AUTHORS: Yolanda F. Wiersma*, Memorial University of Newfoundland; R. Troy McMullin, Canadian Museum of Nature
ABSTRACT: Understanding how rare species are distributed across their range can be difficult due to heterogeneity between landscape units. We propose that lichen thalli along the trunk of a tree are analogous to habitat patches in a kilometers-extent landscape and hence can function as a model system. Applying the patch-mosaic model at smaller spatial extents allows for increased statistical power. We use this system to test whether landscapes with rare species are different from those without. We sampled macrolichen diversity along the trunk of 24 balsam fir trees in a stand on the Avalon Peninsula, Newfoundland, Canada, along with microclimate variables. We analysed difference in pattern by aspect and along the gradient of 1 m up the trunk as well as between trees containing a species of globally rare lichen vs. those that did not. We found no difference in total patch richness or abundance between the micro-landscapes. We found significantly consistent patterns in lichen patches along the trunk when we controlled for aspect and by aspect when we controlled for height up the tree. These patterns were similar on the trees with the rare species. Lichen species richness did not differ between trees containing the rare species vs. those that did not. Thus, lichen patch pattern is statistically similar between trees and, as such, these can be considered replicate landscape units and be used as model systems for observational and manipulative experiments to test questions about spatial pattern and process, such as those concerning distribution of rare species.
AUTHORS: Joel S. Brown1,2, Audrey R. Freischel1,2, and Mehdi Damaghi3Moffitt Cancer Center, 12902 USF Magnolia Drive, Tampa, FL 1Department of Integrated Mathematical Oncology2Cancer Biology and Evolution Program3Department of Cancer Physiology
ABSTRACT: Tumors are ecosystems populated by normal and cancer cells. Both can exhibit considerable diversity and spatial heterogeneity. Cancer cells vary in their resource uptake, modes of metabolism (glycolytic and acid producing versus more efficient oxidative phosphorylation), motility and susceptibility to the immune system. Distance to vasculature establishes gradients in cell densities and nutrients. Like riparian ecosystems, near blood may lead to denser packing of cancer cells under essentially “mesic” environments. Farther away, cells become less dense and subject to hypoxia, low glucose and acidic conditions (“xeric” environments). To examine competition between cancer cell types, we grew labeled MDA-mb-231 (viewed as highly glycolytic, aggressive, and glucose profligate) and MCF-7 breast cancer cells (viewed as less aggressive but more glucose efficient) in 96 well plates. The former were expected to do best under high glucose and low Ph; the latter under low glucose and normal Ph. Spheroids were established under all combinations of high and no glucose, low and normal Ph, and six different starting ratios. Microspheres were 3-D imaged every 24 hours. Unexpectedly, the MDA-mb-231 outcompeted the MCF-7 under all conditions. Spatially, MDA-mb-231cells were more uniformly distributed throughout the spheroid with highest densities at the outer edge where nutrients will be richest and toxic metabolites lowest. The MCF-7 formed a dense clustered ring inside the microsphere away from both the edge and center. This ring broke up into smaller clusters as they declined in numbers. MCF-7 cells’ need to adhere to each other may disadvantage them from accessing nutrients and from moving to the richer edge of the tumor. Competition assays within the context of landscape ecology permitted a more sophisticated and detailed measure of what makes for “aggressive” and successful cancer cells.
