Ecosystem feedbacks contribute to geographic variation in plant–soil eco‐evolutionary dynamics across a fertility gradient

Abstract

Plant species that occur in different soil nutrient environments often vary in plant functional traits that affect soil nutrient cycling, creating positive feedbacks that reinforce nutrient availability. Variation in such nutrient feedbacks could affect plant fitness, but few studies have explored the eco‐evolutionary dynamics of within‐species nutrient feedbacks, and the role that soil communities may play in mediating plant evolutionary responses. We investigated whether a widespread Populus species regulates soil fertility, resulting in positive nutrient feedbacks that influence population divergence. Here, we combined field observations, a reciprocal transplant experiment, and soil community sequencing to test how nutrient feedbacks might facilitate plant adaptation to soil environments. We find that: (1) Plant populations exist along a soil fertility gradient in the field and display trait variation consistent with positive nutrient feedbacks, (2) populations assemble distinct soil prokaryotic communities that are structured by soil chemistry, (3) populations show inconsistent patterns of local adaptation to their soil communities, however (4) genetic variation in plant–soil interactions reinforces soil nutrient differences. Soil communities may mediate variation in plant adaptation across a soil fertility gradient such that plant–soil interactions reinforce positive nutrient feedbacks. Identifying the eco‐evolutionary links between plant trait variation, soil nutrients and microbial communities demonstrates that plants modify and adapt to soil environments. These results support the hypothesis that eco‐evolutionary feedbacks may result when plant–soil interactions (eco) affect ecosystem‐level processes that sustain population‐level feedbacks and selective gradients (evo). A http://onlinelibrary.wiley.com/doi/10.1111/1365-2435.13259/suppinfo is available for this article.

Feedbacks link ecosystem ecology and evolution across spatial and temporal scales: Empirical evidence and future directions

“…In terrestrial systems, growing evidence shows the importance of ecosystem legacy effects in plant–herbivore and plant–soil interactions (e.g. Fitzpatrick et al, ; Pregitzer, Bailey, Hart, & Schweitzer, , Van Nuland et al, , ). Further integrating large‐scale, geographic approaches with ecosystem perspectives (including ecosystem state factors) will likely improve our understanding of how the interplay among phenotypes, trophic dynamics and environmental context influences both ecosystem and evolutionary processes in aquatic and terrestrial ecosystems.…”

“…The best observational example for FB3 is evidenced by the long‐term soil nutrient gradient across the Hawaiian Islands in which divergent populations of Metrosideros polymorpha resulted from differences in litter traits that accelerated or slowed nutrient cycling depending on their position along a soil fertility gradient (Treseder & Vitousek, ; Vitousek, ). However, Van Nuland, et al () provide further direct evidence integrating FB's 1 – 3 across a landscape‐level soil fertility gradient.…”

“…Feedback 3 starts with the original abiotic gradient of ecosystem state factors that underlie FB1 and FB2 . Feedback 3 is driven by and can change due to the slow alteration of the ecosystem pool from the legacy effects of FB1 (Van Nuland, Ware, Bailey, & Schweitzer, ; Vitousek, ; Wooliver, Pfennigwerth, Bailey, & Schweitzer, ; Wooliver et al, ; Figure ). The ecosystem pool is composed of the abiotic factors (e.g.…”

“…abiotic origin ). The historical ecosystem pool can change through time due to gradual effects of past species interactions related to FB1 and FB2 , as the ecosystem pool and the historical ecosystem pool covary (Van Nuland et al, ; Vitousek, ). The eco‐evo feedback at this scale can be thought of as historical contingency because contemporary interactions are contingent on the ecosystem‐wide effects of prior interactions that vary geographically (Van Nuland et al, ; Senior et al, ).…”

The diversity of eco‐evolutionary dynamics: Comparing the feedbacks between ecology and evolution across scales

“…In this case, intraspecific interactions between individuals affect population‐level processes that sustain population‐level feedbacks and selection gradients. In contrast, the approach by Van Nuland, Ware, Bailey, and Schweitzer () encompassed several ecological levels. To isolate and evaluate the eco‐to‐evo and evo‐to‐eco pathways separately, they conducted a combination of field observations and reciprocal transplant experiments.…”