Demographic Estimates from a Muskrat Population in the Flint Hills Caleb M. Bomske1, Matthew S. Peek2, and Adam A. Ahlers1* Abstract - Apparent declines in Ondatra zibethicus Linnaeus (Muskrat) populations across North America necessitate information regarding key demographic parameters. In the Flint Hills ecoregion in Kansas, USA, Muskrat habitat is generally characterized by man-made wetlands used for ranching operations. The relative quality of these habitats for Muskrat populations is poorly understood. We marked and recaptured Muskrats (2020–2022) in the Flint Hills to quantify daily rates in apparent survival. Trap success was low, resulting in a small sample size for inference (n = 22). Our most-supported Cormack-Jolly-Seber model suggested daily recapture probabilities were greater in 2021 (p = 0.67) than in 2022 (p = 0.36). Daily apparent survival rates (Φ = 0.92, 95% CI = 0.75–0.98) were lower than reported in other studies, suggesting poorer habitat conditions for Muskrats in the Flint Hills. We did not capture or detect kits during our study, possibly revealing drought-induced Allee effects in a low-density Muskrat population. Introduction Ondatra zibethicus Linnaeus (Muskrat) are small semiaquatic mammals (0.7–1.8kg; Willner et al. 1980) that are ecologically (Bomske and Ahlers 2021), culturally (Straka et al. 2018, Turner et al. 2018), and economically important species (Ahlers et al. 2016). Although widely distributed across North America, recent evidence suggests Muskrat populations have declined across much of their native range (Ahlers and Heske 2017, Greggory et al. 2019, Roberts and Crimmins 2010, Sadowski and Bowman 2021). A mechanistic explanation for observed population declines remains unknown. However, several hypotheses are plausible. For example, wetland loss, wetland isolation, and changing patterns in hydrology may be affecting key population demographic parameters such as survival and population growth (Ahlers and Heske 2017). Muskrat survival and population growth are affected by both biological and environmental factors. Extreme fluctuations in hydrology, such as drought and flooding, can negatively affect Muskrat populations (Ahlers et al. 2010b, Errington 1939, 1963). Climate change is increasing the frequency and duration of extreme precipitation and drought events (Dai 2013, Easterling et al. 2000), likely posing increased risks for Muskrat populations (Ahlers et al. 2015). Receding water levels during drought conditions can expose burrows to predators, resulting in greater adult and kit mortality rates (Errington 1963). Predation of Muskrats within burrows or lodges has generally been attributed to Neogale vison Schreber (American Mink; Errington 1943, 1963), though Canis latrans Say (Coyote) predation is also a significant cause of mortality (Ahlers et al. 2010b). Muskrats occurring in drought-affected wetlands are reluctant to move into upland areas and search for other good-quality habitats (Errington 1939). Space use by Muskrats is largely restricted to within-water areas (Ahlers et al. 2010a, Ganoe et al. 2021, Matykiewicz et al. 2021), and individuals using terrestrial areas can incur greater mortality risks (Ahlers et al. 2015). Muskrat population abundances are dynamic (Ahlers et al. 2021, Erb et al. 2000), though little is known about survival rates or population growth. Wetlands occurring in the Flint Hills, Kansas, USA, are largely man-made and designed for water retention to support Bos taurus Linnaeus (Cattle) ranching operations. Historically, wetlands large enough to support Muskrat populations in the Flint Hills ecoregion were relatively rare outside of naturally occurring riparian areas. Constructed wetlands now support wetland-obligate biodiversity in the Flint Hills (Swartz and Miller 2021), including semiaquatic mammals such as Muskrat. Relative habitat quality of these wetlands and survival rates of Muskrat occurring in them are unknown. We conducted a mark-recapture study of a Muskrat population occurring in isolated wetlands in the Flint Hills from 2020–2022 to quantify apparent survival estimates. Our goal was to estimate daily Muskrat apparent survival rates to provide insight on the relative quality of these habitats to Muskrat populations. Materials and Methods All trapping and handling methods were approved by Kansas