Predicting species distributions: unifying model selection and scale optimization for multi‐scale occupancy models

Domains of scale in cumulative effects of energy sector development on boreal birds

Context

Industrial development in Canada's boreal forest creates cumulative environmental effects on biodiversity. Some effects may be scale-dependent, creating uncertainty in understanding and hindering effective management.

Objectives

We estimated cumulative effects of energy sector development on distributions of sixteen migratory songbird species at multiple spatial scales within the boreal region of Alberta, Canada, and evaluated evidence for scale domains in species responses.

Methods

We used a hierarchical, multi-scale sampling and modelling framework to compare effects of oil and gas footprint on songbirds at five spatial scales. We used Bayesian Lasso to facilitate direct comparison of parameter estimates across scales, and tested for differences in grouped parameter estimates among species.

Results

We found consistent scale-dependent patterns across species, showing variable responses to development occurring at the smallest scale, little effect at intermediate scales, and stronger, mainly positive effects at the largest scales. Differences in grouped parameter estimates across scales showed strong evidence for scale domains in the response of songbirds to energy sector development.

Conclusions

We concluded that variable effects at the smallest scale represented individual habitat selection, while larger scale positive effects reflected expanding distributions of open habitat- and disturbance-associated species in areas of high oil and gas footprint. Our results show that single-scale analyses do not reflect population processes occurring at other scales. Future research on linking patterns at different scales is required to fully understand cumulative effects of land use change on wildlife populations.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10980-023-01779-8.

Implications of tree expansion in shrubland ecosystems for two generalist avian predators

Shrublands globally have undergone structural changes due to plant invasions, including the expansion of native trees. Removal of native conifer trees, especially juniper (Juniperus spp.), is occurring across the Great Basin of the western U.S. to support declining sagebrush (Artemisia spp.) habitats and associated wildlife species, such as greater sage-grouse (Centrocercus urophasianus). One justification for conifer removal is that it may improve survival of sagebrush-associated wildlife by reducing the abundance of avian predators. However, the relationship between conifer expansion and predator distributions has not been explicitly evaluated. Further, although structural characteristics of habitat are important for generalist predators, overall prey abundance may also affect habitat use by predators. We examined habitat use of common ravens (Corvus corax) and red-tailed hawks (Buteo jamaicensis), two generalist predators whose populations are increasing in western North America, to variation in structural characteristics and prey distributions in sagebrush habitat that has experienced conifer expansion. Structural characteristics of habitat were important predictors of habitat use for both ravens and red-tailed hawks, whereas measures of prey abundance were unimportant for both species likely because generalist predators can use a wide variety of food resources. Ravens, but not red-tailed hawks, responded positively to increasing cover of juniper and the probability of habitat use was highest (> 0.95) where juniper cover within 100 m was > 20%. Habitat use by red-tailed hawks, but not ravens, was greater near cliffs but was not associated with juniper cover. Our study suggests that the removal of conifer in similar environments may lower the probability of habitat use for ravens, a common predator with significant impacts on many prey species. Therefore, we suggest conifer removal may improve sage-grouse reproductive success and survival depending on responses to conifer removal from other predators. Our results may be reflective of similar changes in rangeland ecosystems around the world undergoing expansion of conifer and other woody vegetation. Though species identities differ from sagebrush habitats, generalist avian predators in other habitats may have similar relationships with structural resources.

Effects of large-scale disturbance on animal space use: Functional responses by greater sage-grouse after megafire

Global change has altered the nature of disturbance regimes, and megafire events are increasingly common. Megafires result in immediate changes to habitat available to terrestrial wildlife over broad landscapes, yet we know surprisingly little about how such changes shape space use of sensitive species in habitat that remains. Functional responses provide a framework for understanding and predicting changes in space use following habitat alteration, but no previous studies have assessed functional responses as a consequence of megafire. We studied space use and tested for functional responses in habitat use by breeding greater sage-grouse (Centrocercus urophasianus) before and after landscape-level changes induced by a >40,000 ha, high-intensity megafire that burned sagebrush steppe in eastern Idaho, USA. We also incorporated functional responses into predictive resource selection functions (RSFs) to map breeding habitat before and after the fire. Megafire had strong effects on the distribution of available resources and resulted in context-dependent habitat use that was heterogeneous across different components of habitat. We observed functional responses in the use and selection of a variety of resources (shrubs and herbaceous vegetation) for both nesting and brood rearing. Functional responses in the use of nesting habitat were influenced by the overarching effect of megafire on vegetation, whereas responses during brood rearing appeared to be driven by individual variation in available resources that were conditional on nest locations. Importantly, RSFs built using data collected prior to the burn also had poor transferability for predicting space use in a post-megafire landscape. These results have strong implications for understanding and predicting how animals respond to a rapidly changing environment, given that increased severity, frequency, and extent of wildfire are consequences of global change with the capacity to reshape ecosystems. We therefore demonstrate a conceptual framework to better understand space use and aid habitat conservation for wildlife in a rapidly changing world.

