Surface Albedo/BRDF Parameters (Terra/Aqua MODIS)

Spatially and temporally complete surface spectral albedo/BRDF products over the ARM SGP area were generated using data from two Moderate Resolution Imaging Spectroradiometer (MODIS) sensors on Terra and Aqua satellites. A landcover-based fitting (LBF) algorithm is developed to derive the BRDF model parameters and albedo product (Luo et al., 2004a). The approach employs a landcover map and multi-day clearsky composites of directional surface reflectance. The landcover map is derived from the Landsat TM 30-meter data set (Trishchenko et al., 2004a), and the surface reflectances are from MODIS 500m-resolution 8-day composite products (MOD09/MYD09). The MOD09/MYD09 data are re-arranged into 10-day intervals for compatibility with other satellite products, such as those from the NOVA/AVHRR and SPOT/VGT sensors. The LBF method increases the success rate of the BRDF fitting process and enables more accurate monitoring of surface temporal changes during periods of rapid spring vegetation green-up and autumn leaf-fall, as well as changes due to agricultural practices and snowcover variations (Luo et al., 2004b, Trishchenko et al., 2004b). Albedo/BRDF products for MODIS on Terra and MODIS on Aqua, as well as for Terra/Aqua combined dataset, are generated at 500m spatial resolution and every 10-day since March 2000 (Terra) and July 2002 (Aqua and combined), respectively. The purpose for the latter product is to obtain a more comprehensive dataset that takes advantages of multi-sensor observations (Trishchenko et al., 2002). To fill data gaps due to cloud presence, various interpolation procedures are applied based on a multi-year observation database and referring to results from other locations with similar landcover property. Special seasonal smoothing procedure is also applied to further remove outliers and artifacts in data series.

基于Terra和Aqua卫星上的两台中分辨率成像光谱仪（MODIS）传感器数据，生成了ARM南部大平原（ARM SGP）区域的时空完整地表光谱反照率/双向反射分布函数（BRDF）产品。研究开发了一种基于土地覆盖的拟合（LBF）算法，用于推导BRDF模型参数和反照率产品（Luo等，2004a）。该方法采用土地覆盖图和多日晴空方向地表反射率合成数据。土地覆盖图源自陆地卫星专题制图仪（Landsat TM）30米数据集（Trishchenko等，2004a），地表反射率数据则来自MODIS 500米分辨率8天合成产品（MOD09/MYD09）。为与NOVA/AVHRR和SPOT/VGT等其他卫星产品兼容，MOD09/MYD09数据被重新整理为10天间隔。LBF方法提高了BRDF拟合过程的成功率，能够更准确地监测春季植被快速返青期、秋季落叶期的地表时间变化，以及农业活动和积雪变化引起的地表变化（Luo等，2004b；Trishchenko等，2004b）。Terra卫星MODIS、Aqua卫星MODIS以及Terra/Aqua联合数据集的反照率/BRDF产品，分别从2000年3月（Terra）和2002年7月（Aqua及联合数据集）开始，以500米空间分辨率每10天生成一次。后者（联合数据集）的目的是获取更全面的数据集，以利用多传感器观测的优势（Trishchenko等，2002）。为填补因云层存在导致的数据缺口，研究基于多年观测数据库，并参考具有相似土地覆盖属性的其他区域结果，应用了多种插值程序。此外，还应用了特殊的季节性平滑程序，以进一步去除数据序列中的异常值和伪影。