties from the GOCI and Meteo Imager (MI) onboard the COMS. He is a principal investigator on the Geostationary Environment Monitoring Spectrometer (GEMS) mission to be launched in 2017 or 2018. He is a member of the Atmospheric Composition Constellation of the Committee on Earth Observation Satellites. He is also co-chair of Working Group 8 (Atmosphere, Climate, and Weather) of Commission VIII on Remote Sensing Applications and Policies of the International Society for Photogrammetry and Remote Sensing.

Jaehwa Lee
Department of Atmospheric Sciences
Yonsei University
Seoul, South Korea
Chul H. Song
Department of Environmental Engineering
Gwangju Institute of Science and Technology
Gwangju, South Korea
Joo-Hyung Ryu, Yu-Hwan Ahn
Ocean Satellite Research Group
Korea Ocean Research and Development Institute
Ansan, South Korea
Chang-Keun Song
National Institute of Environmental Research
Incheon, South Korea