State University Institutional Animal Care and Use Committee (Protocol #4074) and were consistent with guidelines provided by the American Society of Mammalogists (Sikes et al. 2016). Our study occurred in the Flint Hills tallgrass prairie ecosystem, near Manhattan, Kansas, USA (39°11′N, −96°35′W). The Flint Hills represent one of the largest remaining tracts of tallgrass prairie landcover in North America and are characterized by shallow, rocky soils overlaying layers of limestone and flint. Our study area (Fig. 1) ranged from 300–500 m in elevation, and landcover was dominated by native tallgrass prairie (~45%). However, landscape change is prevalent, and much of the landcover now includes agriculture (21%) and woody encroachment (7%; Hamilton et al. 2000, Ratajczak et al. 2012). During our study, monthly precipitation ranged from 20–184 cm (mean 105 cm) and the temperature ranged from -14°–37° C (mean 23° C). All wetlands included in our study were human-made retention ponds constructed for watering cattle (average depth = 3.3 m, range = 0.5–10 m). Emergent plant species included Cattail (Typha sp.), Spikerush (Eleocharis sp.), Milfoil (Myriophyllum sp.), Pondweed (Potamogeton sp.), Sedges (Cyperus sp.), Arrowhead (Sagittaria sp.), Knotweed (Polygonum sp.), Bullrushes (Schoenoplectus sp.), and Duckweed (Lemna sp.). Woody encroachment was also prevalent and characterized by Juniperus virginiana Linnaeus (Eastern Redcedar), Rhus glabra Linnaeus (Smooth Sumac), and Cornus drummondii Mey (Roughleaf Dogwood). Potential Muskrat predators in our study area included American Mink, Buteo jamaicensis Gmelin (Red-tailed Hawk), Coyotes, and Bubo virginianus Gmelin (Great Horned Owl), among others. Landowners confirmed that Muskrat trapping was not occurring on their properties during the duration of our study and that Muskrat trapping was rare at their wetlands after 2010. Average spatial separation from sampled wetlands was 7.74 km (range: 1.63–23.89 km), and average wetland area was 5,234 m2 (range: 278–25,424 m2). Captures and marking We live-trapped Muskrats using single-door, collapsible traps (Tomahawk Live Trap Inc., Hazelhurst, WI; Model 202) at wetlands throughout our study area. We chose wetlands based on probability of Muskrat occupancy the previous fall (information we collected during a concurrent study) and likelihood of Muskrat occurrence at the time of trapping (determined by presence of Muskrat signs at that time). We attempted to target wetlands with the greatest likelihood of Muskrat capture, as our preliminary work suggested Muskrat population abundances were depressed in the region following severe drought. We conducted trapping from 2 November–11 December 2020, 1 March–9 April 2021, and 14 March–10 June 2022, dependent upon temperature and precipitation. We did not trap when daily low temperatures were forecasted ≤ 0°C or during rain events. We fastened live-traps to floating rafts (modified track boards; Larreur et al. 2020, Matykiewicz et al. 2021, Reynolds et al. 2004, Schooley et al. 2012) kept buoyant by Styrofoam sandwiched between two plywood sheets (122 × 61 × 4 cm). We floated trap boards along wetland edges near active Muskrat burrows or in areas with Muskrat signs (e.g., Muskrats, tracks, scat, clippings) and tethered them to sturdy vegetation or structure (e.g., rocks, Eastern Redcedar trees or stumps, Cattails). We covered traps with local vegetation to conceal them and provide cover from adverse weather and direct sunlight (Matykiewicz et al. 2021). We baited traps with apple (e.g., Ahlers et al. 2010a) and checked twice daily, once in the morning and again in the evening. All unmarked individuals were immediately transferred to a handling bag and marked with a 10-mm passive integrated transponder (PIT) tag (Biomark, Boise, Idaho) inserted subcutaneously between the shoulders. We recorded unique identity of all recaptured individuals using a handheld PIT tag reader (Biomark, Boise, Idaho) passed along the outside of the trap or handling bag. We measured weight (g) of all captured Muskrats (average = 1.25 kg, range = 0.82–1.93 kg) and determined sex, though we could not identify sex with complete certainty in all individuals (see Results). All Muskrat handling was done at the wetland edge without use of anesthesia (< 5 min per individual), and marked and