Baltimore Ecosystem Study Institute of Ecosystem Studies

2010 BES Annual Meeting Presentation and Poster Abstracts

Correlating spatial heterogeneity in diurnal eddy fluxes with remote sensing and land cover during the growing and non-growing seasons in suburban Baltimore
Saliendra, Nicanor
Co-Authors: John Hom, Rich Pouyat, Matt Patterson,Ian Yesilonis, Gordon Heisler and Dave Nowak

Abstract: The Cub Hill tower in suburban Baltimore has a heterogeneous footprint that does not conform to traditional sites for micrometeorological measurements. We tested the hypothesis that footprint-specific land cover and remote sensing characteristics will be correlated to spatial heterogeneities in eddy fluxes. Thus, our main objective was to identify and develop empirical models between eddy fluxes, remote sensing metrics and Land Use-Land Cover (LULC). We stratified the eddy flux, remote sensing and LULC data into 24 wind sectors as shown in Fig. 1a. Footprint in each wind sector was characterized by extracting the LULC data from a previous mapping project, and calculating remote sensing metrics from the Landsat-TM (Thematic Mapper). Daily means of diurnal eddy fluxes were averaged every 16-day period to match the 23 Landsat-TM repeat cycles per year. Lastly, eddy flux and remote sensing data were stratified into growing and non-growing seasons, and averaged according to the 24 wind sectors. We found that Normalized Difference Vegetation Index (NDVI) was tightly coupled to LULC (%Vegetation) during the growing season (Fig. 1b), while Normalized Difference Water Index (NDWI) and Land Surface Temperature (LST) were correlated to %Vegetation during both seasons. Additionally, both seasonal means of CO2 and sensible heat fluxes were linked to %Vegetation and %Tree, respectively. Results suggest that the empirical models developed here could be used to facilitate the scaling up of eddy fluxes from tower to landscape and perhaps to regional level.