Retrieval of Reflected Shortwave Radiation at the Top of the Atmosphere Using Himawari-8/AHI Data

Sang-Ho Lee,Se-Hun Rim,Kyu-Tae Lee,Hyun-Seok Jung,Il-Sung Zo,Bu-Yo Kim

doi:10.3390/rs10020213

Sang-Ho Lee, Se-Hun Rim + Show 4 more

Open Access

https://doi.org/10.3390/rs10020213

Copy DOI

Journal: Remote Sensing	Publication Date: Feb 1, 2018
Citations: 17	License type: CC BY 4.0

Affiliation: Gangneung–Wonju National University

Abstract

This study developed a retrieval algorithm for reflected shortwave radiation at the top of the atmosphere (RSR). This algorithm is based on Himawari-8/AHI (Advanced Himawari Imager) whose sensor characteristics and observation area are similar to the next-generation Geostationary Korea Multi-Purpose Satellite/Advanced Meteorological Imager (GK-2A/AMI). This algorithm converts the radiance into reflectance for six shortwave channels and retrieves the RSR with a regression coefficient look-up-table according to geometry of the solar-viewing (solar zenith angle, viewing zenith angle, and relative azimuth angle) and atmospheric conditions (surface type and absence/presence of clouds), and removed sun glint with high uncertainty. The regression coefficients were calculated using numerical experiments from the radiative transfer model (SBDART), and ridge regression for broadband albedo at the top of the atmosphere (TOA albedo) and narrowband reflectance considering anisotropy. The retrieved RSR were validated using Terra, Aqua, and S-NPP/CERES data on the 15th day of every month from July 2015 to February 2017. The coefficient of determination (R2) between AHI and CERES for scene analysis was higher than 0.867 and the Bias and root mean square error (RMSE) were −21.34–5.52 and 51.74–59.28 Wm−2. The R2, Bias, and RMSE for the all cases were 0.903, −2.34, and 52.12 Wm−2, respectively.

Highlights

Reflected shortwave radiation at the top of the atmosphere (RSR) is affected by the surface characteristics (15%); atmosphere gases such as aerosols, vapors, etc. (20%); and clouds (65%) [1].In particular, a clear-sky area is greatly influenced by short-wavelength ultraviolet and near-infrared rays depending on surface characteristics, whereas a cloudy-sky area is affected more by the cloud properties [2]
The anisotropy of the atmosphere differs depending on the geometry geometry of the solar-viewing, as characteristics well as the characteristics the Earth’s surface, the results of the solar-viewing, as well as the of the Earth’sofsurface, the results improved in improved in comparison with the data when anisotropy was considered; the errors listed comparison with the Clouds and the Earth Radiant Energy System (CERES) data when anisotropy was considered; the errors are listed inare
This study used Himawari-8/AHI data to develop an algorithm for retrieving RSR

Summary

Introduction

A clear-sky area is greatly influenced by short-wavelength ultraviolet and near-infrared rays depending on surface characteristics, whereas a cloudy-sky area is affected more by the cloud properties [2]. Aerosols such as particulate matter affect cloud distribution and characterization [3], and increase the planetary albedo in relation to absorption and reflection of RSR [4], causing energy imbalance and global cooling [5,6]. Since the 1970s, many studies have been measuring and analyzing RSR using radiative transfer models and satellite-based broadband or narrowband sensor data.

Methods

Discussion

Conclusion