recaptured individuals were released at the point of capture. Survival analyses We used Cormack-Jolly-Seber models (Program MARK ver. 9.0; White and Burnham 1999) to estimate apparent daily survival rates (Φ) of Muskrats. Apparent survival rates were estimated over 10 days each year. We considered single and additive effects of ‘Sex’ (male = 1, female = 0) and ‘Year’ (2022 = 0, 2021 = 1) on recapture probabilities (p) resulting in four models in this candidate set (Sex, Year, Sex + Year, Null). Because of our small sample size, we did not consider time-dependent effects in either Φ or p, nor did we consider models including interactions. We also did not model covariate effects on Φ as not to overfit models. We ranked models using Akaike’s Information Criterion (AIC), corrected for small sample sizes (AICc), and assessed model-support using an information-theoretic approach (Arnold 2010, Burnham and Anderson 2002). We considered models with ΔAICc < 2.00 as competitive. Results From 2020–2022, we attempted to live trap Muskrats at 23 wetlands (2020 = 11, 2021 = 12, 2022 = 14) resulting in 1,766 trap nights (2020 = 290, 2021 = 640, 2022 = 836). Overall trap success was 0.03/trap night but varied by year (2020 = 0.003, 2021 = 0.04, 2022 = 0.02). We caught and individually marked 22 Muskrats across all years (2020 = 0, 2021 = 14, 2022 = 8). A single Muskrat caught in 2020 escaped the handling bag prior to being marked. Sex could not be determined with complete confidence for 10 individuals, thus our sample included 8 males and 4 females with remaining individuals’ sex unknown. All captured Muskrats were adults and we did not capture or detect kits at wetlands. Average recaptures per individual was 0.86 (range = 0–5) and varied between years (2020 = 0; 2021 = 0.86, range = 0–5; 2022 = 0.88, range = 0–2). We did not recapture any individual Muskrats in multiple years, including at wetlands where we marked Muskrats in previous years. Our top model indicated Muskrat recapture probabilities varied by year (β = 1.12, SE = 0.70; 2021 = 0.67, 2022 = 0.36; Table 1). Our null model was also competitive, though it had poorer model fit (Table 1). After accounting for differences in recapture probabilities between years, daily apparent survival rates were 0.92 (95% CI = 0.75–0.98). Discussion Our study provides information on survival rates of Muskrats in the Flint Hills ecoregion—a key demographic that was previously unknown. Our estimated daily apparent survival rates, when extrapolated to weekly survival rates (0.56), were lower than those reported by other studies (e.g., Ahlers et al. 2010b, Ganoe et al. 2021, Kanda and Fuller 2004, Matykiewicz et al. 2021). Further, we did not catch or detect signs of Muskrat kits or lactating females at any of our wetlands in 1,766 trap nights across 3 years. This suggests reproductive success is limited or absent, and highlights potential demographic Allee effects (Stephens and Sutherland 1999) in a low-density Muskrat population. If species’ populations fall below an abundance threshold average, fitness of individuals within that population can decrease (Allee 1931). Allee effects may reduce Muskrat population fitness for many reasons, including if Muskrats are dispersing away from already low-density populations, mate limitation at wetlands, or inbreeding depression. Future work should focus on identifying mechanistic explanations for reduced population fitness in low-density Muskrat populations including assessing the potential for Allee effects to impede population growth or recovery. Large areas of the Great Plains in North America experienced significant drought conditions prior to our study (summer and fall 2018). In the Flint Hills ecoregion in Kansas, USA, drought was severe (Palmer drought severity index -3 to -3.99), resulting in substantial wetland drying across the region (e.g., Hopper 2019). Additionally, streams and other riparian areas had reduced or no water flow during this time (e.g., Hopper et al. 2020). These drought conditions could have negatively affected Muskrat populations during our study, though we could not empirically test this hypothesis with our data. Extreme hydrological events are becoming more frequent and intense (Dai 2013, Easterling et al. 2000), with potential to negatively affect Muskrat populations (Ahlers et al. 