Uncertainty of Large-Area Estimates of Indicators of Forest Structural Gamma Diversity: A Study Based on National Forest Inventory Data. Winter, S., Böck, A., & McRoberts, R. E. Forest Science, 58(3):284-293, 2012.
Uncertainty of Large-Area Estimates of Indicators of Forest Structural Gamma Diversity: A Study Based on National Forest Inventory Data [link]Paper  doi  abstract   bibtex   
Tree diameter and height are commonly measured forest structural variables, and indicators based on them are candidates for assessing forest diversity. We conducted our study on the uncertainty of estimates for mostly large geographic scales for four indicators of forest structural gamma diversity: mean tree diameter, mean tree height, and standard deviations of tree diameter and tree height. We had three objectives: to estimate the number of national forest inventory (NFI) plots whose data must be aggregated to obtain precise estimates of the indicators of gamma forest structural diversity; to assess differences in structural associations with respect to different forest categories; and to assess the effects of geographic distance on estimates of indicators of forest structural diversity. Uncertainty assessments were conducted using a bootstrap resampling approach and a geospatial approach to assess the effects of geographic distances. Our study relied on a database populated with NFI plot data (12,536 plots and 256,513 trees) from 13 European countries and three ecoprovinces of the United States. The results were threefold. First, data from approximately 150 NFI plots randomly selected from a large database are generally sufficient to estimate standard deviations (SDs) of diameter and height as indicators of structural forest diversity at large scales with acceptable precision. Second, the uncertainties of the estimates were generally greater for forest categories in Europe than in the United States with the exception of Douglas fir forests in the Pacific Northwest of the United States. Third, diameter and height gamma diversities expressed by the means and SDs of distributions of tree diameter and tree height were more similar for geographic areas separated by smaller distances.
@article{RN363,
   author = {Winter, Susanne and Böck, Andreas and McRoberts, Ronald E.},
   title = {Uncertainty of Large-Area Estimates of Indicators of Forest Structural Gamma Diversity: A Study Based on National Forest Inventory Data},
   journal = {Forest Science},
   volume = {58},
   number = {3},
   pages = {284-293},
   abstract = {Tree diameter and height are commonly measured forest structural variables, and indicators based on them are candidates for assessing forest diversity. We conducted our study on the uncertainty of estimates for mostly large geographic scales for four indicators of forest structural gamma diversity: mean tree diameter, mean tree height, and standard deviations of tree diameter and tree height. We had three objectives: to estimate the number of national forest inventory (NFI) plots whose data must be aggregated to obtain precise estimates of the indicators of gamma forest structural diversity; to assess differences in structural associations with respect to different forest categories; and to assess the effects of geographic distance on estimates of indicators of forest structural diversity. Uncertainty assessments were conducted using a bootstrap resampling approach and a geospatial approach to assess the effects of geographic distances. Our study relied on a database populated with NFI plot data (12,536 plots and 256,513 trees) from 13 European countries and three ecoprovinces of the United States. The results were threefold. First, data from approximately 150 NFI plots randomly selected from a large database are generally sufficient to estimate standard deviations (SDs) of diameter and height as indicators of structural forest diversity at large scales with acceptable precision. Second, the uncertainties of the estimates were generally greater for forest categories in Europe than in the United States with the exception of Douglas fir forests in the Pacific Northwest of the United States. Third, diameter and height gamma diversities expressed by the means and SDs of distributions of tree diameter and tree height were more similar for geographic areas separated by smaller distances.},
   keywords = {international reporting on biodiversity
national forest inventory
spatial autocorrelation
tree diameter and height},
   ISSN = {0015-749X},
   DOI = {10.5849/forsci.10-076},
   url = {https://doi.org/10.5849/forsci.10-076},
   year = {2012},
   type = {Journal Article}
}

Downloads: 0