Evaluation of Sentinel-2 Spectral Reflectance and Indices to Estimate Grass LAI and CCC in Heterogeneous Grassland

Estimation of biophysical variables such as leaf area index (LAI) and canopy chlorophyll content (CCC) at high spatiotemporal resolution is important for managing natural environments. However, accurate estimation of biophysical variables particularly over natural heterogeneous landscapes remains a challenge. Stepwise multi-linear regression (SMLR) and Random forest (RF) were used to find optimal models for estimating LAI and CCC within Marakele National Park (MNP) in South Africa during peak productivity. Results show that SMLR yielded better LAI estimation with root mean squared error (RMSE) of 0.67 m2 m−2 and mean adjusted error (MAE) of 0.54, explaining 48% of LAI variability, when bands and indices are combined. Further, RF gave better CCC estimation i.e. RMSE and MAE of 17.08 µg cm−2 and 13.18 respectively, explaining about 40% of CCC variability with Sentinel-2 bands only. This study prompts a need for development of locally parameterized types of models over natural heterogeneous environments.