Multi-temporal LiDAR and hyperspectral data fusion for classification of semi-arid woody cover species

Authors: NORTON, C.L. - University Of Arizona; HARTFIELD, K. - University Of Arizona; Holifield Collins, Chandra; VAN LEEUWEN, W.J.D. - University Of Arizona; METZ, L. - Natural Resources Conservation Service (NRCS, USDA)

Submitted to: Remote Sensing
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/16/2022
Publication Date: 6/17/2022
Citation: Norton, C., Hartfield, K., Holifield Collins, C.D., van Leeuwen, W., Metz, L. 2022. Multi-temporal LiDAR and hyperspectral data fusion for classification of semi-arid woody cover species. Remote Sensing. 14(12). Article 2896. https://doi.org/10.3390/rs14122896.
DOI: https://doi.org/10.3390/rs14122896

Interpretive Summary: Mapping the distribution of woody plants is important for monitoring, managing, and studying woody invasion in grasslands. However, in dry regions, discrimination of tree species is difficult primarily due to the tree similarities, small and sparse canopy cover, but may also be due to overlapping woody canopies as well as seasonal leaf retention (deciduous versus evergreen) characteristics. The objective of this study was to investigate the use of airborne remote sensing data for tree species classification in a desert region. This study produces highly accurate maps by combining multi-temporal fine spatial resolution hyperspectral imagery and Light Detection and Ranging (LiDAR) data (~1 m) from the National Ecological Observatory Network (NEON) at the Santa Rita Experimental Range (SRER) site through a reproducible statistical programing approach that can be applied to larger areas and similar datasets. The influence of fusing spectral and structural information in a random forest classifier for tree identification is clear. Additionally, a multi-temporal dataset slightly increases classification accuracies over a single data collection. Our results show a promising way for tree species classification in a desert region using remote sensing data.

Technical Abstract: Mapping the spatial distribution of woody vegetation is important for monitoring, managing, and studying woody encroachment in grasslands. However, in semi-arid regions, remotely sensed discrimination of tree species is difficult primarily due to the tree similarities, small and sparse canopy cover, but may also be due to overlapping woody canopies as well as seasonal leaf retention (deciduous versus evergreen) characteristics. Similar studies in different biomes have achieved low accuracies using coarse spatial resolution image data. The objective of this study was to investigate the use of multi-temporal, airborne hyperspectral imagery and light detection and ranging (LiDAR) derived data for tree species classification in a semi-arid desert region. This study produces highly accurate classifications by combining multi-temporal fine spatial resolution hyperspectral and LiDAR data (~1 m) through a reproducible scripting and machine learning approach that can be applied to larger areas and similar datasets. Combining multi-temporal vegetation indices and canopy height models led to an overall accuracy of 95.28% and kappa of 94.17%. Five woody species were discriminated resulting in producer accuracies ranging from 86.12% to 98.38%. The influence of fusing spectral and structural information in a random forest classifier for tree identification is evident. Additionally, a multi-temporal dataset slightly increases classification accuracies over a single data collection. Our results show a promising methodology for tree species classification in a semi-arid region using multi-temporal hyperspectral and LiDAR remote sensing data.