Combination of Cross- and Inter-Band Radiometric Calibrations for a Hyperspectral Sensor Using Model-Based Spectral Band Adjustment

Hiroki Mizuochi,Hirokazu Yamamoto,Satoshi Tsuchida,Kenta Obata,Satoru Yamamoto

doi:10.3390/rs12122011

Abstract

Recently, the growing number of hyperspectral satellite sensors have increased the demand for a flexible and robust approach to their calibration. This paper proposes an operational method for the simultaneous correction of inter-sensor and inter-band biases in hyperspectral sensors via the soil line concept for spectral band adjustment. Earth Observing-1 Hyperion was selected as an example, with the Terra Moderate Resolution Imaging Spectroradiometer (MODIS) as a reference. The results over the Railroad Valley Playa calibration site indicated that the discrepancy in the analogous bands between Hyperion and MODIS during 2001–2008 was approximately 4–6% and 7–9% of the root-mean-square error in the top-of-atmosphere (TOA) radiance at the visible and near-infrared region and shortwave infrared region, respectively. For all Hyperion bands, the relative cross-calibration coefficients during this period were calculated (typically ranging from 0.9 to 1.1) to correct the Hyperion TOA radiance to be consistent with the MODIS and the other Hyperion bands. The application of the proposed approach could allow for more flexible cross-calibration of irregular-orbit sensors aboard the International Space Station.

Highlights

The increasing number of hyperspectral satellite sensor missions have revealed a wide range of applications in Earth observation [1]
The largest RMSEk in the visible and near-infrared (VNIR) region was found in the blue band (MODIS band 3) because of the underestimation of the TOA radiance by Hyperion, which slightly overestimated the TOA radiance in the other three bands
This study proposed an operational approach for the simultaneous correction of inter-sensor and intra-sensor biases in hyperspectral sensors, using Earth Observing-1 (EO-1) Hyperion and Terra Moderate Resolution Imaging Spectroradiometer (MODIS) as examples

Summary

Introduction

The increasing number of hyperspectral satellite sensor missions have revealed a wide range of applications in Earth observation [1]. The synergy between multiple satellite sensors offers further opportunities for long-term and broad-scale terrestrial monitoring but requires the sensors to be radiometrically consistent with each other (e.g., [5]). For this purpose, researchers have developed various cross-calibration methods, including a comparison between sensors aboard the same satellite platform [6,7], those aboard different platforms but with near-simultaneous nadir overpass (SNO) [8,9,10], and those without SNO but with similar sensor and solar geometries [11], as well as a statistical comparison of pseudo-invariant calibration sites [12] and deep convective clouds [13]. The limited opportunity to find ideal matched pairs between spatially narrow-swath sensors is often an issue for hyperspectral sensors [1]

Objectives

Methods

Results

Discussion

Conclusion