AUTHORS: Cathy D. Collins*, Bard College; Michelle H. Hersh, Sarah Lawrence College
ABSTRACT: Disease plays a key role in maintaining local plant diversity, but how and why pathogen pressure varies across landscapes is not well understood. Host and pathogen distributions in fragmented landscapes reflect not only the influence of area, isolation, and abiotic factors, but also the biotic interactions between the pathogen and its host. We examined plant richness and composition, fungal-mediated seed mortality, and seed-borne fungal diversity in an experimentally fragmented landscape in Kansas, U.S.A. We buried seeds of six plant species for a year in three landscape treatments: small patches, edges of large patches, and centers of large patches. Half of the seeds in each location were treated with fungicide. We measured germination success of exhumed seeds, and cultured fungi from seeds for identification using ribosomal DNA sequencing. If plant-pathogen interactions influence landscape-scale patterns of plant community structure in fragments, we expected to see: 1) patch size differences in the composition of seedbanks, above-ground plant communities, and seed fungal communities; and 2) pathogen-induced seed mortality that depended on where in landscape seeds were buried. For seeds buried without fungicide, landscape location was the strongest predictor of germination success. Across species, seed germination was lower in small patches, where soil temperatures were more extreme. In contrast, for seeds protected by fungicide, plant richness was a better predictor of seed mortality than patch size: fungicide provided the greatest benefits where host richness was high. Both plant and fungal community structure differed with fragment size. We also found that fungi show host preferences—a necessary condition for diversity-maintaining negative feedbacks. Our results suggest that by altering plant species richness and community composition, fragmentation alters pathogen pressure on seeds indirectly via changes in host community structure. Consequently, local plant-pathogen interactions may contribute to patterns of diversity in fragmented landscapes.
AUTHORS: Kimberly A. With*, Kansas State University
ABSTRACT: Although habitat loss and fragmentation occur at a landscape scale, their effects are typically studied at a patch scale (patch area and isolation effects), thereby agitating debate over whether it is the amount or configuration of habitat that has the greater ecological effect. This patch-based view is reflected in the design of most fragmentation experiments, where the size and relative isolation of patches are manipulated to create landscape patterns (a “bottom-up” approach). Conversely, one can adopt a landscape perspective, in which the total habitat amount and overall fragmentation of the entire landscape are the focus of study (a “top-down” approach). Using fractal neutral landscape models, in which both the amount and fragmentation of habitat are adjusted independently, I created an experimental model landscape system (EMLS) in the field to investigate the relative effects of habitat amount versus fragmentation on arthropod diversity and species interactions. The EMLS comprised a replicated series of plots (16 m x 16 m) that consisted of different amounts of red clover (10-80%) arrayed as either a clumped or fragmented distribution. Although habitat amount generally had a greater effect on patterns of species diversity in this system, the effect of fragmentation was more nuanced and complex, involving both direct and indirect effects. Contrary to conventional wisdom, the effect of fragmentation was not always greatest when habitat was limiting. Furthermore, the scaling of diversity in this system demonstrates why it is not always possible to scale-up from patches to make inferences about fragmentation effects at the landscape scale. This results of this EMLS thus underscore the need for more landscape-based studies—at traditionally defined scales—that investigate the relative effects of habitat loss and fragmentation.
AUTHORS: Amanda E Martin*, Lenore Fahrig – Geomatics and Landscape Ecology Research Laboratory, Carleton University
ABSTRACT: Human land use intensification – including urbanization, intensive agriculture, and increasing road density – can put wildlife species at risk of extinction. Most researchers expect that sedentary species are more sensitive to land use intensification than dispersive species, because dispersive species can recolonize or rescue local populations in a patchy landscape. This has led some conservation biologists and land managers to prioritize conservation of sedentary over dispersive species. Here we ask whether this is a reasonable strategy. Should we expect that sedentary species are generally more sensitive to the negative effects of land use intensification than dispersive species? And, under what conditions (if any) should we expect the opposite? We conducted a quantitative review using the results from 83 cross-species observational, experimental, or meta-analytic studies reporting 486 effects of species dispersal ability on their responses to land use intensification. As expected, sedentary species were typically more sensitive to the negative effects of land use intensification than dispersive species. Nevertheless, ~30% of the time dispersive species were more sensitive than sedentary species. Our analyses also suggest that the effect of dispersal ability on species’ sensitivity to land use intensification varies among taxa. For invertebrates and plants, sedentary species were typically more sensitive than dispersive species. In contrast, for vertebrates, dispersive species were typically more sensitive than sedentary species, and the same was true when comparing across taxonomic groups. Our findings suggest that one cannot simply assume that sedentary species are more sensitive to land use intensification than dispersive species.
