Development of a Global 1-km Vegetative Canopy Dataset Using Multi-platform Satellite Measurements

The vegetative canopy (e.g. forests, grassland and crops) plays an essential role in the Earth’s weather and climate. For instance, water vapor released from vegetation during photosynthesis could alter the surface heat flux and condensation process in the air, leading to potential cloud formation and eventually precipitation. Volatile organic compounds (VOC) emitted from vegetation impact trace gas and aerosol formation, while the canopy also serves as a major sink of such concentrations via dry deposition. Furthermore, previous studies have shown that the canopy environment has its own distinct physicochemical characteristics, which affect the overall physical and chemical predictions in coupled numerical weather prediction and atmospheric composition models. Thus, it is important for such models to include a comprehensive and consistent set of canopy parameters to properly address the effects of the canopy environment on the Earth System.
Here we present a novel global, satellite-based dataset containing major canopy parameters that can be used in coupled NWP and atmospheric composition models, including monthly leaf area index (LAI) and canopy clumping index (CLU), as well as annual forest fraction (FC), forest canopy height (FH), and vertical distribution of plant area volume density (PAVD). Products from various satellites are used, including the Moderate Resolution Imaging Spectroradiometer (MODIS), the Visible Infrared Imaging Radiometer Suite (VIIRS), and the Global Ecosystem Dynamics Investigation (GEDI). All variables share a consistent rectangular global grid coordinate (-90° – 90° in latitude, 0° – 360° in longitude), with 1 km (~0.01°) spatial resolution. Given the instrument limitations in high latitude regions, gap-filling techniques (e.g. Kriging) are used to get comprehensive global coverage while keeping the spatiotemporal and vegetation-dependent variabilities. Moreover, canopy parameters are cross-compared to better understand the uncertainties of the dataset. One of the preliminary results shows that integrated GEDI PAVD (i.e. the plant area index (PAI)) is generally lower than the PAI values based on VIIRS LAI; thus, it may affect the quality of the extended PAVD in high latitude regions.