Midwest Fish & Wildlife Conference 2019

AUTHORS: Curtis P. Wagner, Kevin S. Page – Ohio Division of Wildlife

ABSTRACT: Muskellunge fisheries in Ohio are maintained through stocking.  The Ohio Division of Wildlife (ODOW) stocks approximately 20,000 advanced-fingerling (10–12 inches) Muskellunge annually among nine reservoirs (1 fish/acre).  Currently, voluntary angler reports of Muskellunge catches provide managers with information on the locations, numbers, sizes, and harvest of Muskellunge.  However, this voluntary reporting approach potentially misses critical information on population dynamics metrics such as survival, escapement, and the probability of catching a fish.  To provide additional information on which to evaluate Muskellunge population dynamics in Ohio, the ODOW initiated a long-term tagging study.  Starting in 2013, all Muskellunge stocked into four study reservoirs (>43,000) have been implanted with passive integrated transponders (PIT).  Escapement of Muskellunge is monitored using in-stream PIT detection systems stationed within dam spillways.  Anglers report in-reservoir recaptures using handheld PIT tag readers.  To date, more than 850 implanted Muskellunge have been detected or reported. Focusing on the escapement component of the study, we found that escapement of Muskellunge appears to vary seasonally and depend on the type of dam water control structure.  For one reservoir, the probability of escapement was 4 – 36%, annually.  Together, these estimates provide a more comprehensive picture of Muskellunge fisheries in Ohio reservoirs.

AUTHORS: Joseph D. Schmitt, Christopher S. Vandergoot, Richard T. Kraus – USGS Great Lakes Science Center, Lake Erie Biological Station

ABSTRACT: Populations of coldwater fishes such as burbot Lota lota, lake whitefish Coregonus clupeaformis, and lake trout Salvelinus namaycush have declined in Lake Erie, while cisco Coregonus artedi have been extirpated. Warming temperature regimes and the re-eutrophication of Lake Erie have increased the frequency of harmful algal blooms and hypoxic events, which can reduce oxythermal habitat availability. Using vertical profile data collected in the central basin from 2008-2017, we developed generalized additive models to explore spatial, seasonal, and interannual trends in oxythermal habitat availability for lake trout, cisco, lake whitefish, and burbot based on published oxythermal niche benchmarks. Habitat availability was usually temperature-limited rather than oxygen-limited, and significant (P<0.05) monthly and interannual variations in habitat availability were detected for most species. In general, oxythermal habitat was most limited during August and September; moreover, significant interannual trends in habitat availability were also detected, with 2016 having the most extreme habitat reduction for many species due to record high temperatures. Understanding the spatiotemporal dynamics of oxythermal habitat availability will be important for the conservation and restoration of these fishes in our changing climate. Moreover, these models can be integrated with climate predictions to better understand how warming temperatures will affect coldwater habitat in the future.

AUTHORS: Emy Monroe, Erica Mize – U.S. Fish and Wildlife Service; Richard Erickson, Christopher Merkes – US Geological Survey; Nicholas Berndt, Katherine Bockrath, Jeena Credico, Nikolas Grueneis, Jenna Merry, Kyle Mosel, Maren Tuttle-Lau, Kyle Von Ruden, Zebadiah Woiak, Kelly Baerwaldt, Sam Finney – U.S. Fish and Wildlife Service; Jon Amberg, US Geological Survey

ABSTRACT: Natural resource managers use data from survey or monitoring efforts that use a wide variety of tools. Environmental DNA (eDNA) is a genetic surveillance tool for detecting species and holds potential as a tool for large-scale monitoring programs.  Two challenges of eDNA-based studies are imperfect capture of eDNA in collection samples (e.g., water field samples) and imperfect detection of eDNA using molecular methods (e.g., quantitative PCR), which create uncertainty about sample designs for eDNA-based monitoring.  We used an occurrence model to address these challenges and determine how many  samples were required to detect species using eDNA and to examine when and where to take samples.  Water samples were collected from three different habitat types in the Upper Mississippi River when both Bighead Carp and Silver Carp were known to be present based on telemetry detections.  Each habitat type was sampled during April, May and November.  Detections of eDNA for both species varied across sites and months, but were generally low, 0 - 19.3% of samples were positive for eDNA.  Additionally, we found statistical artifacts where sample eDNA capture probabilities would artificially inflate estimates of molecular detection probabilities.  Overall, we found that eDNA-based sampling holds promise to be a powerful monitoring tool for resource managers, however, limitations of eDNA-based sampling include different biological and ecological characteristics of target species as well as aspects of different physical environments that impact the implementation of these methods.

