Look for complete geospatial metadata in this layer's associated xml document available from the download link * Metric Name: Northern Spotted Owl Nesting/Roosting Forest Cover Type Suitability Index * Tier: 2 * Data Vintage: 2023 * Unit Of Measure: 0−10000 = relative suitability index from 0 to 1 (with a x10000 scalar applied). A value of 10,000 would indicate perfect habitat suitability of a pixel. Maximum value found (within the California portion of the range of NSO) is 8,628. * Metric Definition and Relevance: This is a measure, expressed as a relative index (0-1), of suitability (or similarity) of forest structure and composition for Northern Spotted Owl (NSO) ( _Strix occidentalis caurina_ ) nesting/roosting. The data were developed by the Regional Ecosystem Office for the Northwest Forest Plan monitoring program. These data were produced using machine learning software Maxent and trained/tested with nesting/roosting pair locations (1993). As illustrated in the diagram below, an index near zero indicates forest structure/composition dissimilar to where NSO pairs nest/roost. An index nearer to one indicates similar conditions. Raster values are as follows: • -1 = no data (non-forested) • 0−10000 = relative suitability index from 0 to 1 (with a x10000 scalar applied). A value of 10,000 would indicate perfect habitat suitability of a pixel. Maximum value found (within the California portion of the range of NSO) is 8,628. Annual forest vegetation structure and composition maps (30-m pixel resolution) for forest-capable lands from 1986 to 2023 were generated using the gradient nearest neighbor (GNN) imputation modeling and mapping methodology developed by Oregon State University Department of Forest Ecosystems and Society’s Landscape Ecology, Modeling, Mapping, and Analysis program (LEMMA 2020). GNN is a multivariate, nonparametric modeling and mapping framework that inputs forest inventory plot data to individual map pixels based on Landsat surface reflectance and environmental similarity in the gradient space (Ohmann and Gregory 2002). The version of GNN used in this analysis was based on the composite Landsat images produced to map the forest disturbances above, matching plot measurements to Landsat image years (Bell et al. 2021). Methodological changes, described in detail in the late-successional and old- growth monitoring report (), improved the quality of GNN compared to previous monitoring reports. This included using a consistent type of forest inventory plot for imputations, the ensemble LandTrendr imagery described above, imagery stabilization, and bootstrapped approximations utilizing multiple neighbors (k = 7) with weighted means proportional to the probability that a bootstrap sample would result in that plot being the nearest neighbor for a pixel. * Creation Method: Maps of forest types associated with owl nesting and roosting were produced following methods from previous monitoring reports (Davis et al. 2011, 2016), and that methodology is described briefly below. Open-source machine learning software Maxent (Phillips et al. 2006, 2017, 2021) was used to develop a forest cover type model for each modeling region using 10 bootstrapped random samples. We used 75 percent of NSO locations for model training and 25 percent for model testing. Training locations were analyzed against a random sampling of 10,000 background locations from forest-capable pixels within the modeling region. The authors used a logistic model output as the relative index of forest suitability for nesting and roosting by NSO pairs. The forest suitability index ranged from 0 to 1.0, where values closer to zero represent forest structure and species composition unsimilar to that found at NSO locations and higher values are more similar. See Davis et al. 2022 for more details on the methods. * Credits: Regional Ecosystem Office (REO) - Northwest Forest Plan Bell, D.M.; Acker, S.A.; Gregory, M.J.; Davis, R.J.; Garcia, B.A. 2021. Quantifying regional trends in large live tree and snag availability in support of forest management. Forest Ecology and Management. 479: Article 118554. . 2020.118554. Glenn, E.M., Lesmeister, D.B., Davis, R.J., Hollen, B., Poopatanpong, A. 2017. Estimating density of a territorial species in a dynamic landscape. Landscape Ecol. 32:563–579. Davis, R.J.; Dugger, K.M.; Mohoric, S.; Evers, L.; Aney, W.C. 2011. Northwest Forest Plan—the first 15 years (1994–2008): status and trends of northern spotted owl populations and habitats. Gen. Tech. Rep. PNW-GTR-850. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station. 147 p. https://doi.org/10.2737/PNW-GTR-850. Davis, R.J.; Hollen, B.; Hobson, J.; Gower, J.E.; Keenum, D. 2016. Northwest Forest Plan—the first 20 years (1994–2013): status and trends of northern spotted owl habitats. Gen. Tech. Rep. PNW-GTR-929. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station. 54 p. https://doi.org/10.2737/PNW-GTR-929. Davis, Raymond J.; Lesmeister, Damon B.; Yang, Zhiqiang; Hollen, Bruce; Tuerler, Bridgette; Hobson, Jeremy; Guetterman, John; Stratton, Andrew. 2022. Northwest Forest Plan—the first 25 years (1994–2018): status and trends of northern spotted owl habitats. Gen. Tech. Rep. PNW-GTR-1003. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station. 38 p. . Phillips, S.J.; Anderson, R.P.; Shapire, R.E. 2006. Maximum entropy modeling of species geographic distributions. Ecological Modelling. 190(3–4): 231–259. . Phillips, S.J.; Anderson, R.P.; Dudik, M.; Shapire, R.E.; Blair, M.E. 2017. Opening the black box: an open-source release of Maxent. Ecography. 40(7): 887–893. . Phillips, S.J.; Dudík, M.; Shapire, R.E. 2021. Maxent software for modeling species niches and distributions (Version 3.4.1). http://biodiversityinformatics.amnh.org/open_source/maxent. (31 January 2021).