Woodlands across the globe have expanded in recent centuries in response to shifts in climate, land use, and disturbance patterns, but are expected to be particularly vulnerable to decline as a result of environmental change. Gaps in our understanding of the drivers of woodland spatial dynamics, however, limit our ability to forecast range shifts. While the effect of extrinsic factors, such as fire and climate, on woodlands are well-documented, we lack knowledge on the role of intrinsic factors such as density-dependence in driving tree abundances and distributions across space and time. We used two remotely-sensed datasets of tree cover and climate predictors in Bayesian state-space models to test the effects of climate and density on the spatiotemporal changes in pinyon pine (Pinus monophylla) cover in the Great Basin, USA. We compared models with only climate predictors to models including both climate and tree cover (i.e. density-dependence). Models including density-dependence better explained observed dynamics than climate alone. Pinyon pine cover increased at the fastest rate in areas with intermediate precipitation and low maximum temperatures. Spatial effects of climate interacted with temporal variation in climate. In cool, wet areas, growth was higher in warm, dry periods. In warm, dry areas, it was favorable to have cool periods. However, positive direct effects of climate at mid to high elevation were partially offset by density-dependence. Because extant density was positively correlated with average precipitation and negatively correlated with temperature, pinyon pine cover often declined in climatically favorable areas. Our results indicate that direct effects of warming and drying climate will likely lead to increases in cover at high elevations and declines at low elevations in the long-term. However, the added effects of density-dependence may create counter-intuitive patterns with increases in low density, low elevation areas in the near term and declines in seemingly favorable high density cooler-wetter locations. As a result, management intended to increase woodland resilience through silvicultural treatments like density reductions may have the greatest impact at intermediate elevations where increasing climate stress coincides with high tree densities.
Read full abstract