2010 NORTHEASTERN NATURALIST 17(2):229–238 Do Trends in Muskrat Harvest Indicate Widespread Population Declines? Nathan M. Roberts1,* and Shawn M. Crimmins2 Abstract - Ondatra zibethicus (Muskrat) is one of the most widely distributed furbearers in eastern North America. Anecdotal evidence suggests that Muskrats are experiencing a regional decline in numbers, although little empirical evidence exists to support this claim. Our objectives were to document temporal trends in Muskrat harvest in eastern North America, and to use the relationship between harvest and pelt price to infer potential trends in regional Muskrat populations. Muskrat harvest has declined by approximately 75% since 1986 in eastern North America, despite a recent resurgence in pelt prices. Recent harvest rates showed little correlation (r2 = 0.355–0.559) with current or time-lagged pelt prices, despite large correlations (r2 = 0.785–0.823) between pelt price and harvest from historic data (1948–1968). These results suggest that, at low harvest levels, there is only a weak correlation between harvest and pelt price. These results may be indicative of regional declines in Muskrat abundance, although future research is needed to substantiate this hypothesis. Introduction Ondatra zibethicus L. (Muskrat) is one of the most widely distributed and heavily harvested furbearers in North America (Boutin and Birkenholz 1987, Errington 1963). Muskrats occupy almost every type of freshwater aquatic habitat in eastern North America and are often the dominant herbivore in these systems (Erb and Perry 2003). As such, Muskrats are critical to the structuring of marshland vegetation and chemical cycles (Connors et al. 2000, Weller 1981, Weller and Fredrickson 1973) and can often completely restructure vegetative communities (Danell 1979, Smirnov and Tretyakov 1998). The influence of Muskrats on vegetative structure can affect invertebrate communities (de Szalay and Cassidy 2001, Nelson and Kadlec 1984, Nummi et al. 2006) as well as bird abundance (Kaminski and Prince 1981) and diversity (Bishop et al. 1979). Muskrats can also serve as indicators of ecosystem health by responding to various toxins and chemicals that commonly degrade aquatic habitats (Erickson and Lindzey 1983, Halbrook et al. 1993, Stevens et al. 1997). Furthermore, Muskrat predation can reduce the abundance of exotic species such as Dreissena polymorpha Pallas (Zebra Mussel) (Sietman et al. 2003). As wetland loss becomes more prevalent (Dahl 2000), maintenance of the remaining functioning wetland and aquatic systems is of great importance to managers. As such, conserving viable Muskrat populations may be of critical importance. 1Department of Natural Resources, 207 Fernow Hall, Cornell University, Ithaca, NY 14853. 2Department of Forest Management, 207 Forestry Building, University of Montana, Missoula, MT 59812. *Corresponding author - nmr25@cornell.edu. 230 Northeastern Naturalist Vol. 17, No. 2 Muskrats are used in the wild fur trade and are managed as a game or furbearer species in most jurisdictions (Erb and Perry 2003). As recently as 1992, fur trappers in the northeastern United States considered Muskrats the most important of local furbearing species (International Association of Fish and Wildlife Agencies 2005). Trappers have claimed that Muskrat populations and distributions have been declining for over a decade. Numerous anecdotal sources suggest that Muskrat densities have declined at sites once inhabited by thriving populations and that little population expansion has occurred into previously uninhabited areas, despite their ability to rapidly colonize previously uninhabited areas (Hansen 1965, Matis et al. 1996, Wood 1974). Similarly, wildlife managers have recorded substantial declines in Muskrat harvests throughout the northeast United States and eastern Canada (Northeast Fur Resources