Biological processes dominate seasonality of remotely sensed canopy greenness in an Amazon evergreen forest

• DOI: 10.1111/nph.14939
• Publication Date: 2017-12-24T11:07:52Z
• Authors: Jin Wu; Hideki Kobayashi; Scott C. Stark; Ran Meng; Kaiyu Guan; Ngoc Nguyen Tran; Sicong Gao; Wei Yang; Natalia Restrepo‐...; Tomoaki Miura; Raimundo Cosme Ol...; Alistair Rogers; Dennis G. Dye; Bruce W. Nelson; Shawn P. Serbin; Alfredo R. Huete; Scott R. Saleska
• Keywords: Seasonality; Tree canopy; Enhanced vegetation index; Evergreen forest

Summary Satellite observations of Amazon forests show seasonal and interannual variations, but the underlying biological processes remain debated. Here we combined radiative transfer models ( RTM s) with field forest leaf canopy characteristics to test three hypotheses for satellite‐observed reflectance seasonality: changes in area index, canopy‐surface leafless crown fraction and/or demography. Canopy s PROSAIL FL i ES ), driven by these factors combined, simulated patterns well, explaining c . 70% variability a key reflectance‐based vegetation index MAIAC EVI, which removes artifacts that would otherwise arise from clouds/aerosols sun–sensor geometry). Leaf demography independently accounted 1, 33 66% ‐simulated EVI seasonality, respectively. These also strongly influenced modeled near‐infrared NIR ) reflectance, why both observed , is especially sensitive captures dynamics well. Our improved analysis canopy‐scale biophysics rules out satellite as significant causes at this site, implying aggregated phenology explains larger scale remotely patterns. This work significantly reconciles current controversies about satellite‐detected phenology, improves our use study climate–phenology relationships tropics.