Current prediction of the ecological value of northern caribou ranges is based on the Caribou Habitat Assessment and Supply Estimator (CHASE) framework (McNay and Zimmerman 2002). Several range types are considered in this framework: 1) calving and summer range; 2) pine-lichen winter range; 3) high-elevation winter range; and 4) movement corridors. Each range type is influenced by the risk of predation by wolves, for which wolf density was estimated based on its relationship to moose density. The objective was to develop a conceptual model for projecting populations of Northern Caribou in the Mackenzie area of B.C. using the projected habitat quality and predation risk indicators from the CHASE seasonal range models, and population parameters measured from the area. This involved simulating a population by randomly sampling demographic and environmental variables from defined probabilities and then analyzing their cumulative effects on the state of the population over time. Model needs to capture the response of the population to changes in the distribution and amount of habitat within each seasonal range type. Therefore a spatial population model was necessary. This report provides documentation on significant modifications made to the conceptual model for spatial population modelling for the northern caribou population model, as well as a description of the semi-spatial stage based population model prototype developed for efficient model testing and sensitivity analysis. Included in the report are the results from a series of sensitivity analyses designed to test model response to various parameter states. Specifically, I tested the response of the model to 1) static vs. dynamic landscape; 2) density dependant mortality in avoided habitat; 3) sensitivity to systematic variation in annual vital rates; 4) sensitivity of the model to systematic variation in carrying capacity in each habitat category; 5) response to different partitioning of annual mortality among seasonal ranges; 6) response to varying the proportioning of mortality in predation risk classes.
O?Brien, Dan. 2005. Model for assessment of Northern Caribou population response to projections of seasonal range conditions and predation risk: model description & outputs for sensitivity analysis and testing. Forest Investment Account (FIA) - Forest Science Program. Forest Investment Account Report. FIA2005MR127