Technical Committee 2005). However, most survey techniques for Muskrats are only feasible across small geographic extents (e.g., Schooley and Branch 2005) and may not be useful for examining large-scale population trends. Fur harvest dynamics and the value of pelts are major factors influencing fur harvest effort (Erickson 1981, Erickson and Sampson 1978). A positive relationship between fur harvest and pelt prices is well documented for species including Castor canadensis Kuhl (Beaver) (Bailey 1981), Martes pennanti Erxleben (Fisher) (Powell 1993), and Lontra canadensis Schreber (River Otter) (Chilelli et al. 1996). However, this relationship has not been examined in Muskrats. If changes in Muskrat harvests are merely an artifact of changing market conditions, it would be logical to assume a correlation between pelt price and harvest. Conversely, if changes in Muskrat harvests are reflective of underlying population fluctuations, we would not expect to see as strong a correlation between harvests and pelt prices. Because of the relationship between pelt price and fur harvest, the use of pelt prices to model furbearer populations and harvest dynamics is common (e.g., Scognamillo and Chamberlain 2006). The objectives of our study were to 1) document Muskrat harvest trends in the northeastern United States and eastern Canada, 2) examine the relationship between Muskrat harvest and pelt price during the past 20 years, and 3) examine historical relationships between Muskrat harvest and pelt price between 1948 and 1968. Materials and Methods We obtained annual Muskrat harvest data from 1986–2006 (hereafter referred to as contemporary data), when available, from three jurisdictions in eastern Canada (New Brunswick, Ontario, and Quebec) and nine jurisdictions in the northeastern United States (Connecticut, Massachusetts, Maine, New Jersey, New Hampshire, Pennsylvania, Rhode Island, Vermont, Virginia, and West Virginia) (Table 1). Harvest data beyond the 1996/1997 season for Massachusetts were excluded from analysis because trapping activity since this season was severely affected by legislative actions. Historical Muskrat harvest data (1948–1968) were acquired from five jurisdictions for 2010 N.M. Roberts and S.M. Crimmins 231 Table 1. Annual Muskrat harvest from 1986/1987–2004/2005 trapping seasons by jurisdiction. Year CT MA NB NJ NH ON PA RI VA VT WV QC 1986 10,893 28,404 22,915 199,056 6115 540,106 440,880 1337 64,579 * 34,397 322,643 1987 11,179 28,656 18,229 184,805 6871 491,067 346,558 1211 63,464 * 34,643 307,396 1988 5024 13,780 27,859 177,402 5809 172,715 229,958 728 * * 13,234 113,335 1989 3033 19,958 12,074 126,807 3746 95,143 141,577 284 13,080 * 6669 81,838 1990 3686 13,519 12,037 66,349 2381 65,965 112,358 473 15,734 2117 4692 59,658 1991 3754 12,517 17,226 72,909 3886 50,564 156,014 442 21,961 3308 11,148 58,078 1992 3226 9474 16,028 78,228 2525 84,344 135,533 461 10,380 2632 7074 45,978 1993 1710 9595 17,396 42,274 2273 87,687 121,657 367 14,832 4442 5661 80,076 1994 4512 11,341 23,075 56,737 4389 83,069 178,145 472 21,353 4647 8419 102,209 1995 3159 7,873 22,306 82,506 2731 65,432 130,442 356 7010 2002 4233 9103 1996 3104 7,062 29,650 74,837 2976 124,877 146,013 417 23,925 9475 9440 11,547 1997 3222 712 39,386 81,351 3980 97,365 216,066 454 20,045 11,218 7474 13,648 1998 2767 1017 28,617 66,732 3517 85,316 148,205 512 3888 7389 2833 10,593 1999 1568 747 22,063 33,185 1714 63,820 94,215 243 9673 4010 1734 7377 2000 2443 667 17,064 70,882 2137 60,054 79,880 275 18,343 4351 2857 7214 2001 3022 917 23,472 49,590 3604 95,574 121,994 283 11,743 4420 5785 13,702 2002 1347 649 19,306 36,402 1453 59,670 75,340 216 7739 3355 4160 10,805 2003 2249 1419 19,741 56,413 1929 57,106 71,368 177 9683 2415 3210 38,539 2004 1966 1063 20,121 39,208 2396 53,553 * 197 8654 4219 2523 43,608 2005 * * * 34,465 * 68,215 70,995 * 8451 3097 * * 2006 * * * 33,747 * 84,107 121,167 * 10,308 5802 * * *Data unavailable. 