Obata Kenta, Tsuchida Satoshi, Yamamoto Hirokazu, Thome Kurtis
National Institute of Advanced Industrial Science and Technology (AIST), The Institute of Geology and Geoinformation, Tsukuba, Ibaraki 305-8567, Japan.
NASA Goddard Space Flight Center, Greenbelt, MD 20771, USA.
Sensors (Basel). 2017 Aug 4;17(8):1793. doi: 10.3390/s17081793.
Radiometric cross-calibration between the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) and the Terra-Moderate Resolution Imaging Spectroradiometer (MODIS) has been partially used to derive the ASTER radiometric calibration coefficient (RCC) curve as a function of date on visible to near-infrared bands. However, cross-calibration is not sufficiently accurate, since the effects of the differences in the sensor's spectral and spatial responses are not fully mitigated. The present study attempts to evaluate radiometric consistency across two sensors using an improved cross-calibration algorithm to address the spectral and spatial effects and derive cross-calibration-based RCCs, which increases the ASTER calibration accuracy. Overall, radiances measured with ASTER bands 1 and 2 are on averages 3.9% and 3.6% greater than the ones measured on the same scene with their MODIS counterparts and ASTER band 3N (nadir) is 0.6% smaller than its MODIS counterpart in current radiance/reflectance products. The percentage root mean squared errors (%RMSEs) between the radiances of two sensors are 3.7, 4.2, and 2.3 for ASTER band 1, 2, and 3N, respectively, which are slightly greater or smaller than the required ASTER radiometric calibration accuracy (4%). The uncertainty of the cross-calibration is analyzed by elaborating the error budget table to evaluate the International System of Units (SI)-traceability of the results. The use of the derived RCCs will allow further reduction of errors in ASTER radiometric calibration and subsequently improve interoperability across sensors for synergistic applications.
先进星载热发射和反射辐射仪（ASTER）与陆地-中等分辨率成像光谱仪（MODIS）之间的辐射交叉校准已部分用于推导ASTER辐射校准系数（RCC）曲线，该曲线是可见光至近红外波段日期的函数。然而，交叉校准不够准确，因为传感器光谱和空间响应差异的影响没有得到充分缓解。本研究试图使用改进的交叉校准算法评估两个传感器之间的辐射一致性，以解决光谱和空间效应问题，并推导基于交叉校准的RCC，从而提高ASTER校准精度。总体而言，在当前的辐射率/反射率产品中，ASTER第1和第2波段测量的辐射率平均比在同一场景中用MODIS对应波段测量的辐射率分别高3.9%和3.6%，ASTER第3N波段（天底）比其MODIS对应波段小0.6%。两个传感器辐射率之间的百分比均方根误差（%RMSE）对于ASTER第1、第2和第3N波段分别为3.7、4.2和2.3，略高于或低于所需的ASTER辐射校准精度（4%）。通过详细阐述误差预算表来分析交叉校准的不确定性，以评估结果的国际单位制（SI）可追溯性。使用推导得到的RCC将进一步减少ASTER辐射校准中的误差，并随后提高传感器之间协同应用的互操作性。
