A nearest-neighbor imputation approach to mapping tree species over large areas using forest inventory plots and moderate resolution raster data

Wilson, Barry T.; Lister, Andrew J.; Riemann, Rachel

doi:10.1016/j.foreco.2012.02.002

Cited by 156 publications

(164 citation statements)

References 52 publications

(52 reference statements)

Supporting

Mentioning

158

Contrasting

Order By: Relevance

“…For less common species there was no evidence that one strategy worked better than the other. Previous studies have differed in their conclusions with respect to the effects of stratification on k-NN predictions (e.g., [39,40]), although the studies did not look at dbh predictions. Differing sample sizes between studies likely played a role in the different findings between studies.…”

Section: K-nn Imputation Strategiesmentioning

confidence: 98%

An Examination of Diameter Density Prediction with k-NN and Airborne Lidar

Strunk

Gould

Packalén

et al. 2017

Forests

View full text Add to dashboard Cite

While lidar-based forest inventory methods have been widely demonstrated, performances of methods to predict tree diameters with airborne lidar (lidar) are not well understood. One cause for this is that the performance metrics typically used in studies for prediction of diameters can be difficult to interpret, and may not support comparative inferences between sampling designs and study areas. To help with this problem we propose two indices and use them to evaluate a variety of lidar and k nearest neighbor (k-NN) strategies for prediction of tree diameter distributions. The indices are based on the coefficient of determination (R 2 ), and root mean square deviation (RMSD). Both of the indices are highly interpretable, and the RMSD-based index facilitates comparisons with alternative (non-lidar) inventory strategies, and with projects in other regions. K-NN diameter distribution prediction strategies were examined using auxiliary lidar for 190 training plots distribute across the 800 km 2 Savannah River Site in South Carolina, USA. We evaluate the performance of k-NN with respect to distance metrics, number of neighbors, predictor sets, and response sets. K-NN and lidar explained 80% of variability in diameters, and Mahalanobis distance with k = 3 neighbors performed best according to a number of criteria.

show abstract

Section: K-nn Imputation Strategiesmentioning

confidence: 98%

An Examination of Diameter Density Prediction with k-NN and Airborne Lidar

Strunk

Gould

Packalén

et al. 2017

Forests

View full text Add to dashboard Cite

show abstract

“…The SII integrates three facets of fragmentation affecting some aspect of forest ecosystem functioning-patch size, local forest density, and patch connectivity to core forest areas-to create a single metric for comparison. Since even acceptably low misclassification rates in the source land cover data can be magnified into substantial errors in metric values (Langford et al 2006, Shao and Wu 208), we have calculated spatial integrity at the two scales corresponding to two of the most reliable and widely available sources of data-the 30 m (98.4 ft) scale of the 2011 National Land Cover Dataset (NLCD 2011) (Jin et al 2013), and the 250 m (820 ft) scale of the 2009 FIA forest cover dataset (Wilson et al 2012). Both scales fall within the 10 to 1000 km 2 (2,471 to 247,091 acres) scale at which pattern process linkages are often of greatest management interest (Forman and Godron 1986).…”

Section: Introductionmentioning

confidence: 99%

New York Forests, 2012

Widmann¹,

Crawford²,

Kurtz³

et al. 2015

View full text Add to dashboard Cite

This report summarizes the second annual inventory of New York's forests, conducted in 2008-2012. New York's forests cover 19.0 million acres; 15.9 million acres are classified as timberland and 3.1 million acres as reserved and other forest land. Forest land is dominated by the maple/beech/ birch forest-type group that occupies more than half of the forest land. The sound wood volume on timberland has been rising and is currently 37.4 billion cubic feet, enough to produce saw logs equivalent to 93.7 billion board feet. On timberland, the average annual growth in volume of live trees outpaced removals by a ratio of 2.1:1. The net change in volume averaged 1.1 percent per year. This report includes additional information on forest attributes, land use, forest fragmentation, forest ownership, forest health indicators, timber products, statistics, and quality assurance of data collection. Detailed information on forest inventory methods and data quality are available online at http://dx

show abstract

“…A geographic information system (GIS) and various geospatial datasets were used to produce the maps in this report. Maps were constructed using (1) categorical coloring of Illinois counties according to forest attributes (such as forest land area), (2) a variation of the k-nearest-neighbor (KNN) technique to apply information from forest inventory plots to remotely sensed MODIS imagery (250 m pixel size) based on the spectral characterization of pixels and additional geospatial information (see Wilson et al 2012 …”

Section: How Do We Produce Maps?mentioning

confidence: 99%

Illinois Forests 2015

Crocker¹,

Butler²,

Kurtz³

et al. 2017

Self Cite

View full text Add to dashboard Cite

The third full annual inventory of Illinois' forests reports more than 4.9 million acres of forest land and 99 tree species. Forest land is dominated by oak/hickory and elm/ash/cottonwood forest types, which make up 92 percent of total forest area. The volume of growing stock on timberland has been rising since 1948 and currently totals 7.0 billion cubic feet. Average annual net growth of growing stock from 2010 to 2015 was about 146.1 million cubic feet per year. This report includes additional information on forest attributes, land-use change, carbon, timber products, and forest health. The following information is available online at https://doi.org/10.2737/NRS-RB-113: 1) detailed information on forest inventory methods, statistics, and quality assurance of data collection; 2) tables that summarize quality assurance; 3) a core set of tabular estimates for a variety of forest resources; and 4) a Microsoft® Access database that represents an archive of data used in this report, with tools that allow users to produce customized estimates. AcknowledgmentsThe authors would like to thank the many individuals who contributed to both the inventory and the analysis of Illinois' forest resources. Primary field crew and quality assurance staff over the [2010][2011][2012][2013][2014][2015]

show abstract

A nearest-neighbor imputation approach to mapping tree species over large areas using forest inventory plots and moderate resolution raster data

Cited by 156 publications

References 52 publications

An Examination of Diameter Density Prediction with k-NN and Airborne Lidar

An Examination of Diameter Density Prediction with k-NN and Airborne Lidar

New York Forests, 2012

Illinois Forests 2015

Contact Info

Product

Resources

About