Multi-species occupancy modeling provides novel insights into amphibian metacommunity structure and wetland restoration

A fundamental goal of community ecology is to understand species-habitat relationships and how they shape metacommunity structure. Recent advances in occupancy modeling enable habitat relationships to be assessed for both common and rare species within metacommunities using multi-species occupancy models (MSOM). These models account for imperfect species detection and offer considerable advantages over other analytical tools commonly used for community analyses under the elements of metacommunity structure (EMS) framework. Here, we demonstrate that MSOM can be used to infer habitat relationships and test metacommunity theory, using amphibians. Repeated frog surveys were undertaken at 55 wetland sites in southeastern Australia. We detected 11 frog species from three families (Limnodynastidae, Myobatrachidae, and Pelodryadidae). The rarest species was detected at only one site whereas the most common species was detected at 42 sites (naive occupancy rate 0.02-0.76). Two models were assessed representing two competing hypotheses; the best-supported model included the covariates distance to the nearest site (connectivity), wetland area, presence of the non-native eastern mosquitofish (Gambusia holbrooki), proportion cover of emergent vegetation, an interaction term between Gambusia and emergent vegetation cover, and the proportion canopy cover over a site. Hydroperiod played no detectable role in metacommunity structure. We found species-habitat relationships that fit with current metacommunity theory: occupancy increased with wetland area and connectivity. There was a strong negative relationship between occupancy and the presence of predatory Gambusia, and a positive interaction between Gambusia and emergent vegetation. The presence of canopy cover strongly increased occupancy for several tree frog species, highlighting the importance of terrestrial habitat for amphibian community structure. We demonstrated how responses by amphibians to environmental covariates at the species level can be linked to occupancy patterns at the metacommunity scale. Our results have clear management implications: wetland restoration projects for amphibians and likely other taxa should maximize wetland area and connectivity, establish partial canopy cover, and eradicate Gambusia or provide aquatic vegetation to mitigate the impact of this non-native fish. We strongly advocate the use of MSOM to elucidate the habitat drivers behind animal occupancy patterns and to derive unbiased occupancy estimates for monitoring programs.

Histoecology: Applying Ecological Principles and Approaches to Describe and Predict Tumor Ecosystem Dynamics Across Space and Time

Cancer cells exist within a complex spatially structured ecosystem composed of resources and different cell types. As the selective pressures imposed by this environment determine the fate of cancer cells, an improved understanding of how this ecosystem evolves will better elucidate how tumors grow and respond to therapy. State of the art imaging methods can now provide highly resolved descriptions of the microenvironment, yielding the data required for a thorough study of its role in tumor growth and treatment resistance. The field of landscape ecology has been studying such species-environment relationship for decades, and offers many tools and perspectives that cancer researchers could greatly benefit from. Here, we discuss one such tool, species distribution modeling (SDM), that has the potential to, among other things, identify critical environmental factors that drive tumor evolution and predict response to therapy. SDMs only scratch the surface of how ecological theory and methods can be applied to cancer, and we believe further integration will take cancer research in exciting new and productive directions. Significance: Here we describe how species distribution modeling can be used to quantitatively describe the complex relationship between tumor cells and their microenvironment. Such a description facilitates a deeper understanding of cancers eco-evolutionary dynamics, which in turn sheds light on the factors that drive tumor growth and response to treatment.

Identifying important military installations for continental-scale conservation of marsh bird breeding habitat

Degradation of wetland ecosystems has negatively impacted many species, perhaps none more so than marsh birds that breed in vegetative emergent wetlands throughout North America. The U.S. Department of Defense manages approximately 29 million acres of land within the continental U.S., and many military installations contain wetland complexes that may be important for wetland birds. Thus, failure to adequately manage habitat for marsh birds could result in species extirpations and additional listings under the Endangered Species Act, and may result in regulatory burdens that reduce military readiness. We conducted spatial analyses to identify important breeding habitat on > 500 military installations for 12 species of marsh birds, with the goal of identifying installations that are, and are not, likely to harbor breeding habitat for each species. We also sought to assess the local value of military installations for species of greatest concern by comparing habitat suitability within installations to that in areas directly adjacent to those sites. We built range-wide, spatially-explicit models of species distribution to project suitability of breeding habitat for marsh birds within and adjacent to military installations. Our results demonstrate that installations with the best marsh bird habitat are geographically aggregated (both among and within species), primarily at sites along the eastern seaboard and within the southern U.S. In addition, only a few sites appear to contain high-quality habitat for most species. Five or fewer sites contained most of the high-quality habitat for 9 of 12 species, whereas most of the high-quality habitat for remaining species was found at ≤ 10 sites. This work fills an information gap regarding the distribution of breeding habitat for marsh birds on military lands across the U.S., and should facilitate both strategic conservation of habitat over broad scales and the integration of marsh birds into management efforts at the site level. Our analyses also identify installations that are not likely to harbor breeding habitat for priority species, and thus should help minimize conflicts between needs of the military and marsh-bird conservation.