Skip to Main Content
Statistical inference for forest structural diversity indices using airborne laser scanning data and the k-Nearest Neighbors techniqueAuthor(s): Matteo Mura; Ronald E. McRoberts; Gherardo Chirici; Marco Marchetti
Source: Remote Sensing of Environment
Publication Series: Scientific Journal (JRNL)
Station: Northern Research Station
View PDF (1.0 MB)
DescriptionForest structural diversity plays a major role for forest management, conservation and restoration and is recognized as a fundamental aspect of forest biodiversity. The assessment, maintenance and restoration of a diversified forest structure have become major foci in the effort to preserve forest ecosystems from loss of biological diversity. However, the assessment of forest biodiversity is difficult because it involves multiple components and is characterized using multiple variables. The objective of the study was to develop a methodological approach for predicting, mapping, and constructing a statistical inference for a multiple-variable index of forest structural diversity. The method included three key components: (i) use of the k-NearestNeighbors (k-NN) technique, field plot data, and airborne laser scanning metrics to predict multiple forest structural diversity variables simultaneously, (ii) incorporation of multiple diversity variable predictions into a single index, and (iii) construction of a statistically rigorous inference for the population mean of the index. Three structural diversity variables were selected to illustrate the method: growing stock volume and the standard deviations of tree diameter at breast-height and tree height. Optimization of the k-NN technique produced mean relative deviations less in absolute value than 0.04 for predictions for each of the three structural diversity variables, R2 values between 0.50 and 0.66which were in the range of values reported in the literature, and a confidence interval for the population mean of the index whose half-width was approximately 5% of the mean. Finally, the spatial pattern depicted in the resulting map of forest structural diversity for the study area contributed to validating the proposed method.
- Check the Northern Research Station web site to request a printed copy of this publication.
- Our on-line publications are scanned and captured using Adobe Acrobat.
- During the capture process some typographical errors may occur.
- Please contact Sharon Hobrla, email@example.com if you notice any errors which make this publication unusable.
- We recommend that you also print this page and attach it to the printout of the article, to retain the full citation information.
- This article was written and prepared by U.S. Government employees on official time, and is therefore in the public domain.
CitationMura, Matteo; McRoberts, Ronald E.; Chirici, Gherardo; Marchetti, Marco. 2016. Statistical inference for forest structural diversity indices using airborne laser scanning data and the k-Nearest Neighbors technique. Remote Sensing of Environment. 186: 678-686. https://doi.org/10.1016/j.rse.2016.09.010.
KeywordsBootstrap variance, Optimization, Multivariate
- A Comparison of the Spatial Linear Model to Nearest Neighbor (k-NN) Methods for Forestry Applications
- Estimating areal means and variances of forest attributes using the k-Nearest Neighbors technique and satellite imagery
- Multivariate inference for forest inventories using auxiliary airborne laser scanning data
XML: View XML