232 Northeastern Naturalist Vol. 17, No. 2 which complete harvest data was available including New Hampshire, New Jersey, Ontario, Quebec, and Rhode Island (Table 2). These historic data represented the most complete historic harvest records available for this geographic region. Average pelt prices for each season were provided by North American Fur Auctions, formerly Hudson’s Bay Company (Toronto, ON, Canada). North American Fur Auctions is the largest wild fur auction in the world, thus we assumed that prices and price trends realized at this auction reflected the current state of the overall wild fur market. Mean per-pelt prices were adjusted for inflation to represent 2007 Canadian dollars (Statistics Canada) (Figs. 1A, 2A). We examined the relationship between pelt price and harvest by constructing two linear regression models for recent (1986–2006) and historic (1948–1968) Muskrat harvest. Our first model considered harvest as a function of current pelt price, while our second model considered a one-year time lag in pelt price. Previous research indicated that fur harvest can be affected by lagged pelt prices (Chilelli et al. 1996, Scognamillo and Chamberlain 2006). This lag is due to marketing practices that result in the sale of large quantities of fur that establish international market trends prior to the majority of the harvest effort during a given season. Linear regression models were constructed in SAS (SAS Institute, Carey, IN) using the PROC GLM procedure. Annual pelt harvest was treated as the dependent variable and annual mean pelt price was considered the predictor variable. In all models, Table 2. Annual Muskrat harvest from 1948/1949–1968/1969 trapping seasons by jurisdiction. Year NJ NH ON RI QC 1948 * 13,185 742,761 * 231,062 1949 * 11,165 562,587 10,104 220,082 1950 * 12,313 656,388 9828 180,648 1951 * 12,820 741,814 11,795 172,417 1952 * 20,394 838,392 10,932 176,294 1953 * 20,512 780,090 9884 190,457 1954 * 12,641 841,135 7868 122,709 1955 * 11,975 500,111 8569 141,511 1956 * 9430 564,511 9271 123,530 1957 * 8439 446,578 6524 136,205 1958 * 9767 337,986 5234 162,608 1959 * 8431 320,287 7836 206,195 1960 * 9125 304,731 5205 204,011 1961 * 8753 377,888 3627 124,677 1962 * 6103 345,428 5320 118,239 1963 * 6811 497,091 4566 191,626 1964 287,982 9993 251,795 4573 176,489 1965 215,000 6917 390,685 4468 103,835 1966 184,807 7271 359,142 4396 126,310 1967 120,471 8235 369,848 3113 121,376 1968 209,818 9505 539,034 4320 155,945 *Data unavailable. 2010 N.M. Roberts and S.M. Crimmins 233 harvests were treated as nested within each jurisdiction to account for regional variation in harvest. Model fit was assessed by examining the mean squared error (MSE) of the least squared estimates of the model parameters and the associated r² values. We examined for temporal trends in Muskrat harvest independent of pelt price using simple regression analysis. Because we did not have harvest values for each jurisdiction every year we could not use total annual Figure 1. Inflation-adjusted Muskrat pelt prices in 2007 Canadian dollars (A) and harvest totals (B) for historic period (1948–1968). 234 Northeastern Naturalist Vol. 17, No. 2 harvest (see Table 1). Rather, we used total harvest per area (km2) each year only from those jurisdictions for which we had harvest data (Figs 1B, 2B). Thus, our model would not be subject to variability caused by jurisdictions that did not report or have access to total harvest values, or by those that did not allow Muskrat harvest. Data on jurisdiction size were gathered from the National Atlas (National Atlas of the United States 2008) and the Atlas of Canada (NRC 2008). Figure 2. Inflation-adjusted Muskrat pelt prices in 2007 Canadian dollars (A) and harvest totals (B) for modern period (1985–2006). 