Phenology-adaptive / static MODIS composites: seamless multi-annual (2005+-2) and seasonal (POS, EOS, MOS) image composites across Zambia

This dataset contains seasonal, multi-annual MODIS reflectance composites generated across Zambia for 2005 +-2 years. The data is associated to following paper: D. Frantz, A. Röder, M. Stellmes, and J. Hill (2017): Phenology-Adaptive Pixel-Based Compositing using Optical Earth Observation Imagery. Remote Sensing of Environment 190, 331-347. DOI: 10.1016/j.rse.2017.01.002 A parametric compositing technique was employed to produce composites from very dense MODIS reflectance images (MOD09GA product, 1-2 day temporal resolution). Composites were generated for three phenological seasons: peak of season (POS), end of season (EOS) and minimum of season (MOS). Two different sets were produced: (1) phenology-adaptive composites that explicitly consider the land surface phenology of each pixel and (2) static composites that use a fixed and global target DOY representative for the seasons. The images are in Standard ENVI Format. Naming convention: 20050228_PBC_INF_zambia.dat - First 8 digits: Mean date of selected observations; the temporal sequence is POS, EOS, and MOS - PBC_INF / PBC_REF: composite criteria / reflectance composite The reflectance composites are 6-band (0.469µm, 0.555µm, 0.645µm, 0.859µm, 1.640µm, 2.130µm), 16bit bsq images. The composite criteria images are 4-band (number of observations, selected DOY, diff. to target DOY, diff. to target year), 16bit bsq images. Acknowledgements The MODIS MOD09GA data products were retrieved from the online Data Pool, courtesy of the NASA Land Processes Distributed Active Archive Center (LP DAAC), USGS/Earth Resources Observation and Science (EROS) Center, Sioux Falls, South Dakota, https://lpdaac.usgs.gov/data_access/data_pool. The MODIS MOD13Q1/MYD13Q1 data products were retrieved from the online Data Pool, courtesy of the NASA Land Processes Distributed Active Archive Center (LP DAAC), USGS/Earth Resources Observation and Science (EROS) Center, Sioux Falls, South Dakota, https://lpdaac.usgs.gov/data_access/data_pool. Land Surface Phenology was inferred from MOD13Q1/MYD13Q1 data with the Spline Analysis of Time Series (SpliTS) algorithm, courtesy of Dr. Sebastian Mader, Trier University, Trier, Germany.本数据集包含2005±2年期间覆盖赞比亚全境生成的季节性多年MODIS（中分辨率成像光谱仪，Moderate Resolution Imaging Spectroradiometer）反射率合成产品。本数据集关联以下学术论文：D. Frantz、A. Röder、M. Stellmes与J. Hill（2017）：《基于光学对地观测影像的物候自适应像素级合成方法》，《环境遥感（Remote Sensing of Environment）》190卷，第331-347页。DOI: 10.1016/j.rse.2017.01.002 研究采用参数化合成技术，从时间分辨率为1-2天的高密度MODIS反射率影像（MOD09GA产品）中生成合成影像。合成影像覆盖3个物候季节：生长季峰值（Peak of Season, POS）、生长季末期（End of Season, EOS）与生长季谷值（Minimum of Season, MOS）。本次共生成两组合成产品：（1）逐像素考虑地表物候特征的物候自适应合成产品；（2）采用固定全球目标年积日（Day of Year, DOY）表征对应季节的静态合成产品。 所有影像均采用标准ENVI格式存储。命名规则示例：20050228_PBC_INF_zambia.dat —— 前8位数字为所选观测数据的平均日期；影像时间序列依次为POS、EOS与MOS；PBC_INF与PBC_REF分别对应合成判据产品与反射率合成产品。 反射率合成产品为6波段影像，波段中心波长分别为0.469μm、0.555μm、0.645μm、0.859μm、1.640μm与2.130μm，采用16位波段顺序（Band Sequential, bsq）格式存储。合成判据影像为4波段影像，分别包含观测数量、所选年积日、与目标年积日的差值、与目标年份的差值，同样采用16位bsq格式存储。 致谢 MODIS MOD09GA数据产品从美国国家航空航天局（NASA）陆地过程分布式主动存档中心（LP DAAC）、美国地质调查局（USGS）地球资源观测与科学（EROS）中心（南达科他州苏福尔斯）的在线数据池获取，获取链接：https://lpdaac.usgs.gov/data_access/data_pool。 MODIS MOD13Q1/MYD13Q1数据产品同样从上述机构的在线数据池获取。 地表物候信息通过样条时间序列分析（Spline Analysis of Time Series, SpliTS）算法从MOD13Q1/MYD13Q1数据中提取，该方法由德国特里尔大学的Sebastian Mader博士提供。