A High Spatiotemporal Enhancement Method of Forest Vegetation Leaf Area Index Based on Landsat8 OLI and GF-1 WFV Data

The leaf area index (LAI) is a crucial parameter for analyzing terrestrial ecosystem carbon cycles and global climate change. Obtaining high spatiotemporal resolution forest stand vegetation LAI products over large areas essential an accurate understanding of ecosystems. This study takes the northwestern part Inner Mongolia Autonomous Region (the northern section Greater Khingan Mountains) in China as research area. It also generates time series product 8-day 30 m growth period from 2013 to 2017 (from 121st 305th day each year). Simulated Annealing-Back Propagation Neural Network (SA-BPNN) model was used estimate Landsat8 OLI, multi-period GaoFen-1 WideField-View satellite images (GF-1 WFV) adaptive reflectance fusion mode (STARFM) predict by combining inversion Global LAnd Surface Satellite-derived (GLASS LAI) products. results showed following: (1) SA-BPNN estimation has relatively accuracy, with R2 = 0.75 RMSE 0.38 model, 0.74 0.17 2016 model. (2) fused good correlation verification measured sample site (R2 0.8775) similarity GLASS product. (3) product, compared interannual trend line, it accords plants seasons. provides theoretical technical reference based on high-score data, important role exploring primary productivity cycle changes future.