2010 N.M. Roberts and S.M. Crimmins 235 Results Muskrat harvest decreased substantially during our study, declining by approximately 75% since 1986. Declines occurred primarily between 1986 and 1990, with harvest levels stabilizing after 1990 (Fig. 2B). Harvest varied substantially during the historic period (Fig. 1B). All models of Muskrat harvest were highly significant (P < 0.0001), although predictive accuracy varied greatly (Table 3). Models of historic Muskrat harvests predicted by pelt price exhibited high model fit (r2 = 0.785–0.823), performing better than models of current harvest (r2 = 0.355–0.559). Models incorporating a oneyear time lag did not perform as well as models predicting harvests as a function of current pelt price (Table 3). Discussion As with many furbearers, assessing population trends of Muskrats is challenging (Erb and Perry 2003). The high costs associated with large-scale mark-recapture and population-dynamics studies often preclude their use for monitoring furbearer populations, particularly for species with relatively low monetary value such as Muskrats. Despite their ecological importance, little is known of the population status of Muskrats (Erb and Perry 2003). The correlation between pelt prices and harvest levels has been suggested as a means of monitoring furbearer populations and has been used by previous researchers (Bailey 1981, Butler 1942). However, geographic and speciesspecific variability in such relationships is poorly understood (Chilelli et al. 1996). Despite these factors, it is generally thought that a strong relationship between pelt prices and harvest levels can often be expected. Our comparison of historic and current Muskrat harvests in relation to pelt price demonstrates that, while historically strong, this relationship has declined in recent years. Although other factors can influence trapper behavior (e.g., pelt prices of other species, pelt handling time, access to animals), our findings for Muskrat harvests are unique in that a relationship between pelt value and harvest existed historically but weakened over time. Similar observations have not been reported in other investigations examining pelt value and harvest relationships for other species, suggesting that, even if these other factors did influence the pelt value and harvest relationship, the general trend of pelt value driving harvest effort should still hold given relatively stable population status. Given that this correlation was apparent for Muskrat harvest in our historical data set (1948–1968) but not Table 3. Model parameters for Muskrat harvest in the northeastern US and eastern Canada. Model r2 CV Mean F P-value Current Muskrat harvest 0.559 126.4 41,692 21.58 <0.0001 Current Muskrat harvest with time lag 0.355 152.9 41,692 9.35 <0.0001 Historic Muskrat harvest 0.823 54.5 176,521 76.25 <0.0001 Historic Muskrat harvest with time lag 0.785 60.1 176,521 59.77 <0.0001 236 Northeastern Naturalist Vol. 17, No. 2 in our contemporary data set (1986–2006), anecdotal reports of declining Muskrat populations in the northeastern United States and Eastern Canada may be valid. Although the relationship between pelt price and harvest has been used for other furbearers (Scognamillo and Chamberlain 2006), this relationship has not previously been examined in Muskrats. Our models are, to our knowledge, the only attempt to describe Muskrat population status over such a large geographic extent. However, the relatively poor fit of models for contemporary harvest suggests that this relationship may break down at low harvest levels. As such, caution should be exercised if pelt prices are used to predict harvest or population trends, especially when harvest levels are low. It is impossible to know if the observed breakdown in the pelt price-harvest relationship is the result of population declines or simply