Mapping forest structure for wildlife habitat analysis using waveform lidar: Validation of montane ecosystems

摘要

Consistent and accurate measurements of forest structure at the landscape scale are required by forest ecologists and managers for a variety of applications. Lidar remote sensing has proven to be a valuable tool for measuring these attributes in many ecosystems, including tropical, boreal, and mid-latitude forests. However, there have been few studies in montane forests. Here, we examine the ability of a large footprint lidar system to retrieve forest structural attributes in the highly variable terrain and canopy conditions of the Sierra Nevada mountains in California. Specifically, we examined the impact of slope, elevation, aspect, canopy cover, crown shape, and the spatial arrangement of canopy-forming trees on the accuracy of a large footprint lidar system in retrieving canopy height, canopy cover, and biomass. We found good agreement between field and lidar measurements of height, cover, and biomass at the footprint level, and canopy height and biomass at the stand level. Differences between field and lidar measurements are mainly attributable to the spatial configuration of canopy elements and are less sensitive to topography, crown shape, or canopy cover. The accuracy of canopy cover retrieval was highly sensitive to estimates of ground cover reflectivity and to ground sampling density. The accuracy of biomass retrieval was also good, and comparable to previous efforts in other biomes. (c) 2005 Elsevier Inc. All lights reserved.