Conservation and Ecology of Wildlife Communities II

Contributed Paper
ROOM: Room 130 – Cimarron

1:10PM Regenerating Clearcuts and Post-Harvest Treatments Increase Spruce-Fir Avian Assemblages and Richness in Managed Stands
Brian W. Rolek; Daniel J. Harrison; Cynthia S. Loftin; Petra B. Wood
Extent of spruce-fir forest and associated bird abundances have declined in recent decades emphasizing the need to better understand avian responses to forest management and to identify options that conserve or increase habitat for these birds. We sought to identify harvest practices and post-harvest treatments that provide habitat for spruce-fir passerine birds by surveying regenerating stands previously managed with prevalent forestry techniques. We conducted avian point counts and vegetation surveys on lands with known management histories to assess relationships among forest harvest and post-harvest treatments, and avian abundances and richness. We surveyed conifer-dominated and mixed coniferous-deciduous forests that were managed with six prevalent harvest treatments harvested five to 41 years prior to our study and mature reference stands ≥48 years prior. We sampled three harvest treatments including selection, shelterwood, and clearcut stands as well as post-harvest treatments of precommercial thinning (PCT), aerially applied herbicide (e.g. glyphosate), and both PCT and herbicide. Clearcuts with PCT and herbicide were in earlier successional stages, and had notably greater spruce-fir composition than other treatments. Avian assemblages of 49 bird species shifted between treatments, and spruce-fir birds tended to have greater abundances in stands with a greater representation of spruce-fir trees (≥70%), which included clearcut stands, and particularly clearcut stands subsequently treated with PCT and herbicide. Richness of spruce-fir birds and species of concern was greater in clearcuts and clearcuts with herbicide and PCT treatments. Conversely, selection and shelterwood harvests had the least richness among treatments, and were not statistically different from mature reference stands. Spruce-fir bird richness decreased with successional stage and increased with greater spruce-fir composition. Favoring use of clearcutting and post-harvest treatments over selection and shelterwood harvests may enhance habitat for spruce-fir birds in northern New England, southern Quebec, and Maritime Provinces of eastern Canada where transitional northern forests dominate the landscape.
1:30PM Metabarcoding with Environmental DNA to Identify Wildlife Species Potentially Attracted to Uranium Mine Containment Ponds as a Water Source in the Arid Southwest US
Katy Klymus; Catherine Richter; Nathan Thompson; Jo Ellen Hinck
Objectives/Methods: Development of new uranium mines in the Grand Canyon watershed in northern Arizona was restricted in 2009 by the federal government until studies assessing the potential impacts of radionuclide and heavy metal contamination could be completed. Water containment ponds at mines are designed to receive all surface run-off and contain elevated chemical concentrations. The ponds are also a constant water source in an arid region and could result in contaminant exposure to local food webs. To understand the heavy metal exposure pathways, we conducted environmental DNA metabarcoding in parallel with traditional biodiversity surveys via small mammal trapping and acoustic monitoring to identify wildlife using these water sources. With samples from surface water near active mines and mine containment ponds we employed a metabarcoding approach with 12S and 16S rRNA gene markers. Results: Using the 12S markers we recovered large numbers of sequence reads from taxa expected to be in the area and from less common or hard to observe taxa such as the Mexican free-tailed bat and the tiger salamander. Detection of the tiger salamander is of note because this species was not observed by the traditional biological survey techniques used. Due to low phylogenetic resolution of the 12S marker, most taxa were not identified down to species level. Using our 16S markers, we expect to improve our taxonomic resolution. We will compare our metabarcoding survey results with those from the traditional survey methods and also look at seasonal changes in species occurrence. Significance: As eDNA is quickly becoming a popular tool for wildlife surveys; we will discuss the advantages and limitations of this technique based on our experience. Ultimately this tool will enable us to better understand the overall biodiversity of the area and aid risk assessment of resuming new mining activities.
