Recruitment failure, climate change and community interactions in California valley oak woodlands.

摘要

The absence of sapling-sized individuals in many California valley oak (Quercus lobata) populations has perplexed naturalists and ecologists over the past century. Field observations of skewed size class distributions have lead to concern about the long-term viability of this endemic keystone species, and to questions about patterns of recruitment in a long-lived, low-density species. Over the past 30 years, a robust body of experimental work on valley oak seedling survival has increased our understanding of mechanistic drivers of recruitment. However because of the species' complex life-history, wide distribution and frequent occurrence on private land, a comprehensive assessment of recruitment across the distribution has been lacking. Based on a meta-analysis of planting studies, land manager interviews, dendroecology, historic surveys and new recruitment surveys, we attempt to address the questions of where and under what circumstances valley oaks recruit in natural populations. We conducted a resurvey study of ten historically surveyed sites and a field study of demographic patterns in 26 sites with valley oak sapling recruitment across the species distribution. We found sapling recruitment in a variety of systems, climates and land management regimes, and an increase in presence of recruitment over time in the resurveyed sites. Our results indicate that saplings are restricted to a subset of the adult distribution and certain populations may be at risk of decline. However, sapling recruitment in this species is more extensive than had been reported in the literature and its long- term conservation and restoration prospects may be more tractable than previously anticipated, particularly on working rangelands.;Questions on the importance of top-down vs. bottom-up effects on population regulation have long occupied ecologists. Recent work has delved into how these factors may shift in strength across both time and space, highlighting the need for larger scales of study. We looked at bottom-up forces (water availability) and top-down forces (small mammal herbivory) on valley oak ( Quercus lobata) sapling recruitment. We use the results of a meta-analysis on valley oak planting experiments, interviews with land stewards, and an analysis of field data on natural populations comparing recruiting and non-recruiting areas to get at questions about how population regulation may shift across a spatial precipitation gradient Our field study spanned 25 sites across the species distribution, over a gradient of 400-1000 mm average annual precipitation. This approach allowed us to study time-integrated processes in a long-lived system that would be impossible to manipulate in a solely experimental setting. We found that the relative importance of top-down and bottom-up forces in limiting oak recruitment shifts across a precipitation gradient, with small mammalian herbivores (top-down forces) more important at higher-precipitation sites and moisture stress (bottom-up forces) more important at lower-precipitation sites. Our study demonstrates the importance of expanding the spatial scale of studies on community dynamics, in order to better understand the multiple factors regulating species.;The ability to anticipate and prepare for climate change is a major focus in conservation ecology, and species bioclimate models are one of few available predictive tools. Most bioclimate models are parameterized with data from adult life-stages and focus on distribution-scale shifts, limiting their ability to project changes for long-lived species subject to strong local interactions. We assess bioclimate model projections for California valley oaks ( Q. lobata) against actual demographic patterns at the projected expanding and contracting edges of the current distribution, and evaluate whether patterns of oak recruitment at these edges correspond to the projected shifts. The valley oak bioclimate model does not explain the distribution of recruiting sites across projection categories. However it does explain differences in local-scale patterns of recruitment, with increased constriction of saplings around water bodies in the projected contracting areas of the distribution. Key climate parameters associated with oak saplings are different from those associated with adults. Recruitment appears to be restricted to cooler and wetter sites than is reflected in the distribution of adults, suggesting overprediction of adaptive shifts from a model parameterized by adult-stage distribution data. Particularly in species with complex life histories, analyses of life-stage vulnerability and micro-refugia are central to understanding climate-related species movement. Refining bioclimate models with targeted biological fieldwork on emergent changes in spatial demographic patterns could improve the reliability of projected species responses to climate change.