ESTIMATING URBAN FOREST CARBON SEQUESTRATION POTENTIAL IN THE SOUTHERN UNITED STATES USING CURRENT REMOTE SENSING IMAGERY SOURCES

Krista MERRY, Pete BETTINGER, Jacek SIRY, J. Michael BOWKER

ABSTRACT: With an increased interest in reducing carbon dioxide in the atmosphere, tree planting and maintenance in urban areas has become a viable option for increasing carbon sequestration. Methods for assessing the potential for planting trees within an urban area should allow for quick, inexpensive, and accurate estimations of available land using current remote sensing sources. Here we use Landsat 8, launched in February 2013, and the USDA’s NAIP program to perform supervised classification of land cover classes in six southern cities. These supervised classifications were used to determine the availability of plantable open area in each city. The results of the assessment using the two different imagery sources are compared, and in terms of overall accuracy, were found to be similar for the two data sources. Both the producer’s and user’s accuracies when using NAIP imagery were slightly lower than when using Landsat 8 imagery. However, each of the classifications met our desired accuracy levels for open area delineation in five of six cases.

Keywords: Urban forestry, Landsat 8, NAIP, Supervised classification, Carbon sequestration

Full article here