Aerosol optical depth (AOD) retrieval for atmospheric correction in Landsat-8 imagery using second simulation of a satellite signal in the solar spectrum-vector (6SV)

Abdul Basith,Gathot Winarso,Ratna Prastyani,Muhammad Ulin Nuha

doi:10.21924/cst.4.2.2019.122

Abstract

Atmospheric correction has been challenging task in digital image processing. It requires several atmospheric parameters in order to obtain accurate surface reflectance of objects within the image scene. One of the most crucial parameters required for accurate atmospheric correction is aerosol optical depth (AOD). AOD can be obtained by in-situ measurement or estimated from remote sensing observation. In this experiment, atmospheric correction was performed using second simulation of a satellite signal in the solar spectrum-vector (6SV) algorithm on Landsat-8 imagery in which AOD parameter was retrieved from surface reflectance inversion involving daily-global surface reflectance product of moderate resolution imaging spectroradiometer (MODIS). Furthermore, AOD retrieved from surface reflectance inversion was also validated using ground-based sun photometer observation data from aerosol robotic network (AERONET) station in Bandung, Indonesia. Our experiment shows the consistency between AOD from surface reflectance inversion and AOD from ground-based observation. Finally, 6SV was performed on Landsat-8 imagery to obtain the surface reflectance. We further compared surface reflectance of 6SV atmospheric correction and surface reflectance of Landsat-8 Level 2 product. The atmospherically corrected image also shared agreeable result with Landsat 8 Level-2 product.

Highlights

Atmospheric correction has been a real challenge in preimage processing
The objectives of this study are (1) to demonstrate aerosol optical depth (AOD) retrieval using moderate resolution imaging spectroradiometer (MODIS) products and (2) to perform atmospheric correction using 6SV in Landsat-8 imagery based on the retrieved AOD parameter
Landsat-8 Level 1 product is represented by digital number values which can be converted to Top-of-Atmosphere reflectance (TOA)

Summary

Introduction

Atmospheric correction has been a real challenge in preimage processing. Atmospheric effects due to aerosol in particular can be found extremely difficult to correct [1]. Aerosols are basically liquid or solid particles suspended in the atmosphere with variety of sizes, shapes and compositions [2]. It affects Earth’s climate systems as it plays an essential role in Earth’s solar radiation budget [3]. The presence of aerosols in the atmosphere affect the recorded data by spacebased instruments due to its solar radiation scattering and absorption properties

Objectives

Methods

Results