Cross-scale analysis reveals interacting predictors of annual and perennial cover in Northern Great Basin rangelands.

Exotic annual grass invasion is a widespread threat to the integrity of sagebrush ecosystems in Western North America. Although many predictors of annual grass prevalence and native perennial vegetation have been identified, there remains substantial uncertainty about how regional-scale and local-scale predictors interact to determine vegetation heterogeneity, and how associations between vegetation and cattle grazing vary with environmental context. Here, we conducted a regionally extensive, one-season field survey across burned and unburned, grazed, public lands in Oregon and Idaho, with plots stratified by aspect and distance to water within pastures to capture variation in environmental context and grazing intensity. We analyzed regional-scale and local-scale patterns of annual grass, perennial grass, and shrub cover, and examined to what extent plot-level variation was contingent on pasture-level predictions of site favorability. Annual grasses were widespread at burned and unburned sites alike, contrary to assumptions of annual grasses depending on fire, and more common at lower elevations and higher temperatures regionally, as well as on warmer slopes locally. Pasture-level grazing pressure interacted with temperature such that annual grass cover was associated positively with grazing pressure at higher temperatures but associated negatively with grazing pressure at lower temperatures. This suggests that pasture-level temperature and grazing relationships with annual grass abundance are complex and context dependent, although the causality of this relationship deserves further examination. At the plot-level within pastures, annual grass cover did not vary with grazing metrics, but perennial cover did; perennial grasses, for example, had lower cover closer to water sources, but higher cover at higher dung counts within a pasture, suggesting contrasting interpretations of these two grazing proxies. Importantly for predictions of ecosystem response to temperature change, we found that pasture-level and plot-level favorability interacted: perennial grasses had a higher plot-level cover on cooler slopes, and this difference across topography was starkest in pastures that were less favorable for perennial grasses regionally. Understanding the mechanisms behind cross-scale interactions and contingent responses of vegetation to grazing in these increasingly invaded ecosystems will be critical to land management in a changing world.

Restoration ecology through the lens of coexistence theory.

Advances in restoration ecology are needed to guide ecological restoration in a variable and changing world. Coexistence theory provides a framework for how variability in environmental conditions and species interactions affects species success. Here, we conceptually link coexistence theory and restoration ecology. First, including low-density growth rates (LDGRs), a classic metric of coexistence, can improve abundance-based restoration goals, because abundances are sensitive to initial treatments and ongoing variability. Second, growth-rate partitioning, developed to identify coexistence mechanisms, can improve restoration practice by informing site selection and indicating necessary interventions (e.g., site amelioration or competitor removal). Finally, coexistence methods can improve restoration assessment, because initial growth rates indicate trajectories, average growth rates measure success, and growth partitioning highlights interventions needed in future.

Application of modern coexistence theory to rare plant restoration provides early indication of restoration trajectories.

Restoration ecology commonly seeks to re-establish species of interest in degraded habitats. Despite a rich understanding of how succession influences re-establishment, there are several outstanding questions that remain unaddressed: are short-term abundances sufficient to determine long-term re-establishment success, and what factors contribute to unpredictable restorations outcomes? In other words, when restoration fails, is it because the restored habitat is substandard, because of strong competition with invasive species, or alternatively due to changing environmental conditions that would equally impact established populations? Here, we re-purpose tools developed from modern coexistence theory to address these questions, and apply them to an effort to restore the endangered Contra Costa goldfields (Lasthenia conjugens) in constructed ("restored") California vernal pools. Using 16 years of data, we construct a population model of L. conjugens, a species of conservation concern due primarily to habitat loss and invasion of exotic grasses. We show that initial, short-term appearances of restoration success from population abundances is misleading, as year-to-year fluctuations cause long-term population growth rates to fall below zero. The failure of constructed pools is driven by lower maximum growth rates compared with reference ("natural") pools, coupled with a stronger negative sensitivity to annual fluctuations in abiotic conditions that yield decreased maximum growth rates. Nonetheless, our modeling shows that fluctuations in competition (mainly with exotic grasses) benefit L. conjugens through periods of competitive release, especially in constructed pools of intermediate pool depth. We therefore show how reductions in invasives and seed addition in pools of particular depths could change the outcome of restoration for L. conjugens. By applying a largely theoretical framework to the urgent goal of ecological restoration, our study provides a blueprint for predicting restoration success, and identifies future actions to reverse species loss.

Disentangling key species interactions in diverse and heterogeneous communities: A Bayesian sparse modelling approach.

Modelling species interactions in diverse communities traditionally requires a prohibitively large number of species-interaction coefficients, especially when considering environmental dependence of parameters. We implemented Bayesian variable selection via sparsity-inducing priors on non-linear species abundance models to determine which species interactions should be retained and which can be represented as an average heterospecific interaction term, reducing the number of model parameters. We evaluated model performance using simulated communities, computing out-of-sample predictive accuracy and parameter recovery across different input sample sizes. We applied our method to a diverse empirical community, allowing us to disentangle the direct role of environmental gradients on species' intrinsic growth rates from indirect effects via competitive interactions. We also identified a few neighbouring species from the diverse community that had non-generic interactions with our focal species. This sparse modelling approach facilitates exploration of species interactions in diverse communities while maintaining a manageable number of parameters.