1:50PM A Global Estimate of Net Secondary Productivity: Wildlife in a Changing World
Garrett M. Street; Gordon G. McNickle
It is critical to estimate the fluxes of global carbon pools to understand the implications of human activities and policy on potential climate change scenarios. The terrestrial carbon pool is currently the largest source of uncertainty in balancing the global carbon budget. Most estimates of the terrestrial biosphere rely on remote sensing methods that only detect Net Primary Productivity (NPP) by plants. NPP is the majority of carbon in any terrestrial ecosystem; however, plants support a food chain of consumers, and Net Secondary Productivity (i.e., production of animals; NSP) at the global scale has the potential to contribute to the carbon budget of the terrestrial biosphere. Here we use gridded global NPP estimates to generate gridded estimates of global NSP based on 4 competing models of trophic efficiency assuming a simple 3 trophic level system. We estimate that terrestrial NPP could potentially support 4.28-24.0 Pg C/yr of NSP globally; however, major declines in animal population abundance have been observed across terrestrial biomes due to anthropogenic degradation of habitat. Correcting for these losses, we estimate that current global NSP is only 2.29-13.04 Pg C/yr, meaning that degradation and loss of natural habitat has resulted in the loss of storage capacity in global NSP of 1.94-10.95 Pg C/yr, equivalent to the loss of the entire desert and xeric shrubland biome on average every year. Our data show that many biodiversity hotspots across the globe are also NSP hotspots. These magnitudes mean that globally animal tissues store as much carbon as the vegetation in many individual biomes, and that failed conservation efforts have significantly reduced the capacity of the terrestrial biosphere to store carbon through the loss of animal habitat.
2:10PM Development of Multi-Species, Long-Term Monitoring Programs for Resource Management
Jeremy A. Baumgardt; Michael L. Morrison; Leonard A. Brennan; Brian L. Pierce; Tyler A. Campbell
Among landowners with conservation goals, there is a growing interest in long-term wildlife monitoring on their lands. Proper planning for monitoring includes determining the species, metrics, sampling methods, experimental design, and level of effort necessary to achieve the desired power for detecting meaningful changes. Failure to give these decisions proper attention often leads to wasted resources with insufficient information to meet the objectives. Our primary objectives were to develop multiple alternative scenarios for a monitoring program including power to detect changes in abundance and/or occupancy for rodents on 88,000 ha of rangelands in south Texas owned by the East Foundation. It was also our goal, to present steps that can be replicated elsewhere for developing customized monitoring programs. We trapped rodents using ~28,000 trapnights each year with 2 trap configurations from 2014 to 2016 resulting in 13,183 captures of 9 species. We estimated abundances and occupancy in each year for each species and conducted power analyses using simulation routines. We used these results to develop 4 multi-species monitoring scenarios, including costs to implement. The most effort-intensive scenario we designed requires trapping 40 locations with square grids for 6 consecutive nights each year. With this effort, we predict it would be possible to detect annual changes in abundance of 10% or greater over 10 years for 4 species and net declines in occupancy rates of 50% or greater over 10 years for 5 species. Our least-intensive scenario requires trapping 30 transects for 4 consecutive nights each year. We predict this effort would allow for the detection of annual changes in occupancy rates between 35 and 55% over 10 years for 5 species. Although there is no single best scenario, our study design, analyses, and results serve as a framework that can be used as a starting point for other such efforts.
2:30PM Predicting the Fundamental Niche and Global Distribution of Hermatypic Coral Reefs Using Machine Learning Algorithms
Vitalis Dubininkas; Stanley Mastrantonis
Recent events of widespread coral bleaching allude to the significant impact of climate change and anthropogenic pressure upon the shallow-water marine ecosystems known as coral reefs. Given the limitations of satellite imagery and predictions of average sea surface temperature rise to be 2°C by 2060, novel methods that aid conservation on a large scale are required. Previous studies have attempted to map the global distribution of tropical corals by directly calculating the area of reefs using nautical charts and by calculating continental shelf area using predictive techniques. Conversely, the present study aimed to predict the spatial distribution of tropical coral reefs by incorporating environmental data and elements of machine learning into three different species distribution models. By using habitat suitability of hermatypic corals as a proxy for reef presence, this study was able to predict the global distribution of tropical coral reefs with a 20 x 20 km resolution. It was observed that bathymetry, sea-surface temperatures, salinity, and pH significantly affected the likelihood of observing coral reef presence in all three models. The models also performed favourably under sensitivity analysis (AUC > 0.84), especially when compared to empirical observations of known coral reef systems. Areas which were predicted to have marginal-to-high suitability were used to approximate the spatial coverage on a global scale. Using the combined results of the models, it was estimated that suitable habitat for hermatypic corals is approximately 3.53 ± 0.18% of the world’s oceans. Nevertheless, calculations of pristine habitat (i.e. p ≥ 0.9 of finding a coral reef) were noted to be closer to 1.0 % of the world’s oceans. The outputs of the models, as well as the data that was generated as a result of this study, should be of interest to coral ecologists, wildlife biologists, fisheries managers, and groups within the petroleum industry.
