Skip to Main Content
U.S. Forest Service
Caring for the land and serving people

United States Department of Agriculture

Home > Search > Publication Information

 Due to a lapse in federal funding, this USDA website will not be actively updated. Once funding has been reestablished, online operations will continue.


  1. Share via EmailShare on FacebookShare on LinkedInShare on Twitter
    Dislike this pubLike this pub
    Author(s): Carolyn B. Meyer; Sherri L. MillerC. John Ralph
    Date: 2004
    Source: In: S. Huzurbazar, editor. Resource Selection Methods and Applications. Omnipress, Madison, Wisconsin: p. 94-106
    Publication Series: Miscellaneous Publication
    PDF: View PDF  (503 KB)

    Description

    The scale at which habitat variables are measured affects the accuracy of resource selection functions in predicting animal use of sites. We used logistic regression models for a wide-ranging species, the marbled murrelet, (Brachyramphus marmoratus) in a large region in California to address how much changing the spatial or temporal scale of variables changes accuracy in predicting sites occupied by murrelets (sites believed to be used for nesting). This seabird forages in the ocean and nests inland in large, old trees. Classification accuracy of independent plots, assessed at 4 spatial scales (patch, landscape, subregional, and regional) and 2 time periods (present and previous decade), was highest (by 10%) for the model that incorporated all scales. Of the individual spatial scales, landscape was most accurate probably because it contained the most limiting factors for the murrelet, which were old-growth forest fragmentation and isolation. For temporal scale, there was a time lag before birds showed a negative response to fragmentation, as they still occupied plots in the 1990s that were recently (after 1985) fragmented. Adding the time lag improved accuracy by 4%. When absence data from plot locations beyond the apparent geographic nesting range (delineated by presence of frequent fog) were removed from models, prediction accuracy improved within the nesting range, mostly due to improved optimal classification cutoffs. We more rigorously evaluated our multi-scale model by assessing accuracy within geographic subsections of the nesting range and found it was still high to very high (86-100), as most studies rarely exceed 85% accuracy. The results confirm that logistic regression can be very useful for predicting animal use when variables are measured at multiple spatial and temporal scales.

    Publication Notes

    • You may send email to psw_communications@fs.fed.us to request a hard copy of this publication.
    • (Please specify exactly which publication you are requesting and your mailing address.)
    • 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.

    Citation

    Meyer, Carolyn B.; Miller, Sherri L.; Ralph, C. John. 2004. Logistic regression accuracy across different spatial and temporal scales for a wide-ranging species, the marbled murrelet. In: S. Huzurbazar, editor. Resource Selection Methods and Applications. Omnipress, Madison, Wisconsin: p. 94-106

    Keywords

    Brachyramphus marmoratus, California, habitat, logistic regression, marbled murrelet, prediction accuracy, spatial scale, temporal scale

    Related Search


    XML: View XML
Show More
Show Fewer
Jump to Top of Page
https://www.fs.usda.gov/treesearch/pubs/31939