AUTHORS: Michael J. Dreslik, Illinois Natural History Survey; John A. Crawford, National Great Rivers Research and Education Center; Sarah J. Baker, Illinois Natural History Survey; Christopher A. Phillips, Illinois Natural History Survey

ABSTRACT: For effective conservation and recovery, an adaptive management framework is often best when paired with monitoring population-level responses. In many species, monitoring abundances over time using traditional capture-mark-recapture (CMR) methods is logistically challenging. N-mixture models are an extension of the occupancy and detection probability framework and can estimate abundances across multiple populations. The models use raw abundance counts taken during surveys, model the distributions of capture frequencies, incorporate density-dependent effects and can provide population estimates when recaptures are too few. When validated with traditional CMR estimates, they can provide robust estimates for multiple populations across the landscape. We chose to determine the effectiveness of an N-mixture modeling approach to generate population size estimates for the Eastern Massasaugas within the Carlyle Lake region in Illinois. Our results will be used to determine regional population trends and provide a foundation to assess the effectiveness of conservation actions.

AUTHORS: Seth J. Fopma, South Dakota State University; Brian D.S. Graeb, South Dakota State University; Tammy Wilson, National Park Service

ABSTRACT: Described as an “island on the prairie” the Black Hills are a small range of mountains arising from a sea of short and mid-grass prairies. Upwelling from the center of the hills, flowing outward are numerous, cold-water streams lacking connectivity to regional cold-water networks. Many species that inhabit local streams are subsequently isolated from conspecific populations, posing unique management challenges. Climate change, predicted to alter local climatic patterns (intensified wet and dry cycles, general warming), potentially further limits species distributions within the island. Mountain Sucker (Catostomus platyrhynchus) is listed as a management indicator species for the Black Hills of South Dakota by South Dakota’s department of Game Fish and Parks, and is used as a proxy for regional ecosystem health. Surveys conducted between 2008 and 2010 revealed that populations had been in decline in both distribution and local abundance. Population surveys conducted between 2014 and 2017 were used to generate species distribution models (SDMs) for this regionally imperiled species. Model predictions are expected to be driven by stream permanency and connectivity. Predictions were compared to 2018 empirical observations to assess model accuracy. Accurate models allow managers to more efficiently identify local populations, impacts of climate change and target conservation efforts.

AUTHORS: Christopher Jacques, Western Illinois University; Evan Phillips, Colorado Parks and Wildlife; Angela Jarding, National Park Service; Susan Rupp, Enviroscapes Ecological Consulting, LLC; Robert Klaver, U.S. Geological Survey; Chadwick Lehman, South Dakota Game, Fish and Parks; Jonathan Jenks, South Dakota State University

ABSTRACT: Since 1993, elk (Cervus canadensis nelsoni) abundance in the Black Hills of South Dakota has been estimated using a detection probability model previously developed in Idaho, though are likely negatively biased because of a failure to account for visibility biases under local conditions. To correct for this bias, we evaluated the current detection probability across the Black Hills during January and February 2009-2011 using radiocollared elk. We used logistic regression to evaluate topographic features, habitat characteristics, and group characteristics relative to their influence on detectability of elk. Elk detection probability increased with less vegetation cover (%), increased group size, and snow cover (%); overall detection probability was 0.60 (95% CI = 0.52-0.68) with 91 of 152 elk groups detected. Predictive capability of the selected model was excellent (ROC = 0.807), and prediction accuracy ranged from 70.2% to 73.7%. Cross-validation of the selected model with other population estimation methods resulted in comparable estimates. Application of our model should be applied cautiously if characteristics of the area (e.g., vegetation cover > 50%, snow cover > 90%, group sizes > 16 elk) differ notably from the range of variability in these factors under which the model was developed.

