Estimation of Urban Forest Canopy Using Non-parametric Methods and GeoEye-1 Imagery Data (Comparison of BRT and RF Regression Algorithms)

The forest canopy is the medium for energy، mass، and momentum exchanges between the forest ecosystem and the atmosphere. Tree crown size is a critical aspect of canopy structure that significantly influences these biophysical processes in the canopy. Tree crown size is also strongly related to other canopy structural parameters، such as tree height، diameter at breast height and biomass. But information about tree crown sizes is difficult to obtain and rarely available from traditional forest inventory. In this study، relationship between GeoEye-1 multispectral imagery data and urban forest canopy was investigated in region 3 of Tehran. Too، the aim of this study is evaluation the sufficiency of GeoEye-1 data and image texture features and Boosted Regression Tree method (BRT) and Random Forest algorithm (RF) to delineate the urban forest canopy. At first، we confided of geometric rectification using a road network map. By using full inventory take 100 plots with 20× 20 (m) dimension. In digital canopy extraction using texture analyze factors and main band data، BRT and RF algorithms was used for analyze and evaluation relationship between canopy area and satellite data. The BRT method estimated the canopy cover by adjustment determinate coefficient and root mean square error respectively 97%، 38. 34 (m2. plot-1). The mentioned values for RF are 93% and 38. 24 (m2. plot-1) respectively. This study presents that the strengths of the GeoEye-1 imagery data and the potentials of the image texture features and BRT and RF methods، which may help the urban planners، monitor and interpret complex urban characteristics