| dc.description.abstract | The benefits of an urban forest canopy have become more readily understood in recent years. They provide a wealth of environmental, social, and economic benefits such as improved water quality, air temperature regulation, carbon sequestration, air pollution reduction, enhanced human health, better aesthetics, increased property value, and reduction of energy consumption. For this study, a multi-scale analysis is used to determine the efficiency and efficacy of varying remote sensing platforms to capture, model, and measure standard forest metrics. These include industry-standard measurements such as diameter at breast height (DBH) and total stem height. Remotely sensed data platforms such as airborne light detection and ranging (LiDAR), ground-based LiDAR, stereoscopic imaging, and image object construction using object based image analysis have existed for some time, and provide ways to reduce cost and increase efficiency over standard field data collection methods. For this study, data of the same urban forest canopy was examined using data collected from each remote sensing platform and compared directly with data captured using standard in-field forest measurement techniques. The specific objectives being to determine if remote sensing platforms which provide the highest resolutions, specifically spatial resolution, correlate into higher relative accuracy when compared to in-field measurements. These measurements include: 1) at the individual tree scale, DBH, total height, crown width, and total biomass taken for each individual branch along with the bole 2) at the city block scale, these measurments include DBH, total height, species identification, and crown width. 3) at the city scale, these measurments include total height and species identification. The study areas include: 1) a series of open-grown Nuttall oaks (Quercus texana) located approximately 5km northwest of Auburn University, Alabama 2) A single city block located on the campus of Auburn Unversity 3) an urban forested area of the City of Auburn approximately 310 hectares in size. At the local scale each of a series of 7 Nuttall oak trees were scanned using a terrestrial LiDAR scanning system (TLS) from a distance of 5 to 10 meters and from 7 different vantages. Once these data were captured and registered into a comprehensive vector point cloud the trees were destructively sampled in order to aquire accurate linear and volumetric measurements. This information was then compared with those generated from models of the vector point cloud. No significant differences in linear measurements were observed, but a consistent underestimation of overall biomass was observed across all 7 specimen. At the medium scale, a city block of Auburn University’s campus had aiborne imagery taken of the entire area from multiple vantage points using oblique oriented passive sensor cameras. These data were used to create a vector point cloud of the study area that were compared to field measurments. No significant differences in linear measurement were observed, minor underestimation of overall biomass was observed compared with estimates based on field measurements. At the city scale, airborne LiDAR (ALS) data were collected and used in conjunction with airborne orthographic imagery to generate a surface model that identifies individual trees. This was accomplished using geographic object based image analysis (GeOBIA) techniques. Once generated the total heights for each individual tree were attributed and compared to a sampling of trees measured in the field. Total heights were found to be underestimated in most observations, but were greater for shorter coniferous trees when compared to taller deciduous and coniferous trees. In general, this study was useful in comparing the accuracy and relative efficiency of these different remote sensing platforms as they pertain to urban forest measurements. Modern urban forestry relies heavily upon the use of these and other forms of remote sensing to quickly and accurately measure and maintain these vital natural resources. | en_US |