AUTHORS: Chad R. Williamson, Henry Campa III, Scott R. Winterstein – Michigan State University; Alexandra B. Locher, Grand Valley State University; Dean E. Beyer, Jr., Michigan Department of Natural Resources

ABSTRACT: To determine current habitat suitability for elk (Cervus elaphus) in Michigan, we developed a stand-level (fine-scale) habitat suitability index (HSI) model for public lands, and a landscape-level (coarse-scale) HSI model for public and private lands. For our stand-level HSI model, we used forest compartment inventory data to identify cover types important to elk, and assigned suitability values (0=low, 1=high) to each cover type for elk life requisites (spring food, winter food, winter thermal cover). Additionally, we modified suitability values based on stand conditions (e.g., stand size, age of aspen [Populus spp.], % canopy closure). For our landscape-level HSI model, we used satellite imagery to classify cover types and assigned suitability values to cover types for each life requisite. Our HSI models indicate a heterogeneous arrangement of high suitability for spring food (openings, aspen) and winter food (aspen, hardwoods, conifers) throughout our study area, and several large areas of high suitability for winter thermal cover (conifers) in the southern edge of our study area. Our landscape-level model provided suitability for private lands, but overestimated areas of high suitability in comparison to our stand-level model. Habitat potential was modeled by delineating habitat types by overlaying digital spatial data layers (soils, land-type associations, vegetation) and identifying successional trajectories using habitat classification guides and literature. We assigned suitability values to each habitat type for life requisites at early to late successional stages. Comparisons between current elk habitat suitability and habitat potential identify key areas where managers can maximize management efforts for elk in Michigan. Areas determined to have higher habitat potential may become focus areas if they are not currently being managed or have low suitability. Conversely, areas with low habitat potential may be avoided for continued or future elk habitat management.

AUTHORS: David L. Smith, Aaron Cupp, Christa Woodley, Aaron Urbanczyk – U.S. Army Engineer R&D Center

ABSTRACT: In the United States the Asian carps threaten the Great Lakes via the Illinois Waterway.  The United States Army Corps of Engineers and partners are investigating the use of carbon dioxide   based deterrence barrier.  Carbon dioxide acts as an anesthetic that leads to immobilization and death in fishes.  We have been developing carbon dioxide response rules for Bighead carp  (Hypophthalmichthys nobilis) in a laboratory.  We are applying those rules in a hydrochemodynamic numerical model representing a carbon dioxide barrier at Brandon Road Lock and Dam, a component of the Illinois Waterway.  We explore fish response to the carbon dioxide barrier using a fish movement model.  In the model we implemented rules driven by variable water velocities and carbon dioxide concentrations and produced a movement track.  We measured fish numbers that that would 1) leave the barrier in the downstream direction, 2) become immobilized, and 3) successfully pass the barrier.   The results suggest that a carbon dioxide barrier is another fish deterrence technology that has applicability in the management of invasive species.  However, additional research and development is required to better understand fish response to carbon dioxide gradients and cost of deploying an operational barrier. 

AUTHORS: Joel G. Putnam, Diane Waller, Justine Nelson– US Geological Survey Upper Midwest Environmental Sciences Center; Tammy J. Clark, Viterbo University

ABSTRACT: The search for new chemical controls for aquatic invasive species (AIS) that are efficacious and selective is needed to expand the arsenal of AIS control tools for resource managers. Chemical control options for dreissenid mussels (Dreissena polymorpha and D. bugensis) currently rely heavily on molluscicides that can be costly and/or harmful to nontarget species. The Environmental Protection Agency ECOTOX Knowledgebase was used to gather toxicity data for over 400 taxa covering five kingdoms and 7700 chemicals. Our search used structural activity relationships (SARs) to correlate chemical information with biological activity and predict new chemicals that are effective against dreissenid mussels. A database of chemical descriptors, such as molecular weight, solubility, and polar surface area, was created and published to link the chemical structure/information with species-specific toxicity. Toxicity trials have been initiated using a category of chemicals with high selective toxicity towards dreissenid mussels. Chemicals that produced significant mortality of dreissenid mussels were also tested on nontarget native freshwater mussels to determine selectivity. The results of toxicity trial will be combined with chemical characteristics (e.g., solubility) to identify toxicants that may be suitable for incorporation into a microparticle that is ingested by dreissenid mussels.