AUTHORS: Paulson Des Brisay*, Patricia Rosa, Christoph Ng, Nicola Koper – University of Manitoba
ABSTRACT: Landscape ecology is often criticized for describing species distribution patterns, but failing to determine the mechanisms that explain those patterns. In addition, landscape ecology tends to focus on species distributions and coarse measures of productivity, to accommodate the vast spatial scales over which sampling must occur; however, this further compromises our abilities to understand mechanisms behind landscape-scale patterns. We illustrate how landscape-scale manipulative experiments can allow us to overcome these problems. As anthropogenic noise becomes more prevalent across habitats, its impacts on wildlife remain difficult to assess. To disentangle effects of industrial noise from confounding factors (i.e. well presence, human activity, roads), we developed a high-fidelity solar-powered broadcasting system to reproduce sound from wells in sites without wells. We are comparing these “noisy” sites with sites with no wells, sites containing broadcasting equipment that is turned off, sites with real and active wells, and sites with real wells that are turned off (silent); we are evaluating impacts of these treatments on abundance, nesting success, clutch size, age structure, body condition, stress physiology, adult and fledgling survival, communication, and parental care at nests of grassland songbirds. We found negative effects of noise exposure per se on nesting success of Chestnut-collared Longspurs, Savannah Sparrows and Sprague’s Pipits, and on abundance of Chestnut-collared Longspurs and Vesper Sparrows. Female Chestnut-collared Longspurs, but not males, showed elevated corticosterone levels near noise playbacks, and noise explains some but not all negative effects of wells on parental care. However, noise did not explain variation in age structure or body size. The experimental design was essential and effective in allowing us to quantify the reasons for edge effects from wells on diverse fitness-related parameters. Our results demonstrate that both noise reduction and minimizing the footprint of above-ground infrastructure are necessary for mitigating effects of oil and gas development.
AUTHORS: Jane Capozzelli*, Masters Student; Dr. Jim Miller, Professor and Director of Graduate Studies – Department of Natural Resources and Environmental Sciences, University of Illinois Urbana-Champaign
ABSTRACT: Due to fire suppression and overgrazing, woody encroachment has emerged as the greatest contemporary source of habitat loss and degradation of grasslands in the region. One strategy that may reduce woody encroachment is the fire-grazing interaction, which has gained traction on large and small reserves because it restores ecosystem function by recoupling fire and grazing. To understand whether this practice reduces woody encroachment we surveyed the composition, abundance, and morphology of woody plants on 11 experimental pastures. We utilized a long-term experiment with three fire and grazing treatments (patch-burn-graze [n = 4], graze-and-burn [n = 4], and burn-only [n = 3]). Each pasture was divided into three patches. In the patch-burn-graze treatment, one patch was burned annually. Cattle had free access to the entire pasture, but preferentially grazed the most recently burned patch more intensively than the other patches. The graze-and-burn treatment had one pasture-wide burn every three years with free access for cattle. The burn-only treatment had one pasture-wide burn every three to five years and no grazing. We established 120 100-x-2-m belt transects (6 – 17 per pasture) and recorded the species, the number of stems, and the maximum height of every plant > 0.5 m tall. In 2017, we sampled 4983 plants of 37 species. The burn-only treatment had the highest abundance of woody plants, driven largely by sprawling patches of Rubus allegheniensis (Allegheny blackberry). Woody plants were slightly less abundant in the patch-burn-graze than in the graze-and-burn treatment. Based on plant morphology, we found that cattle consumed woody plants in the patch-burn-graze treatment, but avoided eating them in the graze-and-burn treatment. Grazing damage likely increases plants’ susceptibility to fire, which may maintain low densities of woody plants long-term.
