Spaceborne Remote Sensing of African Tropical Forest Productivity and Drivers
Russell Doughty, University of Oklahoma, russell.doughty@ou.edu (Presenter)
Michael Wimberly, University Of Oklahoma, Norman, mcwimberly@ou.edu
Dan Wanyama, University of Oklahoma, wanyama@ou.edu
Helene Peiro, University of Oklahoma, helene.peiro@ou.edu
Sean Crowell, University of Oklahoma, scrowell@ou.edu
The intra- and inter-annual variability of global atmospheric carbon dioxide concentrations is largely driven by changes in the terrestrial uptake of carbon dioxide through photosynthesis. Tropical forests account for about one third of global terrestrial photosynthesis, but relatively less is known about the seasonality and drivers of tropical forest productivity in African landscapes due to a lack of eddy covariance and airborne XCO2 data.
In an effort to help fill this knowledge gap, we used (1) Level 2 spaceborne solar-induced chlorophyll fluorescence (SIF) from OCO-2, OCO-3, and TROPOMI; (2) MODIS- and VIIRS-based vegetation indices (EVI, NIRv, NDVI, and LSWI); and (3) CHIRPS precipitation to investigate the seasonality and drivers of tropical forest productivity in 12 of Africa’s tropical forest ecoregions.
We found that SIF was significantly correlated with precipitation across all ecoregions, but that the strength of the correlation was higher for ecoregions with less total annual rainfall. We did not observe this phenomenon with the vegetation indices.
Lag correlations between SIF and precipitation indicated that SIF and precipitation was less coupled in the wetter ecoregions. We also found that SIF was overall more strongly correlated with precipitation than the vegetation indices, and that the distinctive double-peak seasonality of fluorescence, which we observed in most ecoregions, was either muted or not found in the seasonality of the vegetation indices.
Further work is needed to establish whether the seasonality of SIF for each ecoregion is robust across sun-sensor geometries and observations of various cloud-fraction thresholds. Analyses and comparisons of commonly used GPP products and SIF would also provide more insight into the degree to which these data sources agree and constrain our knowledge of the timing and magnitude of the uptake of atmospheric carbon dioxide in African tropical forest landscapes.
Associated Project(s):
Poster Location ID: 3-34
Presentation Type: Poster
Session: Poster Session 3
Session Date: Thu (May 11) 3:00-5:00 PM
CCE Program: TE