Fawn Deaths, Coyotes and the Results of Trapping

In studies where predators were found, they accounted for the most fawn deaths. Yet the researchers described the survival improvement from coyote trapping as modest. At South Carolina's Savannah River Site, trappers took 474 coyotes, reducing the index based on their scat by 78%. Fawn survival through 16-week monitoring reached 0.513, 0.202 and 0.431 across the three treatment years, compared with 0.228 beforehand. Contracts averaged $199 for each coyote. We reconstructed the experiment's figures and calculated a square-mile cost, then compared the results with research from a Delaware area lacking established predators. Together, these studies support identifying the problem before choosing predator removal, habitat improvements or a smaller doe harvest.

Researchers Gingery and coauthors, affiliated with Penn State and the Pennsylvania Game Commission, fitted radio markers to 98 fawns across two Pennsylvania research areas during 2015 and 2016. They combined their findings with published estimates of survival covering 29 populations across 16 states. In contiguous forest without agriculture, the estimated probability of fawns surviving to ages of 3 to 6 months was 41%. Survival was higher by 4.9 percentage points for each additional 10% of the total land area devoted to agriculture. This comparison describes an association among populations; it does not test agriculture's effect on fawns. Within every land-cover category examined, predators accounted for the most deaths. Compared with areas containing only forest, places mixing woods and farmland had both a smaller proportion of deaths attributed to predators and a lower mortality rate from predation. Human causes accounted for a larger proportion of deaths there. The researchers also found no association linking deer density with fawn survival.

Our reading was limited to the abstract because access to the complete paper is closed. The abstract supports a habitat-wide pattern rather than identifying the problem on an individual property. We omit predator-specific death totals because we could not check those figures.

Researchers Kilgo and coauthors tracked 216 fawns wearing radio collars. Monitoring covered four years ahead of trapping (2006 to 2009), followed by three treatment years (2010 to 2012). Work took place within three units measuring 32 km2 each at South Carolina's Savannah River Site. Hired trappers took 474 coyotes: 169 during 2010, 137 during 2011 and 168 during 2012. Annual removal averaged 1.63 coyotes per km2. That rate exceeds the pre-trapping density estimate of 0.8 to 1.5 coyotes per km2 that the researchers cite from howl surveys. Even so, the scat index declined 78% without reaching zero. That 78% measures the decline in an index of droppings counted per kilometer each day, which the researchers used to represent coyote abundance. It does not describe the percentage of coyotes captured. Coyotes arriving from beyond the units may account for the smaller reduction than expected, the authors suggest.

During the first treatment year, survival exceeded twice its starting level. The second year returned to the starting level; the third approached twice that level. However, the third year's 95% confidence interval overlapped the interval from before treatment. Estimates for each treatment year came from 31 to 52 fawns. The researchers point out that trapping rates were somewhat lower during 2011, when survival dropped back. They make no connection between those changes, and three years cannot establish one. There were also no units left untreated for comparison. As a result, the effects of year and treatment cannot be separated. The researchers say yearly variation unrelated to trapping remains a possible explanation.

According to the paper, coyotes led the causes of death throughout the periods shown in the table. They accounted for 80% of deaths ahead of treatment and 73% while trapping was underway. Mortality attributed to coyotes dropped to 0.357 during 2010, then rose again to 0.577 during 2011. In every treatment year, mortality from the remaining causes showed no difference from its level before trapping. The authors therefore describe coyote predation as additive: coyotes kill fawns that would survive without that predation, rather than fawns that would die from other causes. That finding supports the possibility that removal can matter. The researchers say coyote abundance declined less than they expected and suggest that incoming coyotes may account for the result. They did not test whether removing coyotes more thoroughly would have increased survival.

The study likewise did not demonstrate an effect of understory cover. Vegetation cover had no relationship with survival when considered alone or together with coyote abundance. Although the researchers expected thicker understory to benefit fawns, the estimate ran in the opposite direction, with a 95% interval extending across zero.

The researchers characterize removal's overall survival benefit as modest. They conclude that neither controlling coyotes nor managing vegetation seems effective at increasing recruitment. They also say effective control of coyotes may be harder on properties smaller than their research units, each measuring 32 km2. Immigration into those units may explain why coyote numbers fell less than expected, they suggest. They identify a smaller antlerless harvest as a measure that may be necessary, while calling for additional research in the Southeast.

Contract prices ranged between $120 and $250 for each coyote, averaging $199. Another figure in the paper, $123, represents the cost after subtracting the catch's commercial value, if contractors had been allowed to sell it. Applying $199 to 474 coyotes produces a total of about $94,000. Using the annual removal rate of 1.63 coyotes per km2, that average contract price translates to about $324 annually for each km2, or about $840 annually for each square mile. Using the net price of $123 gives about $519 for each square mile. We calculated these rates from the study's figures; an individual landowner would face a different bill.

Researchers Dion and coauthors caught 109 fawns during 2016 and 2017 in Sussex County, Delaware. Within their research area, populations of black bears, bobcats and coyotes were not established. Every one of the 42 deaths observed was attributed to natural causes other than predation, including illness, emaciation and congenital defects. Those deaths usually occurred within 7 days after birth. During that first week, risk was influenced by birth weight, the mother's age class and precipitation.

Capture involved two approaches: searches conducted as opportunities arose, and vaginal implant transmitters (VITs). Does carry these devices, helping field crews locate newly born fawns. Ninety-day survival among fawns located through VITs was 0.44, compared with 0.69 among those found opportunistically. The authors explain the difference as early deaths missed by opportunistic searches. They report that mortality unrelated to predation left survival comparable with that in regions where predators were established. They say mortality from causes other than predation may be the ultimate constraint on survival in other regions, obscured by predation as the more readily visible cause.

Few fawns observed, then, do not establish that coyotes are the problem.

We reviewed the complete Kilgo paper; for the remaining two studies, we read their abstracts. The trapping experiment covered a single location, used a single removal method and ran for three treatment years. Each treatment year's estimate involved 31 to 52 fawns. The location was federally owned, wooded land in the Southeast, and the experiment lacked comparison units without treatment. Our square-mile estimates use the experiment's average price and removal rate. The paper's Table 5 reports 151 deaths, while its text reports 152. We therefore use the authors' percentages rather than calculating death shares ourselves.

Within one of our site's tables, property_metric, we searched 604 metric names. We found none measuring fawn survival, fawn mortality, predator abundance or coyote harvest. Rows concerning fawns contain counts and ratios, including North Dakota's count of fawns for every 100 does. Our search did not extend to the site's other data stores, and this article includes no chart from our data.

Identify the cause of poor fawn recruitment before spending money to address it. Track recruitment over several years, leaving coyote removal as the final and most expensive option. A single year cannot identify the problem: survival at the Savannah River Site ranged from 0.202 to 0.513 while the same trapping program continued. Before attributing losses to predators, examine local habitat and the other causes documented in these studies, including Delaware's illness, emaciation and congenital defects. The trapping researchers say recruitment may require a reduction in doe harvest, meaning hunters fill fewer antlerless tags. They also call for more study of that approach in Southeastern herds. Taking fewer does leaves more females available to produce fawns, although the experiment did not test that approach. To begin examining the intent behind your state's antlerless policy, read the buck-to-doe fingerprint.

If removal still appeals to you, first calculate the expense. The Savannah River removal rate and average contract price imply about $840 annually for each square mile, on property that may be too small to retain the benefit. Our seven-state decline paper demonstrates how herd size, antlerless policy and disease shocks jointly shape a single harvest trend. Those combined influences are why fawn sightings from a single year should not determine the choice.