2:50PM Refreshment Break
3:20PM Emergent Wetland Bee Communites in the Lower Mississippi Alluvial Valley of Arkansas
Phillip L. Stephenson
Farm Bill programs like the wetland reserve program, provide ecological services including water and soil conservation as well as meeting wildlife and fisheries needs. The value of these wetlands to pollinators, in particular bees, has not been examined despite these wetlands producing large numbers of flowering plants. Too, these wetlands are usually surrounded by croplands that might benefit from pollination services provided by wetland bees. In this study, we compared species richness and diversity of pollinator communities between actively and passively managed palustrine emergent wetlands in the lower Mississippi Alluvial Valley in Arkansas throughout the growing season. Active management practices include disking, mowing, water level manipulation while passive management practices include natural drying. Solitary bees (Hymenoptera: Apoidea) were surveyed using pan traps, blue-vane traps, and sweep nets. In 2016, weWe collected 19,615 individual bees that included five families, 31 genera, and 87 species. Of these species, five (Anthophorula asteris, Ceratina cockerelli, Diadasia enavata, Diaunomia triangulifera, Svastra cressonii) were new Arkansas state records. Our estimated species richness in actively managed emergent wetlands (70.7, 95% CI = 61.49-84.33) overlapped with our estimated species richness in passively managed emergent wetland (83.5, 95% CI = 68.00 – 101.66). We found that the probability of detecting a species in actively (0.86, 95% CI = 0.722-0.986) and passively (0.78, 95% CI = 0.637-0.946) managed emergent wetlands were both high. Shannon-Weiner diversity estimates were not significantly different between actively (1.8, 95% CI = 1.72 – 1.94) and passively (1.6, 95% CI = 1.50 – 1.74) managed emergent wetlands. Actively and passively managed emergent wetlands support a similar suite of bee species throughout the growing season. The USDA Farm Bill wetland easement programs can create suitable resources for pollinators and so further justifies this important management program.
3:40PM Data, Sweat, Policy, Law and Ethics – Sharing Wildlife Data in the Instant Access Age
Kimberly Titus
Recent editors of JWM along with WSB papers suggest that wildlife professionals need to do a better job of publishing results and archiving data for broader use. I agree, but also note that open access collected by agencies often comes with various constraints. First, sweat. Agency biologists have sweat and risk collecting wildlife data and feel some inherent right to ‘their’ data, although the public actually own the data held in trust by the agency. Second, policy. Agencies often have policies regarding how data is collected, stored, published and shared. Some agencies embargo data. Third, there are many laws requiring data disclosure and open access. However, many states also prohibit release of sensitive wildlife data. This creates a conundrum because some laws are old and don’t recognize technology changes. Fourth, industry data can be proprietary and subject to a contract. Fifth, data are provided other agencies for legal decisions – even if it has not been published. I suggest that there are age-related differences among wildlife professionals and expectations on the ease of using data they did not collect. Presently, someone with a user ID and password can access wildlife location data in near real-time. Projects using this technology often have many cooperators/agencies necessitating written agreements to avoid ethical and legal conflict. One agency buys the satellite tags, industry provides partial funding, another agency deploys the tags and there is a graduate student involved. Who owns and interprets the data? The time when agency biologists used handshake agreements to control data access is gone. There is increasing pressure to make data available for management decisions, often before an analysis by the investigators. Get it in writing.
