Estimates of the relative effects of demographic parameters on population rates of change, and of the level of natural variation in these parameters, are necessary to address potential effects of perturbations on populations. We used a stochastic model, based on survival and reproduction estimates of the Porcupine Caribou (Rangifer tarandus granti) Herd (PCH), during 1983-89 and 1989-92 to obtain distributions of potential population rates of change (r). The distribution of r produced by 1,000 trajectories of our simulation model (1983-89, r̄ = 0.013; 1989-92, r̄ = 0.003) encompassed the rate of increase calculated from an independent series of photo-survey data over the same years (1983-89, r = 0.048; 1989-92, r = -0.035). Changes in adult female survival had the largest effect on r, followed by changes in calf survival. We hypothesized that petroleum development on calving grounds, or changes in calving and post-calving habitats due to global climate change, would affect model input parameters. A decline in annual adult female survival from 0.871 to 0.847, or a decline in annual calf survival from 0.518 to 0.472, would be sufficient to cause a declining population, if all other input estimates remained the same. We then used these lower survival rates, in conjunction with our estimated amount of among-year variation, to determine a range of resulting population trajectories. Stochastic models can be used to better understand dynamics of populations, optimize sampling investment, and evaluate potential effects of various factors on population growth.