You are here: Home / Research Topics / Research Highlights / Individual Highlight

Research Highlights

Individual Highlight

Long-Term Seasonality of Greenness in Tropical Forest

Photo of Landsat image greenness metrics for the years 2010-2014 in Puerto Rico. Magenta areas have low minimum and maximum vegetation greenness and less photosynthesis overall. Darker greens are semi-deciduous dry forest patches which have greater minimum greenness. Bright yellow and bright green have high greenness all year.  Landsat image greenness metrics for the years 2010-2014 in Puerto Rico. Magenta areas have low minimum and maximum vegetation greenness and less photosynthesis overall. Darker greens are semi-deciduous dry forest patches which have greater minimum greenness. Bright yellow and bright green have high greenness all year. Snapshot : Seasonal patterns of forest greenness (termed phenology) help guage forest productivity and health. Mapping changes in these patterns with satellite imagery can locate where forests are most affected by hurricanes or drought. This research is refining algorithms for, and mapping, long-term phenology over Puerto Rico and the U.S. Virgin Islands, where such mapping is particularly difficult because cloud-free images are rare and topography is steep.

Principal Investigators(s) :
Helmer, Eileen H. 
Research Location : Puerto Rico and the U.S. Virgin Islands
Research Station : International Institute of Tropical Forestry (IITF)
Year : 2020
Highlight ID : 1636

Summary

Effective forest management requires understanding where forest productivity may be affected by climatic events like hurricanes or drought. Detecting these changes in satellite imagery, however, requires long-term seasonal image data. In persistently cloudy tropical regions like Puerto Rico and the U.S. Virgin Islands, deriving long-term trends from satellite imagery is challenging. Vegetation seasonality is complicated by steep topography and the small number of times any one place is imaged when the sky is also clear. USDA Forest Service scientists and collaborators are developing a long-term dataset of forest greenness that will track seasonal patterns of greenness, which are related to forest productivity, for the years 2000 through 2018. By refining algorithms that they previously developed and applied, they hope to detect seasonal vegetation greenness, even where there is more frequent cloud cover and steeper topography than in previous tests. They will use these data in conjunction with data from the Forest Inventory and Analysis program to characterize the changes in forest productivity related to recent drought and hurricanes.

Forest Service Partners

External Partners

  • Humfredo Marcano-Vega (Southern Research Station)
  • David Gwenzi - Humboldt State University
  • Xiaolin Zhu - Hong Kong Polytechnic University