2015). Subsequent studies assessing resilience of man-made wetlands in the Flint Hills to extreme flooding and drought events are warranted, especially as it relates to conservation of wetland-obligate species occurring in the area. Prior to wetland construction, it is likely Muskrats did not occur in the Flint Hills except in riparian areas or limited naturally occurring wetlands. Based on our results, our sampled wetlands likely only offer marginal-quality habitat to support Muskrat populations. Human-constructed wetlands in the Flint Hills that retain sufficient water levels may provide temporary refugia for Muskrats during drought (e.g., to escape predation) or act as stepping stones between good-quality habitat for dispersing individuals. These hypotheses, however, need to be rigorously evaluated. Unfortunately, we could not assess the importance of local- (e.g., wetland size, amount of vegetation) and landscape-scale (e.g., amount of agriculture or woody encroachment) habitat quality on apparent daily survival rates of Muskrats because of our limited sample size. We did not recapture marked Muskrats between sampling years or at wetlands where they were not initially marked. Studies examining the spatial ecology of Muskrats in the Flint Hills will likely provide insight into factors affecting wetland colonization and extinction of Muskrat populations. Additionally, studies incorporating known-fate survival and population abundance information would likely reveal the relative importance of these wetlands to Muskrat population fitness. Acknowledgements Our research occurred on prairies historically home to Native nations including the Kaw, Osage, and Pawnee. Funding was provided by the Department of Horticulture and Natural Resources at Kansas State University and the Kansas Department of Wildlife and Parks via the U.S. Federal Aid in Wildlife Restoration Fund. We are thankful for technical help provided by M. Zurschmiede and S. Bomske. We also thank A. Ricketts for advice and suggestions on improving this research. We thank private landowners who gave us permission to live-trap Muskrats on their wetlands. Literature Cited Ahlers, A.A., L.A. Cotner, P.J. Wolff, M.A. Mitchell, E.J. Heske, and R.L. Schooley. 2015. Summer precipitation predicts spatial distributions of semiaquatic mammals. PLoS ONE 10: e0135036. Ahlers, A.A., and E.J. Heske. 2017. Empirical evidence for declines in Muskrat populations across the United States. Journal of Wildlife Management 81:1408–1416. Ahlers, A.A., E.J. Heske, and C.A. Miller. 2016. Economic influences on trapper participation and per capita harvest of Muskrat. Wildlife Society Bulletin 40:548–553. Ahlers, A.A., E.J. Heske, R.L. Schooley, and M.A. Mitchell. 2010a. Home ranges and space use of Muskrats, Ondatra zibethicus, in restricted linear habitats. Wildlife Biology 16:400–408. Ahlers, A.A., E.J. Heske, R.L. Schooley, and M.A. Mitchell. 2010b. Effects of flooding and riparian buffers on survival of Muskrats (Ondatra zibethicus) across a flashiness gradient. Canadian Journal of Zoology 88:1011–1020. Ahlers, A.A., T.P. Lyons, and E.J. Heske. 2021. Population dynamics of Muskrat (Ondatra zibethicus) and American Mink (Neovison vison): Investigating contemporary patterns in a classic predator-prey system. Canadian Journal of Mammalogy 99:681–688. Allee, W.C. 1931. Animal aggregations: A study in general sociology. University of Chicago Press, Chicago, IL, USA. 409 pp. Arnold, T.W. 2010. Uninformative parameters and model selection using Akaike’s information criterion. Journal of Wildlife Management 74:1175–1178. Bomske, C.M., and A.A. Ahlers. 2021. How do Muskrats Ondatra zibethicus affect ecosystems? A review of evidence. Mammal Review 51:40–50. Burnham, K.P., and D.R. Anderson. 2002. Model Selection and Multimodel Inference: A practical information-theoretic approach. 2nd ed. Springer, New York, NY, USA. 490 pp. Dai, A. 2013. Increasing drought under global warming in observations and models. Nature Climate Change 2:52–58. Easterling, D.R., G.A. Meehl, C. Parmesan, S.A. Changnon, T.R. Karl, and L.O. Mearns. 2000. Climate extremes: Observations, modeling, and impacts. Science 289:2068–2074. Erb, J., N.C. Stenseth, and M.S. Boyce. 