Wildlife agencies could more easily assess their state’s elk, deer, or pronghorn populations if every animal alive in August fell into one of two groups by Christmas: those registered as hunters’ kills, and those destined to see spring’s fawning or calving seasons.
It’s never that easy, of course, because legal hunting harvest isn’t the only drain on a herd’s bottom line. The long list of other possible deductions includes disease, injury, predation, poaching, starvation, winterkill, roadkill, unregistered kills, wounding loss, and freak accidents.
Despite such unpredictable factors, agency biologists calculate herd sizes and sex ratios with consistent precision each year, thanks to decades of fine-tuned laws, surveys, indexes, aerial counts, registration systems, and population models tailored to each state. In other words, accurately assessing wild, free-ranging herds requires constant data gathering from many sources.
“Deer don’t die between starched sheets in numbered rooms,” said Keith McCaffery, a semi-retired deer biologist who’s spent over 40 years studying and calculating whitetail populations for the Wisconsin Department of Natural Resources. “Except for those brought home by hunters, most dead deer melt back into the ground without notice. But the more data you collect on a herd, and the longer and more systematically you collect it, the better your population estimates become.”
Biologists can estimate some drains through data-driven systems like winter-severity indexes, which use sustained days of snow depths and sub-zero temperatures to estimate regional losses. And more recently, deer fitted with GPS and radio-telemetry collars provide unprecedented insights into disease, predation and hunter-inflicted wounding losses.
Some agencies lump all such deaths as an aggregate loss that adds 15% to hunting’s registered kill. Other agencies estimate wounding losses alone add 15% to the registered kill; and still others add 25% to the registered kill to cover disease, roadkill, wounding losses, and starvation, especially during severe winters in the Northwoods, Great Plains, or West.
Those calculations aren’t random estimates, but neither are they based on bar-coded death certificates or coroner reports.
“It’s hard to untangle many of those mortality factors,” said Dan Storm, a Wisconsin DNR researcher. “Sick, injured, or starving deer are prone to much higher hunting and predation risks. Without a hunter, wolf, or coyote to finish them off, they might die anyway. How deer die might not matter much in population estimates, but you still need a good sense of what percentage of the herd you lose each year beyond the registered harvest.”
Human-driven drains such as roadkills and wounding losses, however, can vary by region, habitat, weather, weapon type, harvest restrictions, hunting pressure, traffic volume, and time of year. Still another complexity is evolving societal norms. Hunters today seldom condone poaching, no matter the violator’s economic situation. That might explain why illegal kills today aren’t big factors in herd dynamics. In Colorado, for example, radio-tracking studies show poaching accounts for 0.4% to 1.5% of doe deaths, and 1.5% to 1.7% of buck deaths.
Hunters today also don’t tolerate “ground checking” and “shoot-and-sort” compared to their counterparts 50 to 80 years ago.
Consider this paragraph from Aldo Leopold’s “A Sand County Almanac,” published in 1949: “A common denominator of all sporting codes is not to waste good meat. Yet it is now a demonstrable fact that Wisconsin deer hunters, in their pursuit of a legal buck, kill and abandon in the woods at least one doe, fawn, or spike buck for every two legal bucks taken out. In other words, approximately half the hunters shoot any deer until a legal deer is killed.”
Some might accuse Leopold of exaggerating, but he was likely citing data later published by Burton Dahlberg and Ralph Guettinger in their 1956 book “The White-Tailed Deer in Wisconsin.” The book cites post-hunt reports, searches, and surveys from Leopold and his era, including these:
Eight doe carcasses on a 300-acre property in 1928; 60 illegal deer per 100 legal bucks in 1938; 68 dead or wounded bucks, does and fawns left behind for each 100 legal bucks taken home in 1939; “an illegal kill equal to the legal kill, plus a crippling loss of legal bucks equal to one-third of the legal kill” in 1941; 130 illegal deer for every 100 legal bucks in 1947; and 67 illegal kills per 100 legal fork-horned bucks in 1948.
Dahlberg and Guettinger wrote: “Almost all evidence collected in Wisconsin since 1941 indicates these estimates … are, if anything, conservative.”
And when Colorado’s wildlife commission imposed statewide protections on spike bulls in 1971, a post-hunt survey of northwestern Colorado’s White River elk herd by the Division of Wildlife documented a record 32 yearling and two mature bulls abandoned. The previous highs were 13 abandoned bulls in 1968 and 1969.
Research since then reported far better compliance. McCaffery reported that research from 1963 to 1989 at central Wisconsin’s Sandhill Study Area estimated the average unretrieved wounding kills by firearms were 6.5% of the legal kill. A mid-1990s study in South Carolina by Charles Ruth reported unretrieved woundings, and deer that wouldn’t have been found without tracking dogs, were about 8% of the total kill. And a recent Wisconsin study of GPS-collared deer in two regions put wounding losses at 3% in the northern forest area and 15.3% in the farmland area, or 10% overall.
Colorado’s 4-point rule for bull elk enjoys high compliance. Conservation wardens issued 131 citations for antler-point violations from 2006 through 2017, an annual average of 11. “It’s easier to see and count points on elk because their antlers are big and they generally live in more open country,” said Scott Wait, senior Colorado Parks and Wildlife biologist in Durango.
