NMSC National Meteorological Satellite Center

GK-2A Visible/Near-infrared Calibration

     The Advanced Meteorological Imager (AMI) on GK2A has four visible channels and two near-infrared channels (Table 1).

Table 1. GK2A AMI visible/near-infrared channels

Onboard Calibrator

     The AMI has a solar diffuser as a solar calibration target (SCT). The SCT was the first solar diffuser installed on the KMA geostationary meteorological satellite imager. It is used as a bright reference for the visible (VIS) and near-infrared (NIR) bands, while space view observations are used as dark reference. By using these calibration targets, linear and bias calibration coefficients are derived to convert the counts of raw detector sample data into radiances. However, if there are issues with the absolute calibration targets, vicarious calibration techniques should be used, along with stable calibration monitoring.

Algorithm

     The NMSC employs vicarious calibration techniques using Earth targets (ocean, desert, water cloud (WC), and deep convective cloud (DCC)), ray-matching, and GSICS DCC. The Earth target algorithm is based on the development of meteorological data processing system for the Communication, Ocean and Meteorological Satellite (COMS) ATBD. The ray-matching and GSICS DCC algorithms have been developed by the GSICS community, and the reference papers can be found at the end of this page.

Table 2. GK2A AMI VIS/NIR calibration methods

※ RTM : Radiative Transfer Model
      6S : Second Simulation of the Satellite Signal in the Solar Spectrum
      SBDART : Santa Barbara DISORT Atmospheric Radiative Transfer

ㆍ Ocean (RTM)

     Ocean targets are usually homogeneous and dark, and are therefore handled only with clear pixels without clouds for sensor calibration. The reflectance of the ocean is affected by aerosol, as the surface reflectance is weak. Therefore, aerosol optical thickness (AOT) should be considered as ancillary data in radiative transfer modeling. Furthermore, the near-homogeneous surface reflectance of the ocean provides an advantage in that the radiance can be calculated without bidirectional reflectance distribution function (BRDF).

ㆍ Desert (RTM)

     The NMSC conducts calibration for 11 fixed desert targets located in the Simpson Desert in Australia. Due to the low impact of aerosols over high surface reflectivity in deserts, there is no need to use AOT as ancillary data from other satellites. However, accurate surface BRDF needs to be used for model input, as the variation of surface BRDF is low with time. The surface BRDF can be updated less frequently compared to other targets.

ㆍ Water Cloud (RTM)

     The atmosphere and surface effects can be minimized in the simulation of TOA radiance over WC targets, due to the strong reflection of the cloud layer. Actual tests for WC reflectance sensitivity show that TOA reflectance is more dependent on cloud optical properties than on surface and atmospheric effects. In addition, the reflectance is more sensitive to cloud optical thickness (COT) than to particle sizes of clouds, as the absorption rate of visible rays into cloud particles is low. Thus, cloud optical properties are required as model input for more accurate results.

ㆍ DCC (RTM)

     Among the targets, which include ocean, desert, WC, and DCC, the DCC target has the highest reflectance among the targets. The vertical distribution of DCC is assumed to span from 1 km to 15 km. COT and particle size of the DCC are assumed to be 200 and 20 µm, respectively. The pixels with the brightness temperature (TB) of IR1 (10.8µm) less than 190K are selected as DCC for calibration.

ㆍ GSICS DCC

     The GSICS DCC method involves utilizing both geostationary earth orbit (GEO) and low earth orbit (LEO) satellites observing DCC targets separately. Using this method, DCC pixels are selected for each month from GEO and LEO observations, and these selected pixels are compared using monthly statistics. This method assumes that both GEO and LEO have observed the same DCC target. By using statistics such as mean, median, and mode, it is possible to monitor the distribution of radiance values or the degradation rate of VIS/NIR channels over time.

ㆍ Ray-Matching

     The ray-matching method involves the intercomparison of the GEO and the well-calibrated LEO satellites. Each grid (pixel) of the GEO-LEO is matched based on satellite and solar angle, and spatial and temporal data are followed by a threshold as shown in Table 4. This method has the advantage of providing results all the time, as it does not require the selection of specific reference targets and monitors a wide and diverse ranges of reflectance. For instance, the DCC method is only applicable for high reflectance targets.

Table 3. Matched channels for GSICS DCC and ray-matching methods.

Table 4. Ray-matching method thresholds

Calibration

     The NMSC provides four types of outcomes on this web page. These include two time series of regression coefficients and ratios, as well as a scatter plot, for methods.

Regression coefficients between the observed and reference reflectance

     The time series of regression slopes (C1) and intercepts (C0) are computed between the observed and reference reflectance. The C1 and C0 values displayed on the web page represent the 29-day moving average value (±14 days).

     Reflectance (Observed)= C1 * Reflectance (Reference) + C0

Ratio of observed reflectance to reference reflectance for each target

     The time series of ratios of observed reflectance to reference reflectance represent the 30-day moving average value.

     Ratio=Reflectance(Observed)/Reflectance(reference)

Scatter plot between observed and reference reflectance

     A comparison of reflectance between observation and simulation is provided in the scatter plot, which includes a regression line.

Reference

Doelling D., Dan Morstad, Rajendra Bhatt, Benjamin Scarino 2011: Algorithm Theoretical Basis Document (ATBD) for Deep Convective Cloud (DCC) technique of calibrating GEO sensors with Aqua-MODIS for GSICS, GSICS ATBDs.

Doelling, D., Rajendra Bhatt, Dan Morstad, Benjamin Scarino, 2011: Algorithm Theoretical Basis Document (ATBD) for ray-matching technique of calibrating GEO sensors with Aqua-MODIS for GSICS, GSICS ATBDs.

Chun, H. W., & Sohn, B. J., 2006: Calibration for the solar channel using MODIS-derived BRDF parameters over Australian desert targets. Target, 4(6), 8.

Ham, S. H. 2006: Improvement of cloud's spectral reflection simulation using MODIS cloud products. 12th Conference on Atmospheric Radiation.

KMA, 2009: Development of Meteorological Data Processing System for Communication, Ocean and Meteorological Satellite(ATBD).

Sohn, B. J., Seung-Hee Ham, Ping Yang, 2009: Possibility of the Visible-Channel Calibration Using Deep Convective Clouds Overshooting the TTL. J. Appl. Meteor. Climatol., 48, 2271-2283.