Modeling Marbled Murrelet (Brachyramphus marmoratus) Habitat Using LiDAR-Derived Canopy Data

To the best of our knowledge, one or more authors of this paper were federal employees when contributing to this work. This is the publisher’s final pdf. The article was published by John Wiley & Sons Ltd. and is in the public domain. The published article can be found at: http://onlinelibrary.wiley.com/journal/10.1002/%28ISSN%291938-5463a.LiDAR (Light Detection And Ranging) is an emerging remote-sensing tool that can provide\ud fine-scale data describing vertical complexity of vegetation relevant to species that are responsive to forest\ud structure. We used LiDAR data to estimate occupancy probability for the federally threatened marbled\ud murrelet (Brachyramphus marmoratus) in the Oregon Coast Range of the United States. Our goal was to\ud address the need identified in the Recovery Plan for a more accurate estimate of the availability of nesting\ud habitat by developing occupancy maps based on refined measures of nest-strand structure. We used murrelet\ud occupancy data collected by the Bureau of Land Management Coos Bay District, and canopy metrics\ud calculated from discrete return airborne LiDAR data, to fit a logistic regression model predicting the\ud probability of occupancy. Our final model for stand-level occupancy included distance to coast, and 5\ud LiDAR-derived variables describing canopy structure. With an area under the curve value (AUC) of 0.74,\ud this model had acceptable discrimination and fair agreement (Cohen’s k=0.24), especially considering that\ud all sites in our sample were regarded by managers as potential habitat. The LiDAR model provided better\ud discrimination between occupied and unoccupied sites than did a model using variables derived from\ud Gradient Nearest Neighbor maps that were previously reported as important predictors of murrelet\ud occupancy (AUC=0.64, k=0.12). We also evaluated LiDAR metrics at 11 known murrelet nest sites. Two\ud LiDAR-derived variables accurately discriminated nest sites from random sites (average AUC=0.91).\ud LiDAR provided a means of quantifying 3-dimensional canopy structure with variables that are ecologically\ud relevant to murrelet nesting habitat, and have not been as accurately quantified by other mensuration\ud methods