AUTHORS: R. Andrew Goodwin, U.S. Army Engineer R&D Center

ABSTRACT: Telemetry projects are often burdened with the need to convert measured fish movement patterns into actionable management guidance that can be quickly used for improving the engineering design of waterways infrastructure, water quality, and/or flowrate. With increasing attention on non-salmonids, one can assess in hindsight the methods that led to significant new insight of juvenile Pacific salmon and how these methods are presently being used to understand fish movement in the Midwest. I explain how interpretations of fish behavior through the lens of ELAM modeling analyses has changed over the past 20 years and how the ELAM is presently being applied to understand species in the Midwest. The ELAM model is a numerical analysis providing an independent viewpoint on fish movement behavior, uniquely separate from traditional statistical insights, which can serve as one of the pillars for informing future design and management of waterways and infrastructure. The ELAM model uses a non-trivial process for converting fish telemetry data into a mechanistic explanation for how/why animal movement patterns emerged the way they did. The ELAM method provides one of the strongest means for forecasting plausible fish response to future design and management actions, recognizing that all methods for predicting behavior are imperfect. The ELAM model is a cost-effective means for vetting future designs and management actions using fish telemetry as input in the early stages of water resource alternatives formulation.

AUTHORS: Gregory J. Soulliere, Mohammed A. Al-Saffar – U.S. Fish and Wildlife Service

ABSTRACT: Targeting conservation to achieve biological objectives for waterfowl and social objectives for people is an emerging priority for bird conservation Joint Ventures implementing the North American Waterfowl Management Plan (NAWMP).  To help achieve NAWMP goals in the Upper Mississippi / Great Lakes Joint Venture (JV) region, we integrated objectives related to waterfowl population demography, conservation supporters (hunters and birders), and ecological goods and services important to society and developed a Decision Support Tool (DST).  Starting with a table of contemporary conservation issues, we transformed related biological and social data into a family of six spatially explicit model-based maps designed to achieve individual objectives.  Output maps were weighted based on discussion with regional decision makers (i.e., the JV Management Board) and then combined, resulting in an aggregate DST to target conservation for waterfowl and people in the JV region.  The tool was designed to be flexible and adaptable; objectives and objective weights may be adjusted and subsequent output maps customized depending on stakeholder priorities.  Current JV objectives to retain and restore high value waterfowl habitats, while enhancing hunting and birding opportunity and addressing watershed impairments, resulted in a spatially explicit DST with solid emphasis in the lower half of the Great Lakes region.

AUTHORS: Tim Lyons, University of Nebraska-Lincoln; T.J. Benson, Illinois Natural History Survey; Wade Louis, Illinois Department of Natural Resources; Mike Ward, University of Illinois at Urbana-Champaign; Richard Warner, National Great Rivers Research & Education Center

ABSTRACT: In agriculturally dominated landscapes, the habitat provided by public and private lands is critical for the conservation and management for non-game as well as game species, such as ring-necked pheasants. Management of these areas to increase pheasant populations has focused on increasing field size, the amount of grassland cover in the landscape, or managing vegetation composition within fields, to improve success during the nesting or brood-rearing stages, or the survival of breeding adults. How these actions will impact overall population growth or which stages or habitat features should be prioritized for management is not always clear. We studied how habitat conditions at the field-and landscape-scale influenced the demography of ring-necked pheasants on public and private grasslands in Illinois. Between 2013-2016, we used radio telemetry to track > 200 ring-necked pheasants and quantified the relationship between habitat features at multiple spatial scales, nest success, chick survival, and adult survival. We then used a simulation study to understand how changes to habitat features important to a particular stage ultimately affected population growth. We also examined how predator identity influenced the relationship between adult survival and habitat conditions. We found that several habitat features had contrasting effects among multiple stages and ultimately restricted population growth when management focused on maximizing performance during one stage. Our results also indicate that raptors may be a more important predator of pheasants than is generally recognized, but the risk of predation can be reduced by the management of vegetation within fields. Collectively our work highlights the importance of full life-cycle studies of demography for the effective management of wildlife and suggests that smaller fields, often overlooked in traditional conservation schemes, can play a role in pheasant management when coupled with appropriate management of vegetation within fields.