4:00PM A General Decision Framework for Quantifying Urbanization Gradients
Benjamin J. Padilla; Chris Sutherland
The field of urban ecology has grown out of decades of increasing interest in how urbanization shapes ecological processes. As our understanding improves, the focus has now shifted towards identifying the mechanisms underlying biological responses along urban gradients, and a desire to test the generality of emerging predictions across space and time. Unfortunately, methods used to define urbanization and produce landscape gradients vary widely, making it challenging to compare across studies and reach general conclusions. Based on a review of 255 papers published between 2007 and 2017 that used a landscape gradient approach to examine the impacts of urbanization on an ecological process, we developed a general decision framework for defining urban gradients. In each study analyzed, we identified key common features including (1) the type of landscape classification (e.g. continuous or categorical), (2) the spatial scale of landscape analysis (e.g. site or landscape level), and (3) the definition of urbanization used (e.g. impervious surface or population density). Despite some patterns in the data, there was substantial variation in the rational and types of decisions made steps, differences that significantly influence results. The general decision framework we develop here describes the key steps for defining urban gradients, taking into consideration the spatial scale, landscape analysis methods, and definition of urbanization, and offers guidance on making context specific choices within this framework allowing for critical comparisons with other studies.
4:20PM Prairie Songbird Responses to Bison Grazing at the National Bison Range and Yellowstone National Park
Danielle A. Fagre; Victoria J. Dreitz, PhD
In prairie ecosystems, grazing is a major disturbance that serves as an ecological process. Grazing can affect plant species diversity and abundance, and vegetative structure. Historically, plains bison (Bison bison) were major grazers in North American prairies. Bison are considered ecosystem engineers, meaning they modify habitat to be more or less suitable for other species. They created and responded to heterogeneity at the landscape level through grazing and other behaviors. In turn, other species responded to this heterogeneity. Contemporary bison grazing is quite different from historic grazing. Bison are restricted to smaller spatial scales, altering their forage and movement behavior compared to historic conditions. Little is known about whether contemporary bison grazing fulfills the same ecological process as it once did, and how it influences other members of the biological community. Prairie songbirds evolved in landscapes that contained a gradient of differentially grazed vegetation. Research has shown prairie songbirds to have associations with vegetation structure. Because grazing alters vegetation, it may create the vegetative structure required by prairie songbirds. This avian guild has declined more than any other, making it imperative to understand this ecological relationship. We investigate prairie songbird abundance in relation to the bison grazing intensity as a means of understanding contemporary bison grazing and songbird habitat selection. We use bird surveys and bison grazing surveys to model the effect of bison grazing intensity on the abundance of multiple bird species. We present results from 2016 at the National Bison Range and Yellowstone National Park, showing species-specific responses to bison grazing. These responses vary by location due to bison management strategies and biotic characteristics of each study site. We conclude that bison grazing intensity may be one of many components of songbird habitat selection. Our results will inform prairie songbird conservation, bison reintroduction efforts, and current bison management.
4:40PM Livestock Grazing Effects on Western Terrestrial Vertebrates: A Systematic Review and Meta-Analysis
Daniel C. Barton; Tim Bean; Abigail Rutrough
Livestock grazing is the predominant land use in western North America. Numerous studies have addressed direct and indirect effects of grazing on western vertebrate wildlife, but inference from these studies is often limited by design challenges. Lack of suitable reference or control sites, partial control of treatment levels, and lack of replication are imposed on grazing effects research by current and historic management of western livestock grazing. We used formal meta-analysis of published treatment-control and before-after designs of livestock grazing effects on terrestrial vertebrates to attempt to surmount these design challenges. Our goals were twofold: first, to estimate effect sizes and variance across broad vertebrate taxonomic groupings, and second, to explore moderators of effect size representing study design, vegetation community, and geographic region. We initially reviewed over 280 studies and identified over 90 for meta-analysis admission using a repeatable and objective search and admission process. In this sample, we found effect size of grazing was highly variable among and within broad taxonomic groupings, and effect size ranged from strongly negative to strongly positive, even within narrower taxonomic levels. The mean effect size was negative, and most studies showed negative effects. Important moderators included study design and community variables. We then recruited several subject-area experts to review our initial sample and to identify papers for inclusion that were not included in the initial sample, to explore the effects of more subjective literature review processes on our results. Overall, we conclude that some groups of terrestrial vertebrates show consistent negative effects of livestock grazing, and suggest design considerations for future studies identified by our meta-analysis.


Contributed Paper
Location: Albuquerque Convention Center Date: September 27, 2017 Time: 1:10 pm - 5:00 pm