Urban tree damage estimation using airborne laser scanner data and geographic information systems: An example from 2007 Oklahoma ice storm

Abstract

This study has applied an integrated airborne laser scanning (ALS) and geographic information systems (GIS) technique to estimate urban tree damages caused by an ice storm that affected the City of Norman, Oklahoma between 8 and 11th of December, 2007. Pre- and post-storm ALS data were collected and processed using GIS. The pre-storm ALS data was used to create a digital elevation model (DEM) and the pre-storm digital surface model (DSM) while the post-storm DSM was created from post-storm ALS data. The DEM were subtracted from the two DSMs to create the pre- and post-storm canopy height models (CHMs). Individual and grouped trees were then separated using a combination of Koukoulas and Blackburn and local maxima tree extraction algorithms. Finally, differences in height and canopy diameter between the pre- and post-storm CHMs within the extracted trees were calculated to measure the damage. Results showed that approximately 9% of the urban trees were completely damaged while almost 6% did not sustain any damage. It also showed that tree damage from ice storms varied with species and that Acer saccarinum, Acer saccharum Marsh, Ulmus americana, and Ulmus pumila were most vulnerable while Platanus occidentalis and Quercus palustris were the most resilient tree species to the ice storm. When compared with the ground reference data, it was found that the proposed methodology can detect tree height/stem damage with >81% accuracy compared to >42% accuracy in canopy damage assessment. Once ALS data is collected after a disaster, the methodology can provide a quick estimate of urban tree damage and hence be useful for city planners. It can also be used to create detailed inventories of tree heights of urban forests and monitor their growth and changes to improve their long-term sustainability.