AUTHORS: Eric J. Gustafson*, Mark E. Kubiske, Brian R. Miranda – USDA Forest Service, Northern Research Station
ABSTRACT: The Aspen-FACE experiment generated 11 years of empirical data on the effect of elevated CO2 and ozone (O3) (alone and in combination) on the growth productivity of model aspen communities (multiple aspen clones, aspen-birch, and aspen-maple) in northern Wisconsin (USA). Uncertainty remains about how these short-term plot-level responses might play out at the landscape scale over multiple decades where climate change, competition, succession and disturbances interact with tree species-level traits such as optimal temperature for photosynthesis and shade and drought tolerance. In this study we used a physiology-based approach (PnET-Succession) within the LANDIS-II forest landscape model to scale the FACE results to broader temporal and spatial scales by mechanistically accounting for the globally changing drivers of temperature, precipitation, CO2 and O3. We calibrated the model using experimental results for each treatment factor (CO2 and O3) alone, and validated the model using the treatment factors in combination. When the experiment was continued by simulation for 80 years, the model predicted if and when various taxa would die from competitive failure and the expected biomass of each taxa under each treatment. All taxa did not occur together in the experiment, but when they were simulated together, the dynamics of competitive exclusion varied by treatment combination. We also used the model to simulate the competitive dynamics of species not included in the Aspen-FACE experiment, both at the site and landscape scale, providing insight into the implications of elevated CO2 and O3 on the dynamics of real forest landscapes of north-central USA.
AUTHORS: Vince L. Butitta, Stephen R. Carpenter, Luke C. Loken – Center for Limnology, University of Wisconsin-Madison; Michael L. Pace, Department of Environmental Sciences, University of Virginia; Emily H. Stanley, Center for Limnology, University of Wisconsin-Madison
ABSTRACT: Advancements in ecological theory have benefited greatly from whole-ecosystem experimentation, particularly in lakes. Lakes provide well-defined boundaries, are highly dynamic in time and space, and are relatively easy to replicate. Experimentation, however, has been an underutilized tool in the attempts to understand ecosystem resilience. Much recent work in ecosystem resilience has focused on detecting early warning signals of weakening resilience prior to experiencing an abrupt transition to an alternative ecosystem regime. In our study, we experimentally tested the detectability of theoretically predicted early warning signals during a whole-lake manipulation. We induced a cyanobacteria bloom by adding nutrients to an experimental lake and mapped fine-resolution spatial patterning of cyanobacteria using a mobile sensor platform. Prior to the bloom, we detected spatial early warning signals based on variance and spatial autocorrelation, as well as a new index based on the extreme values. This successful detection of spatial early warning indicators in a lake manipulation despite its fluid nature suggests the potential utility of spatial metrics for signaling ecological thresholds in other ecosystems.
AUTHORS: Carina Rauen Firkowski*, University of Toronto; Marc Cadotte, University of Toronto Scarborough; Marie-Josée Fortin, University of Toronto
ABSTRACT: Microcosm experiments are accepted as a tractable system that has been widely used to test ecological predictions, and provides biological validation for modelling approaches. Mathematical modelling provides a flexible and robust approach to test the generality of results through sensitivity analyses and incorporation of complexity, providing opportunity for a mechanistic understanding of observed ecological patterns.Here, we seek to determine a metacommunity’s ability to persist in the face of disturbance, while maintaining its structure and function. Since distinct environmental fluctuation regimes are important in defining metacommunity structure, one might predict a differential response between metacommunities to the combined effects of environmental fluctuation and disturbance impacts. Specifically, we investigate how the presence of environmental fluctuation influences the responses of metacommunities to the landscape disturbance of patch depletion.In a microcosm experiment, biological communities consisted of an initial species assemblage of nine protozoan species. Environmental fluctuation was implemented by manipulating light availability, where we compared the control of constant light to a treatment of “high” frequency of fluctuations. After an initial stabilizing period of 12 days, landscape disturbance treatments were applied through the depletion of 0% (control), 40% and 80% of the patches in the metacommunity. The experiment ran for an additional 30 days, allowing for a comprehensive time series assessment of ~50-300 generations. Each treatment level was replicated eight times. The dynamics of this microcosm experiment were simulated through a spatially explicit, species interaction metacommunity (SESIM) model. We assess the role of environmental fluctuations in defining the relative risk of metacommunities to disturbance, and which food web properties are good predictors/promoters of robustness and resilience. The presented methodological approach provides a flexible framework to understand how species respond to multiple disturbances.
