River otters once occupied nearly every major drainage across Canada and the continental United States, but unregulated trapping and habitat disturbance extirpated them from much of their range by the early 1900s. Kentucky was no exception: by the 1950s, otters survived only in the extreme western Jackson Purchase Region. An experimental restocking of 11 animals in the Land Between-the-Lakes area in 1982 and 1983 suggested the species could rebound, prompting a larger restoration that released 355 otters across 14 central and eastern sites from 1991 to 1994. Sightings, damage reports, and sign surveys soon confirmed otters in every major watershed, and the state opened a restricted harvest season in 2004, expanded it statewide in 2006, and progressively liberalized bag limits and zones through 2010.

Despite more than a decade of regulated harvest, wildlife managers lacked a robust estimate of how many otters actually lived in Kentucky. The new study addressed that gap using statistical population reconstruction, an integrated modeling framework that combines age-at-harvest data with estimates of catch-effort and auxiliary information from radio-telemetry. The Kentucky Department of Fish and Wildlife Resources recorded every otter harvested during the regulated season, which runs from mid-November through the end of February, and aged a voluntary subsample of submitted animals, averaging about 15.7 percent annually, using microscopic counts of cementum annuli in teeth. A mailed furbearer trapping survey, with response rates near 30 percent, provided annual estimates of trap-nights devoted to otter trapping.

The modeling approach is technically intricate. Each diagonal of the age-at-harvest matrix represents a cohort of animals born in the same year, and probability mass functions describe how harvest counts distribute across ages within each cohort, conditional on the cohort’s total harvest. Rather than estimating harvest probability directly, the model expresses it as a function of observed annual catch-effort through an age-specific harvest vulnerability coefficient. Because age-at-harvest and catch-effort data alone are often insufficient to uniquely estimate all parameters, the researchers incorporated radio-telemetry data on non-harvest survival from 53 collared river otters in neighboring Ohio and nearby Minnesota, collected between 2002 and 2004, assuming that non-harvest survival rates are similar across this region even though harvest regulations differ.

A key innovation was the incorporation of linear effects that allowed the model to test how interannual fluctuations in pelt prices influenced harvest vulnerability. Because Kentucky’s fur buyer reports were sparse in some years, the researchers combined average otter pelt prices from Kentucky, Missouri, and West Virginia, which were highly correlated, into a single standardized regional price index. The model also used a second-stage abundance estimation procedure: instead of estimating abundance as a parameter within the model, it inflates observed harvest counts by model-estimated harvest probabilities using a Horvitz-Thompson type estimator. This approach reduces the number of parameters, requires less data, and has been shown to produce less biased abundance estimates and more optimal confidence interval coverage than one-stage alternatives.

The results paint a dramatic picture of a population responding to market forces. Pre-harvest abundance declined steadily from an estimated 6,120 otters in 2012 to a low of 3,917 in 2016, then surged to a high of 8,357 by 2019 before falling again to 4,312 by 2023. The decline coincided with the three highest harvest counts of the entire reconstruction, and the researchers attribute it to elevated pelt prices from 2012 to 2014, which drove above-average harvest intensity during those years. Catch-effort ranged from a high of 381,118 trap-nights in 2013 to a low of 122,949 in 2019 and was strongly correlated with pelt prices from the previous season, with a Pearson correlation coefficient of 0.87.

The most counterintuitive finding concerns harvest vulnerability itself. Although higher pelt prices increased the total amount of trapping effort, the model estimated a significant negative linear effect of pelt prices on the likelihood of an otter being harvested per additional unit of effort. When average Kentucky pelt prices nearly doubled from $40.24 to $72.98 in 2012, trap-nights the following year rose from 283,934 to 381,118, yet the vulnerability coefficients for young-of-the-year, yearlings, and adults all decreased. The researchers suggest that soaring prices may have drawn casual or inexperienced trappers away from other furbearers and toward otters to maximize profits, diluting the average skill level of the trapping pool. When prices fell, only dedicated, highly effective trappers remained, so each trap-night became more productive. The dilution effect was real and significant, but smaller than the corresponding increase in effort, meaning higher prices still produce higher overall harvest rates, just at a slightly less-than-proportional amount.

The model also exposed striking age-specific patterns. Yearlings proved far more vulnerable to harvest than either young-of-the-year or adult otters, with baseline harvest probabilities of 0.324 for yearlings versus 0.144 and 0.151 for the other age classes. The researchers attribute this to the dispersal movements typical of that age, as young otters leave their natal ranges and encounter traps more frequently, along with higher nutritional demands, less suspicion of novel objects, and recent release from parental protection. Because yearlings represent the future reproductive segment of the population, disproportionate harvest of these individuals can depress future reproductive potential and hinder population replacement. Non-harvest survival, by contrast, was estimated at 0.835 annually and did not vary by age, consistent with estimates from nearby states ranging from 0.76 to 0.89.

Perhaps the most consequential implication involves how dramatically the new estimates revise earlier thinking. An earlier reconstruction of Kentucky otters, based on only seven years of data and a one-stage estimation approach, suggested a pre-harvest abundance of roughly 22,000 animals in 2012, nearly 250 percent higher than the updated estimate of just over 6,000. The researchers attribute the overestimate to the short data duration, minimal variability in catch-effort, and the single-stage approach. Such large differences drastically change conclusions about the rate at which otters are removed by harvest and could severely impact quota decisions. Fortunately, the state wildlife agency maintained conservative regulations and did not liberalize bag limits based on the earlier overestimates. The new absolute estimates align well with reconstructions from neighboring Ohio and Indiana, which reported abundances between 4,000 and 8,000 animals during comparable periods shortly after harvest resumed.

The study carries clear warnings for wildlife managers across North America. Because pelt prices from the current year strongly predict the following year’s trapping effort, agencies could use market data to preemptively adjust quotas, season lengths, and license numbers before unusually high demand translates into above-average harvest mortality. The researchers recommend continuing to update the reconstruction as new data arrive to ensure the recent decline does not persist, and they caution that Kentucky’s voluntary tooth submission system, which ages only about 16 percent of harvested otters, may skew age distributions if trappers favor certain age classes. Future work incorporating radio-telemetry data collected within Kentucky, particularly on harvest mortality, and random effects capturing natural year-to-year fluctuations in survival could sharpen these models further. For a species whose fate once hinged on unregulated trapping, the otter’s future now depends partly on reading the fur market as carefully as any trapper.

Subject of Research: Effect of pelt prices on river otter abundance and harvest vulnerability estimated with statistical population reconstruction in Kentucky

Article Title: Using statistical population reconstruction to estimate the effect of pelt prices on river otter abundance and harvest vulnerability in Kentucky

Article References: Berg, S. S., & Palmer, L. L. (2025). Using statistical population reconstruction to estimate the effect of pelt prices on river otter abundance and harvest vulnerability in Kentucky. Frontiers in Mammal Science, 4, Article 1634181. https://doi.org/10.3389/fmamm.2025.1634181

Image Credits: AI Generated

DOI: 10.3389/fmamm.2025.1634181

Keywords: river otter, statistical population reconstruction, pelt prices, harvest vulnerability, Kentucky, wildlife management, furbearer, abundance estimation, statistical, population, reconstruction, estimate