poorly performing models. Changes in trapping effort or trapper demographics could influence these results; however, it seems reasonable that the effect would be seen for other furbearers, not just Muskrat harvests. This breakdown further necessitates the need for additional research on Muskrat populations. It has been suggested that Muskrats experience cyclical population fluctuations (Erb et al. 2000). Indeed, historic harvest levels did appear to exhibit mild periodicity in our data; however, the lack of periodic fluctuations in our modern harvest data provides further support for the hypothesis of widespread population declines. The overall lack of information on regional population trends for Muskrats is concerning, although our research does support the hypothesis that Muskrat populations are declining in eastern North America. However, given the tenuous strength of our supporting evidence, we suggest that further research regarding Muskrat population status and potential decline in eastern North America is justified. Acknowledgments We would like to thank the members of the Northeast Association of Fish and Wildlife Agencies, Northeast Furbearer Resources Technical Working Group for providing harvests figures. We thank North American Fur Auctions for providing valuable information on pelt values. Milo Richmond provided useful insight, and Cornell University provided essential logistical assistance for this study. Several anonymous reviewers and guest editor John J. Daigle provided helpful comments and suggestions on an earlier draft of this manuscript. Literature Cited Bailey, T.N. 1981. Factors influencing furbearer populations and harvest on the Kenai National Moose Range, Alaska. Pp. 249–272, In J.A. Chapman and D. Pursley (Eds.). Proceedings of the Worldwide Furbearer Conference. International Fur Trade Federation, Maryland Furtrappers Association. Frostburg, MD. Bishop, R.A., R.D. Andrews, and R.J. Bridges. 1979. Marsh management and its relationship to vegetation, waterfowl, and Muskrats. Proceedings of the Iowa Academy of Science 86:50–56. 2010 N.M. Roberts and S.M. Crimmins 237 Boutin, S., and D.E. Birkenholz. 1987. Muskrat and Round-tailed Muskrat. Pp. 315–325, In M. Novak, J.A. Baker, M.E. Obbard, and B. Malloch (Eds.). Wild Furbearer Management and Conservation in North America. Ontario Ministry of Natural Resources. Toronto, ON, Canada. Butler, L. 1942. Fur cycles and conservation. Transactions of the North American Wildlife Conference 7:463–472. Chilelli, M., B. Griffith, and D.J. Harrison. 1996. Interstate comparisons of River Otter harvest data. Wildlife Society Bulletin 24:238–246. Connors, L.M., E. Kiviat, P.M. Groffman, and R.S. Ostfield. 2000. Muskrat (Ondatra zibethicus) disturbance to vegetation and potential net nitrogen mineralization and nitrification rates in a freshwater tidal marsh. American Midland Naturalist 143:53–63. Dahl, T.E. 2000. Status and trends of wetlands in the conterminous United States 1986 to 1997. US Department of the Interior, Fish and Wildlife Service, Washington, DC. Danell, K. 1979. Reduction of aquatic vegetation following the colonization of a northern Swedish lake by the Muskrat, Ondatra zibethica. Oecologica 38:1432–1439. de Szalay, F.A., and W. Cassidy. 2001. Effects of Muskrat (Ondatra zibethicus) lodge construction on invertebrate communities in a Great Lakes coastal wetland. American Midland Naturalist 146:300–310. Erb, J., and H.R. Perry, Jr. 2003. Muskrats. Pp. 311–348, In G.A. Feldhamer, B.C. Thompson, and J.A. Chapman (Eds.). Wild Mammals of North America: Biology, Management, and Conservation. Johns Hopkins University Press, Baltimore, MD. 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. Erickson, D.W. 1981. Furbearer harvest mechanics: An examination of variable influencing fur harvest in Missouri. Pp. 1469–1491, In J.A. Chapman and D. Pursley (Eds.). Proceedings of the Worldwide Furbearer Conference. International Fur Trade Federation, Maryland Furtrappers Association. Frostburg, MD. Erickson, D.W., and J.S. Lindzey. 