Variation in the abundance of animals has traditionally been explained as the outcome of endogenous forcing from density dependence
and exogenous forcing arising from variation in weather and predation. Emerging evidence suggests that the effects of density
dependence interact with external influences on population dynamics. In particular, spatial heterogeneity in resources and
the presence of capable predators may weaken feedbacks from density dependence to growth of populations. We used the Kalman
filter to analyze 23 time series of estimates of abundance of northern ungulate populations arrayed along a latitudinal gradient
(latitude range of 40°–70°N) to evaluate the influence of spatial heterogeneity in resources and predation on density dependence.
We also used contingency tables to test whether density dependence was independent of the presence of carnivores (our estimate
of predation) and multiple regressions to determine the effects of spatial heterogeneity in resources, predation, and latitude
on the strength of density dependence. Our results showed that the strength of density dependence of ungulate populations
was low in the presence of large carnivores, particularly at northern latitudes with low primary productivity. We found that
heterogeneity in elevation, which we assume acted as a surrogate for spatial heterogeneity in plant phenology, also reduced
effects of density dependence. Thus, we show that external forces created by heterogeneity in resources and predation interact
with internal feedbacks from population density to shape dynamics of populations of northern ungulates.
Keywords Gompertz model - Kalman filter - Large carnivores - Local environmental variability - Predation - Spatial heterogeneity