AUTHORS: Arig Ibrahim-Hashim*, Robert Gillies, Robert Gatenby, Joel Brown – H. Lee Moffitt Cancer Center
ABSTRACT: Malignant cells in solid tumors export metabolically derived acid into surrounding stroma. This micro-environmental acidification can be viewed as a “niche construction”. As an evolutionary strategy, acid-production benefits the cancer cell by decreasing the fitness of non-adapted competitors. At the scale of the tumor ecosystem, acidity promotes the transition from in situ to invasive and metastatic cancer. However, there are significant costs to being an acid-producing/adapted cancer cell including reduced efficiency in energy production and increased energy demands for survival in low Ph. We suggest exploiting these costs as a novel therapeutic intervention. A spatially-explicit, agent-based model of the associated non-linear dynamics demonstrates that small increases in tumor pH, appropriately timed, can reduce the fitness advantage of the aggressive cancer cell, and maintain a non-invasive phenotype as the dominant population. We tested this in a genetically modified (TRAMP) mouse model of prostate cancer, wherein we could identify two competing cell types: C2 and C3. C2 cells exhibited the properties of the niche construction phenotype with significantly increased aerobic glycolysis, glycolytic capacity, acid production, glucose uptake, and invasion. In contrast, C3 cells retained a near-normal metabolic profile. Adding 200 mM NaHCO3 to the drinking water of TRAMP mice at age 4 weeks increased intra-prostatic pHe by 0.2 pH units and prevented transition from in situ to invasive cancer. In established tumors, an increase of intratumoral pH significantly decreased primary tumor growth and dramatically reduced metastases. Furthermore, using an experimental tumor construct, MCF7 and MDA-MB-231 breast cancer cells were co-injected into the mammary fat pad of mice. C2-like MDA-MB-231 cells dominated in untreated animals but C3-like MCF7 cells were selected and tumor growth slowed when intratumoral pH was increased. Our results demonstrate a novel strategy for using perturbations of complex tumor landscapes to steer the system into a less invasive eco-evolutionary trajectory.
AUTHORS: G. Darrel Jenerette*, University of California Riverside
ABSTRACT: The history, current practice, and future needs of landscape ecological research strongly support an expanded role of experimental approaches that complements the rich observational and modeling strengths of the field. Throughout the history of landscape ecology, researchers have used diverse experimental approaches that feature controlled manipulations to answer a broad range of questions. In many cases experiments can provide powerful information on relationships between spatial pattern and ecological functioning. Here I provide an overview of how experiments are used, explore case studies that highlight success with landscape ecological experiments, and suggest new directions for future experiments. In a recent literature survey fifteen different types of landscape ecological experiments were identified. Among these types they included research that identify landscape structure, evaluate how landscape structure influences ecological processes, and evaluate causes of landscape pattern formation. Approaches for conducting landscape experiments vary along multiple axes that represent continuums that reflect constraints from scalability to real landscapes, statistical treatment, and resources for implementation. The next generation of experiments are benefitting from explicit uses of scaling theories and harnessing extensive cyberinfrastructure. Future opportunities for landscape ecologists include expanded collaborations among distributed experiments, better representations of microbial-scale experiments, and direct evaluations of landscape interactions with global changes. Notably, the recent expansion of macrosystem research is providing increased need and opportunities for new landscape experimental designs.