1983. Lead and cadmium in Muskrat and Cattail tissues. Journal of Wildlife Management 47:550–555. Erickson, D.W., and F.W. Sampson. 1978. Impact of market dynamics on Missouri’s furbearer harvest system. Proceedings of the Southeastern Association of Fish and Wildlife Agencies 32:17–29 Errington, P.L. 1963. Muskrat Populations. Iowa State University Press, Ames, IA. Halbrook, R.S., R.L. Kirkpatrick, P.F. Scanlon, M.R. Vaughn, M.R., and H.P. Veit. 1993. Muskrat populations in Virginia’s Elizabeth River: Physiological condition and accumulation of environmental contaminants. Archives of Environmental Contamination and Toxicology 25:438–445. Hansen, E.L. 1965. Muskrat distribution in south-central Oregon. Journal of Mammalogy 45:669–671. International Association of Fish and Wildlife Agencies. 2005. Ownership and the use of traps by trappers in the United States in 2004. Responsive Management. Harrisonburg, VA. Kaminski, R.A., and H.H. Prince. 1981. Dabbling duck and aquatic macroinvertebrate responses to manipulated wetland habitat. Journal of Wildlife Management 45:1–15. 238 Northeastern Naturalist Vol. 17, No. 2 Matis, J.H., T.R. Kiffe, and R. Hengevekd. 1996. Estimating parameters for birthdeath- migration models from spatio-temporal abundance data: Case of Muskrat spread in the Netherlands. Journal of Agricultural, Biological, and Environmental Statistics 1:40–59. National Atlas of the United States. 2008. Available online at http://www.nationalatlas. gov. Accessed 15 September 2007. Natural Resources Canada (NRC). 2008. The Atlas of Canada. Available online at http://atlas.nrcan.gc.ca. Accessed 15 September 2007. Nelson, J.W., and J.A. Kadlec. 1984. A conceptual approach to relating habitat structure and macroinvertebrate production in freshwater wetlands. Transactions of the North American Wildlife and Natural Resources Conference 49:262–270. Northeast Fur Resources Technical Committee. 2005. Minutes of the annual meeting of the Northeast Fur Resources Technical Committee September 7–10, 2005. Modus, CT. Nummi, P., V.M. Vaananen, and J. Malinen. 2006. Alien grazing: Indirect effects of Muskrats on invertebrates. Biological Invasions 8:993–999. Powell, R.A. 1993. The Fisher: Life History, Ecology, and Behavior. Second Edition. University of Minnesota Press, Minneapolis, MN. Schooley, R.L., and L.C. Branch. 2005. Survey techniques for determining occupancy of isolated wetlands by Round-tailed Muskrats. Southeastern Naturalist 4:745–756. Scognamillo, D.G., and M.J. Chamberlain. 2006. Forecasting models for harvest of River Otter in Louisiana. Proceedings of the Southeastern Association of Fish and Wildlife Agencies 60:25–32. Sietman, B.E., H.L. Dunn, J.K. Tucker, and D.E. Kelner. 2003. Muskrat (Ondatra zibethicus) predation on Zebra Mussels (Dreissena polymorpha) attached to unionid bivalves. Journal of Freshwater Ecology 18:25–32. Smirnov, V.V., and K. Tretyakov. 1998. Changes in aquatic plant communities on the island of Valaam due to invasion by the Muskrat Ondatra zibethicus (Rodentia, Mammalia). Biodiversity and Conservation 7:673–690. Stevens, R.T., T.L. Atwood, and J.M. Sleeman. 1997. Mercury in hair of Muskrats (Ondatra zibethicus) and Mink (Mustela vison) from the US Department of Energy Oak Ridge Reservation. Bulletin of Environmental Contamination and Toxicology 58:720–725. Wellner, M.W. 1981. Freshwater Marshes. University of Minnesota Press, Minneapolis, MN. Wellner, M.W., and L.H. Fredrickson. 1973. Avian ecology of a managed glacial marsh. Living Bird 12:269–291. Wood, W. 1974. Muskrat origin, distribution, and range expansion through the coastal areas of Del Norte County, California and Curry County, Oregon. Murrelet 55:1–4.