AUTHORS: Chris Pollentier, Wisconsin Department of Natural Resources; Mike Hardy, University of Wisconsin-Madison; Scott Lutz, University of Wisconsin-Madison; Scott Hull, Wisconsin Department of Natural Resources

ABSTRACT: Eastern wild turkeys (Meleagris gallopavo silvestris) were successfully reintroduced in Wisconsin during the mid-1970s and populations have since expanded beyond their ancestral range throughout the state. Abundance has generally been considered greatest in areas with highly diverse landscapes that include upland woodlands interspersed with agriculture and other open-herbaceous land cover. However, many areas across far northern Wisconsin are comprised of landscapes where the forested area represents > 70% of the land cover. While much research has been focused on areas where populations are generally highest, study of wild turkeys across the far northern reaches of their range in the Upper Midwest and northern Wisconsin has been limited. To better understand wild turkey distribution and habitat relationships across northern Wisconsin, we conducted gobbling call-count surveys along 157 routes from 2013–2017 and instituted a multiseason correlated replicate occupancy modeling approach to link landscape characteristics to patch occupancy. Probability of occupancy was best related to a quadratic function of percentage of open cover (ß = -4.10, SE = 1.07), with probability of occupancy peaking in routes with 30–40% open cover. Probability of colonization was positively associated with the percentage of available agriculture planted in corn (ß = 1.14, SE = 0.42), and also showed a weak negative association with the amount of snow cover (ß = -1.13, SE = 0.62). Our results suggest that even in landscapes where forest cover is pervasive, wild turkeys benefit from the availability of open-herbaceous cover. In addition, corn-crop agriculture serves as an important food resource for wild turkey populations across heavily-forested northern Wisconsin landscapes and influences the probability of colonization into previously unoccupied areas. A better understanding of the distribution of wild turkeys across their northern range will provide much needed information to help guide contemporary management strategies in a post-restoration era.

AUTHORS: Meaghan Gass, Michigan State University Extension and Michigan Sea Grant; Erica Rogers, Michigan State University Extension

ABSTRACT: The Michigan EnviroImpact Tool is a decision support tool for short-term manure application planning that shows daily runoff risk across Michigan. Nutrient runoff from manure application is just one source of harmful algal blooms, but with proper planning, farmers can help keep applied manure nutrients on their fields and reduce nutrient runoff from entering the Great Lakes. The runoff risk forecast is derived from real-time National Weather Service hydrological models. These models rely on precipitation and temperature forecasts as well as simulated snow melt, soil moisture and temperature, and other landscape characteristics. Tool developers used regional testing from edge of field monitoring sites to validate the prediction models. The Michigan EnviroImpact Tool was developed in partnership with the National Ocean and Atmospheric Administration National Weather Service, the Michigan Department of Agriculture and Rural Development, the Michigan Agriculture Environmental Assurance Program, the Michigan State University Institute of Water Research, Michigan Sea Grant and Michigan State University Extension. This tool is part of a regional effort to improve runoff risk decision support tools in the Great Lakes basin supported by the Environmental Protection Agency, Great Lakes Restoration Initiative (GLRI), and National Weather Service North Central River Forecast Center (NCRFC). In this presentation, we will outline the Michigan EnviroImpact Tool’s modeling system, agricultural best management practices for manure nutrient application, the social marketing campaign to Michigan farmers, and lessons learned from stakeholder engagement strategies.

AUTHORS: Tim Lyons, Larkin Powell – University of Nebraska-Lincoln; Mark Vrtiska, Nebraska Game and Parks Commission

ABSTRACT: Bird banding is a basic but important form of population monitoring that is vital to waterfowl management in the U.S. Most often, the data derived from state and federal banding programs are used to estimate survival and harvest rates and distributions across broad geographic regions, but they can also be used to estimate abundance at more local scales. However, absent or limited population surveys, banding efforts, or recoveries, from neighboring states or provinces complicates efforts, leading to ad hoc approaches to deal with these problems during analysis. Here, we use banding and recovery data of Canada geese in Nebraska 2006-2017 to assess the effects of changing harvest regulations on Canada goose demography and estimate abundance. We demonstrate a modified harvest derivation analysis and Lincoln-Peterson estimator to address pitfalls common to these approaches when estimating abundance. Finally, we discuss Jolly-Seber models as an alternative approach to estimating demographic parameters and abundance that circumvents the need for banding or population monitoring programs out-of-state and provides greater detail about the mechanisms responsible for population changes.