A High-Quality Reprocessed MODIS Leaf Area Index Dataset (HiQ-LAI)

The High-Quality Leaf Area Index (HiQ-LAI) is derived from reprocessed MODIS LAI C6.1 product by SpatioTemporal Information Compositing Algorithm (STICA). This method integrates information from multiple dimensions, including pixel quality information, spatiotemporal correlation, and original observations, to improve the raw MODIS LAI retrievals with poor quality. The HiQ-LAI covers the period from 2000 to 2022, with spatial resolutions of 500m/5km for global vegetation area and temporal resolutions of 8 days. Ground-based verification results show that HiQ-LAI performs better than the original MODIS product (MOD15A2H C6.1). Time series curves of the HiQ-LAI exhibit reduced abnormal fluctuations and better alignment with expected phenological patterns. Additionally, the agreement with ground measurements increases gradually as raw data quality decreases. HiQ-LAI was found to be more continuous and consistent than MODIS LAI on a global scale from both spatial and temporal perspectives, especially in the equatorial regions where optical remote sensing usually cannot achieve good performance. Thus, We anticipate that HiQ-LAI with better spatio-temporal continuity will better support varying global LAI time series applications. Here, we offer a product version with a spatial resolution of 5km and a temporal resolution of 8 days. Another version has a spatial resolution of 500 meters and is available through Google Earth Engine (https://code.earthengine.google.com/?asset=projects/verselab-398313/assets/HiQ_LAI/wgs_500m_8d). More details about HiQ-LAI can be found at https://github.com/tiramisu18/HiQ-LAI高质量叶面积指数（High-Quality Leaf Area Index, HiQ-LAI）数据集由时空信息合成算法（SpatioTemporal Information Compositing Algorithm, STICA）对再处理后的MODIS LAI C6.1产品加工生成。该算法整合多维度信息，涵盖像元质量信息、时空相关性与原始观测数据，以此优化质量欠佳的原始MODIS LAI反演结果。HiQ-LAI的时间覆盖范围为2000年至2022年，覆盖全球植被区域，空间分辨率分为500米与5千米两级，时间分辨率为8天。地面验证结果表明，HiQ-LAI的表现优于原始MODIS产品（MOD15A2H C6.1）。其时间序列曲线的异常波动更少，且与预期物候模式的匹配度更佳。此外，随着原始数据质量下降，HiQ-LAI与地面实测值的一致性会逐步提升。研究显示，在全球尺度的时空维度上，HiQ-LAI的连续性与一致性均优于MODIS LAI，尤其在光学遥感通常难以取得良好效果的赤道区域。因此，我们预计这款时空连续性更优的HiQ-LAI数据集，将能更好地支撑各类全球叶面积指数时间序列相关应用。本次发布的产品版本空间分辨率为5千米，时间分辨率为8天。另一版本空间分辨率为500米，可通过谷歌地球引擎（Google Earth Engine）获取，访问链接为：https://code.earthengine.google.com/?asset=projects/verselab-398313/assets/HiQ_LAI/wgs_500m_8d。更多关于HiQ-LAI的详细信息可访问：https://github.com/tiramisu18/HiQ-LAI