2000. Geographic variation in population cycles of Canadian Muskrats (Ondatra zibethicus). Canadian Journal of Zoology 78:1009–1016. Errington, P.L. 1939. Reaction of Muskrat populations to drought. Ecology 20:168–186. Errington, P.L. 1943. An analysis of Mink predation upon Muskrats in north-central United States. Iowa Agricultural Experiment Station Research Bulletin 320:797–924. Errington, P.L. 1963. Muskrat Populations. Iowa State University Press, Ames, IA, USA. 665 pp. Ganoe, L.S., M.J. Lovallo, J.D. Brown, and W.D. Walter. 2021. Ecology of an isolated Muskrat population during regional population declines. Northeastern Naturalist 28:49–64. Greggory, G.J., R. Dibblee, P.Y. Daoust, and M.R. van den Heuvel. 2019. Declines in Muskrat Ondatra zibethicus population density in Prince Edward Island, Canada. Canadian Wildlife Biology and Management 8:46–60. Hamilton, B., W. Ostlie, G. Wingfield, J. Haferman, and D. Ladd. 2000. Ecoregional conservation in the Osage Plains/Flint Hills prairie. The Nature Conservancy, Minneapolis, MN, USA. 48 pp. Hopper, G.W. 2019. An observation of aquatic surface respiration by Central Stoneroller (Campostoma anomalum) in isolated pools during severe drought. Transactions of the Kansas Academy of Science 122:3–4. Hopper, G.W., K.B. Gido, C.A. Pennock, S.C. Hedden, B.D. Frenette, N. Barts, C.K. Hedden, and L.A. Bruckerhoff. 2020. Nowhere to swim: Interspecific responses of prairie stream fishes in isolated pools during severe drought. Aquatic Sciences 82:42. Kanda, L.L., and T.K. Fuller. 2004. Demographic responses of Virginia Opossums to limitation at their northern boundary. Canadian Journal of Zoology 82:1126–1134. Larreur, M., S.K. Windels, B.T. Olson, and A.A. Ahlers. 2020. Cross-scale interactions and non-native Cattails affect the spatial distribution of a wetland-obligate species. Landscape Ecology 35:59–68. Matykiewicz, B.R., S.K. Windels, B.T. Olson, R.T. Plumb, T.M. Wolf, and A.A. Ahlers. 2021. Assessing translocation effects on the spatial ecology and survival of Muskrats Ondatra zibethicus. Wildlife Biology 2:1–11. Ratajczak, Z., J.B. Nippert, and S.L. Collins. 2012. Woody encroachment decreases diversity across North American grasslands and savannas. Ecology 93:697–703. Reynolds, J.C., M.J. Short, and R.J. Leigh. 2004. Development of population control strategies for Mink Mustela vison, using floating rafts as monitors and trap sites. Biological Conservation 120:533–543. Roberts, N.M., and S.M. Crimmins. 2010. Do trends in Muskrat harvest indicate widespread population declines? Northeastern Naturalist 17:229–238. Sadowski, C., and J. Bowman. 2021. Historical surveys reveal a long-term decline in Muskrat populations. Ecology and Evolution 11:7557–7568. Schooley, R.L., L.A. Cotner, A.A. Ahlers, E.J. Heske, and J.M. Levengood. 2012. Monitoring site occupancy for American Mink in its native range. Journal of Wildlife Management 76:824–831. Sikes, R.S., and The Animal Care and Use Committee of the American Society of Mammalogists. 2016. Guidelines of the American Society of Mammalogists for the use of wild mammals in research and education. Journal of Mammalogy 97:663–68. Stephens, P.A., and W.J. Sutherland. 1999. Consequences of the Allee effect for behavior, ecology, and conservation. Trends in Ecology and Evolution 14:401–405. Straka, J.R., A. Antoine, R. Bruno, D. Campbell, R. Campbell, R. Campbell, J. Cardinal, G. Gibot, Q.Z. Gray, S. Irwin, R. Kindopp, W. Ladauceur, J. Lankshear, B. Maclean, S. Macmillan, F. Marcel, G. Marten, L. Marten, J. McKinnon, L.D. Patterson, C. Voyageur, M. Voyageur, G. Whiteknife, and L. Wiltzen. 2018. “We used to say rats fell from the sky after a flood”: Temporary recovery of Muskrat following ice jams in the Peace-Athabasca Delta. Arctic 71:218–228. Swartz, T.M., and J.R. Miller. 2021. The American Pond Belt: An untold story of conservation challenges and opportunities. Frontiers in Ecology and the Environment 19:501–509. Turner, C.K., T.C. Lantz, and Gwich’in Tribal Council Department of Cultural Heritage. 2018. Springtime in the Delta: The socio-cultural importance of Muskrats to Gwich’in and Inuvialuit trappers through periods of ecological and socioeconomic change. Human Ecology 46:601–611. White, G.C., and K.P. Burnham. 1999. Program MARK: Survival estimation from populations of marked animals. Bird Study 46:S120–S139. Willner, G.R., G.A. Feldhamer, E.E. Zueker, and J.A. Chapman. 1980. Ondatra zibethicus. Mammal Species 141:1–8.