Evaluating the Uncertainties in Forest Canopy Height Measurements Using ICESat-2 Data

Forest ecosystems have been identified as major carbon stocks in terrestrial ecosystems; therefore, their monitoring is critical. Forests cover large areas, making it difficult to monitor and maintain up-to-date information. Advances in remote sensing technologies provide opportunities for detailed small-scale monitoring to global monitoring of forest resources. Airborne laser scanning (ALS) data can provide precise forest structure measurements, but mainly for small-scale forest monitoring due to its expensive cost and limited spatial and temporal coverage. Spaceborne lidar (light detection and ranging) can cover extensive spatial scales, but its suitability as a replacement for ALS measurements remains uncertain. There are still relatively few studies on the performance of spaceborne lidar to estimate forest attributes with sufficient accuracy and precision. Therefore, this study aimed at assessing the performance of spaceborne lidar ICESat-2 canopy height metrics and understanding their uncertainties and utilities by evaluating their agreements with ALS-derived canopy height metrics in Mississippi, United States. We assessed their agreements for different forest types, physiographic regions, a range of canopy cover, and diverse disturbance histories using equivalence tests. Results suggest that ICESat-2 canopy height metrics collected using strong beam mode at night have higher agreement with ALS-derived ones. ICESat-2 showed great potential for estimating canopy heights in evergreen forests with high canopy cover. This study contributes to the scientific community’s understanding of the capabilities and limitations of ICESat-2 to measure canopy heights at regional to global scales.