UCLA

Tree Census data is used to quantify forest ecosystem health and services. Current urban forest

methodologies lack a comprehensive approach to integrating 2D and 3D in-situ tree census data

with 2D and 3D products. This research proposes the Kabachnik Tree Census Model (KTCM),

which adds new tree census collection parameters, and a new methodology for processing open

source imagery. The KTCM can work with existing data collection platforms like i-Tree. The

accuracy of TCC products, a USDA Forest Service (USDAFS) Quickbird derived TCC product,

and an open source Google Earth screen capture TCC product, and the TCC KTCM ground truth

are analyzed using an error matrix. A regression analysis is used to understand correlations between DBH, height, and volumetric estimates from the KTCM. The results show that the KTCM is a valuable ground truth model and that vegetation shadow, complex multi-story canopy, multi-colored canopy, seasonality, tree growth, and temporal resolutions, pose the

greatest errors in correctly assessing tree canopy cover and biomass estimates with high

resolution sub-meter imagery.