December 26, 2017
Biological Processes Dominate Seasonality of Remotely Sensed Canopy Greenness in an Amazon Evergreen Forest
Advancing the biophysical understanding of satellite-detected vegetation seasonality in the tropics.
Satellite observations of Amazon forests show seasonal and interannual variation in canopy greenness, but the underlying biological mechanisms leading to a change in greenness have not been resolved. Here a research team from Brookhaven National Laboratory combined canopy radiative transfer models (RTMs) with field observations of Amazon forest leaf and canopy characteristics to test three hypotheses that could explain seasonality in satellite-observed canopy reflectance: (1) changes in the number of leaves per unit ground area (leaf area index), (2) changes in the fraction of the upper canopy that are leafless, and (3) changes in leaf age. They showed that canopy RTMs driven by these three factors closely matched simulated satellite-observed seasonal patterns, explaining ~70% of variability in a key reflectance-based vegetation index. Leaf area index, leafless crown fraction and leaf age accounted for 1%, 33%, and 66% of modeled seasonality.
The analysis of canopy-scale biophysics rules out satellite artifacts as significant cause of satellite-observed seasonal patterns in greenness at this site and implies that leaf phenology can explain large scale remotely observed patterns. Their study reconciles current controversies about satellite-detected canopy greenness and enables more confident use of satellite observations to study climate-phenology relationships in the tropics.
The average annual cycle (2000-2014) of MODIS satellite observed canopy greenness (i.e., MAIAC EVI minimizes the artifacts from clouds/aerosols and sun-sensor geometry) in a Brazilian Amazon evergreen forest, the Tapajos k67 site, shows strong seasonality. This seasonality is primarily driven by canopy near-infrared (NIR) reflectance. Here, the team combined rich, field measurements of leaf and canopy characteristics with a three-dimensional (3D) RTM (i.e., Forest Light Environment Simulator, FLiES) to interpret MAIAC EVI seasonality. The measurements showed that the comprehensive FLiES model with all phenological input (as “P1+P2+P3”) did a good job at simulating MAIAC EVI and NIR reflectance seasonality. This suggests that biological processes dominate canopy-scale reflectance and greenness seasonality in this tropical forest. Further, the research team did model sensitivity analysis to quantify the relative contribution of each of the three phenological factors including “P1” driven by seasonal change in canopy leaf area index only, “P2” driven by seasonal change in canopy-surface leafless crown fraction alone, and “P3” driven by seasonal change in canopy leaf age demography. Their results suggest that canopy-surface leafless crown fraction and leaf age demography control the seasonality in greenness, they did not observe any direct effect of leaf area index on greenness.
Brookhaven National Laboratory
U.S. Department of Energy, Biological and Environmental Research (SC-33)
Environmental System Science
S.P. Serbin, A. Rogers, and J. Wu in part were supported by the Next-Generation Ecosystem Experiments (NGEE)–Tropics) project. The NGEE-Tropics project is supported by the Office of Biological and Environmental Research within the U.S. Department of Energy Office of Science.
Wu, J., H. Kobayashi, S. C. Stark, and R. Meng, et al. "Biological processes dominate seasonality of remotely sensed canopy greenness in an Amazon evergreen forest". New Phytologist 217 (4), 1507–1520 (2017). https://doi.org/10.1111/nph.14939.