LAUSR

The center-periphery hypothesis predicts a decline in population performance toward the periphery of a species' range, reflecting an alteration of environmental conditions at range periphery. However, the rare demographic tests of this hypothesis failed to disentangle the role of geography from that of ecological niche and are biased toward temperate regions. We hypothesized that because species are expected to experience optimal abiotic conditions at their climatic niche center, (i) central populations will have better demographic growth, survival, and fertility than peripheral populations. As a result, (ii) central populations are expected to have higher growth rates than peripheral populations. Peripheral populations are expected to decline, thus limiting species range expansion beyond these boundaries. Because peripheral populations are expected to be in harsh environmental conditions, (iii) population growth rate will be more sensitive to perturbation of survival-growth rather than fertility in peripheral populations. Finally, we hypothesized that (iv) soils properties will drive the variations in population growth rates for narrowly distributed species for which small scale ecological factors could outweigh landscape level drivers. To test these hypotheses, we studied the demography of Thunbergia atacorensis (Acanthaceae), a range-limited herb in West Africa. We collected three years of demographic data to parametrize an integral projection model and estimated population level demographic statistics. Demographic vital rates and population growth rates did not change significantly with distance from geographic or climatic center, contrary to predictions. However, populations at the center of the geographic range were demographically more resilient to perturbation than those at the periphery. Soil nitrogen was the main driver of population growth rate variation. The relative influence of survival-growth on population growth rates exceeded that of fertility at the geographic range center while we observed the opposite pattern for climatic niche. Our study highlights the importance of local scale processes in shaping the dynamics and distribution of range-limited species. Our findings also suggest that the distinction between geographic distribution and climatic niche is important for a robust demographic test of the center-periphery hypothesis. This article is protected by copyright. All rights reserved.