Estimation of canopy clumping index from MISR and MODIS sensors using the normalized difference hotspot and darkspot (NDHD) method: The influence of BRDF models and solar zenith angle

Clumping index (CI) describes the spatial distribution pattern of foliage, and is a critical parameter to characterize the terrestrial ecosystem and model land-surface processes. Global and regional scale CI maps have been generated from POLDER, MODIS, and MISR sensors based on an empirical relationship with the normalized difference between hotspot and darkspot (NDHD) index by previous studies. However, the hotspot, darkspot, and CI values can be considerably different from different bidirectional reflectance distribution function (BRDF) models and solar zenith angles (SZA). In this study, we evaluated the effects of different configurations of BRDF models and SZA values on CI estimation using the NDHD method. CI maps derived from MISR and MODIS were compared with reference data at the VALERI sites. Results show that for moderate to least clumped vegetation (CI > 0.5), CI values retrieved with the observational SZA agree well with field measurements, while SZA = 0° underestimates, and SZA = 60° overestimates. For highly clumped (CI < 0.5) and sparsely vegetated areas (FCOVER < 25%), the Ross-Li model with 60° SZA is recommended for CI estimation. The best NDHD configuration was used to estimate a 15-year time series CI from MODIS BRDF data. The time series CI shows a reasonable seasonal trajectory, and varies consistently with the MODIS leaf area index (LAI). This study enables better usage of the NDHD method for CI estimation, and can be a useful reference for research on CI validation.