A k-nearest neighbor approach for estimation of single-tree biomass

Lutz Fehrmann; Christoph Kleinn

Treesearch

A k-nearest neighbor approach for estimation of single-tree biomass

Informally Refereed

Authors:

Lutz Fehrmann, Christoph Kleinn

Year:

2007

Type:

Other

Station:

Northern Research Station

Source:

In: McRoberts, Ronald E.; Reams, Gregory A.; Van Deusen, Paul C.; McWilliams, William H., eds. Proceedings of the seventh annual forest inventory and analysis symposium; October 3-6, 2005; Portland, ME. Gen. Tech. Rep. WO-77. Washington, DC: U.S. Department of Agriculture, Forest Service: 251-259.

Download (PDF 1.43 MB)

Download (PDF 1.43 MB): https://research.fs.usda.gov/download/treesearch/17051.pdf

Abstract

Allometric biomass models are typically site and species specific. They are mostly based on a low number of independent variables such as diameter at breast height and tree height. Because of relatively small datasets, their validity is limited to the set of conditions of the study, such as site conditions and diameter range. One challenge in the context of the current climate change discussion is to develop more general approaches for reliable biomass estimation. One alternative approach to widely used regression modelling are nonparametric techniques.

Parent Publication

Proceedings of the seventh annual forest inventory and analysis symposium

Citation

Fehrmann, Lutz; Kleinn, Christoph. 2007. A k-nearest neighbor approach for estimation of single-tree biomass. In: McRoberts, Ronald E.; Reams, Gregory A.; Van Deusen, Paul C.; McWilliams, William H., eds. Proceedings of the seventh annual forest inventory and analysis symposium; October 3-6, 2005; Portland, ME. Gen. Tech. Rep. WO-77. Washington, DC: U.S. Department of Agriculture, Forest Service: 251-259.