You are here

Harvard Forest >

Harvard Forest REU Symposium Abstract 2014

  • Title: Measuring Changes in Phenology with Landsat
  • Author: Joshua A Alaniz (University of North Texas)
  • Abstract:

    Human disturbance of forests is increasingly prolific across the U.S. In studying how an ecosystem regrows after human disturbance, changes in phenology can act as a proxy for other ecosystem characteristics, like leaf area or productivity. This study investigates changes in phenology at a 2008 clear-cut at Harvard Forest by determining the date of peak and fastest increase in vegetation for the growing season, estimated by the Normalized Difference Vegetation Index (NDVI). Greater NDVI values indicate greater vegetative cover. By comparing seasonal NDVI curves, the timing of maximum vegetative cover and fastest growing period within the year can be compared over time. This study attempts to test if NDVI derived from Landsat imagery can be used to characterize changes in phenology, given limitations in spatial, spectral, and temporal resolution of Landsat. Images were gathered from USGS’ Earth Explorer from March through November of 2009, 2011, and 2013. Cloud free imagery from Landsat 5, 7, and 8 were used, and each image was atmospherically corrected with the COST model. NDVI and Tasseled Cap (Kauth-Thomas) transformations were calculated with ESRI’s ArcGIS raster calculator. Given the regrowth of vegetation after clear-cut, we predicted that NDVI will start from a higher value each year than the previous and that the peak of NDVI will occur earlier and reach a greater value than the year before. This is inferred by vegetation having more structure and resources available each year increasing the rate and magnitude of leaf area.

  • Research Category: Forest-Atmosphere Exchange