Forest Science Program

See below to download a PDF file.

Validation of modelled habitat classifications for the northern spotted owl in British Columbia using patterns of historical occupancy

Author(s) or contact(s): G.D. Sutherland, J.R. Smith, D.T. O’Brien, F.L. Waterhouse, and A.S. Harestad
Source: Forest Science Program
Subject: Wildlife
Series: Technical Report
Other details:  Published 2010. Hardcopy is available.


Large-scale, integrative decision support tools and models are often used to represent resource management problems and guide management policy. However, such tools and models can be difficult to validate and even more difficult to verify. This creates uncertainties in how best to use model results to inform development of management policy. We used a historical data set of detections of the endangered Northern Spotted Owl (Strix occidentalis caurina) in British Columbia to examine the degree of correspondence between projections of quantity of suitable habitat and its configuration as generated by a spatially explicit modelling framework, and broad classes of area occupancy by owls observed in detection surveys between 1992 and 2001. We found that among spatial scales of analysis, as well as between two time periods (2004 landscape and reconstructed 1984 habitat projections), the projected patterns of habitat suitability and connectivity were as predicted assuming that occupancy of areas by owls is positively related to these factors. However, we could not consistently demonstrate statistical significance for these relationships at all the spatial scales and the two time periods examined.

While our results do not offer direct ecological support for the effect of habitat quantity and configuration on observed trends in Northern Spotted Owl populations in British Columbia, they increase confidence in the parameters and structural assumptions in the habitat portion of the model. We caution that because of data limitations, our results should be treated as an exploration of the implications of important assumptions rather than as strict tests of the influence of habitat variables on population trends. The amount and distribution of habitat is clearly an important issue for conservation and management of Northern Spotted Owls in British Columbia. However, until we better understand the way in which Barred Owls influence Northern Spotted Owls, insights from analyses such as those used in this study may be constrained. Currently, the relative influences of habitat, Barred Owl effects, and other stochastic effects of small population size on detect-ability and distribution patterns of Northern Spotted Owls cannot be fully separated in the type of analysis used in this study.

Download Technical Report 056 PDF file (1452 KB)

To view this document you need the current version of
Adobe Acrobat Reader, available free from the Adobe Web Site.

Updated March 17, 2010