Studies of archery losses also reported respectable numbers. The Camp Ripley study in central Minnesota in the early 1990s found that bowhunters retrieved 87% of deer reported hit. Of the 13% that weren’t found, many likely survived. As Ruth notes, deer are more likely to survive arrows than bullets. Another researcher, Andy Pedersen, studied bowhunting success for nearly 30 years at the Indian Head Navy Base in Maryland. He reported that archers found 85% of deer they arrowed from 1989 through 2018.
McCaffery thinks hunting practices have also changed in recent decades, noting that deer drives have given way to stand-hunting across whitetail country, especially among bowhunters. Therefore, hunters see fewer fleeing deer, and more often shoot deer as they walk or stand.
Wait said Colorado’s biologists use three figures as “starting points” in assessing the percentage of a herd wounded by hunters but not found each year: 25% for cow elk, 15% for bull elk and elk calves, and 10% for deer and pronghorns. He said those figures come from decades of documented statewide surveys, observations, and radio-collar studies, lumping the wounding averages for all methods of take.
Why do those percentages differ by age, species, gender, and body size? Wait said it’s impossible to prove a “why,” but agency biologists follow rational, time-tested, common-sense theories based on years of research, interviews, and observations.
For instance, elk are tougher to kill than deer or pronghorns, and less likely to drop within sight of hunters. Deer and pronghorns are more likely to fall or react visibly to hits. Hunters are more likely to search impact sites after shooting at elk, and track possible hits. Hunters are also more likely to search harder and longer for wounded bulls and bucks than for cows, does, calves, or fawns. They’ll also search harder and longer for big-antlered bulls and bucks than for spikes.
Objective or Subjective?
Wait said Colorado’s biologists don’t lock wounding-loss estimates into their calculations because they can vary by year, location, and weather.
“Population modeling is like balancing your checkbook,” Wait said. “You look at what’s going in and what’s going out, and if your bottom lines don’t match you go back and see what you didn’t account for. Mathematical models use calculations based on biology, aerial flights, on-the-ground surveys, and other objective data. But if local managers think the model isn’t giving realistic projections of what they observe, they make some subjective adjustments.”
Those adjustments might include reassessing bull-to-cow ratios, because bull elk observations are minimum counts based on actual annual sightings, not historical estimates and averages. If biologists conclude they didn’t look in the “right” places during December surveys, when bulls split from cows and calves, they gather more representative samples.
Colorado’s population models also include more general mortality factors such as predation, disease, road-kills, and winter losses. Wait said a good population model draws from at least 10 years of local, regional, and statewide data to ensure accuracy.
“If I created a model from three years of data, I’m just throwing darts at a blank wall,” Wait said. “But 10 years? The model then forces itself to reflect annual survey data like harvest numbers, and age and sex ratios. And when you run models with 20 years of annual data, they’re better yet.”
Agencies run into problems, of course, whenever their models and estimates don’t align with observations of hunters, farmers, ranchers, and other locals.
“I love hearing real observations from people who’ve been on the land 80 years,” Wait said. “If they’re only seeing 20 deer in a field where they used to see 60, we try to figure out why. Are there fewer deer, or did a subdivision down the road change the herd’s distribution? I really enjoy those discussions. But you also must assess whether they’re giving you objective observations, or just advocating for antler-point restrictions or keeping nonresidents out. I get concerned when advocates are blind to real issues.”
Not every wildlife agency finds wounding loss all that complicated, however. Duane Diefenbach, a professor at Penn State University, said research the past 20 years suggests only 1% to 2% of deer shot by hunters aren’t found.
“Pennsylvania loses way more deer to cars and trucks than it does to unrecovered hunting losses,” Diefenbach said. “Outside of our hunting season, we might lose 5% to 12% tops to other factors, and that’s mostly cars, with maybe an occasional freak accident.”
Jim Heffelfinger, wildlife science coordinator for the Arizona Game and Fish Department and a research scientist at the University of Arizona, said only about 1% of hunters responding to annual post-hunt surveys report hitting deer they don’t find. In his book “Deer of the Southwest,” he wrote: “Mortality varies greatly with hunter density, thickness of vegetation, terrain, and hunt structures. Like many mortality factors, the number of deer left in the field after the hunting season would be very difficult to estimate accurately, and any estimate would not be applicable to wide areas or subsequent years.”
Heffelfinger also thinks wounding loss will always be part of predation, whether by human, avian, or four-legged predators.
“It would be unrealistic to expect all shots to result in instantaneous death,” he wrote. “Predators certainly do not attain that ideal, and they also wound deer without bringing them down. Deer bones from Arizona archaeological sites show that Native Americans also didn’t recover every deer they shot. Deer remains from an archaeological dig had healed with a flint arrowhead embedded in its sternum.”
Even so, modern hunters appear to embrace Aldo Leopold’s “one-bullet, one-buck” ideal, and his disdain for those who leave illegal kills where they fall. “Such deer hunting is not only without social value, but constitutes actual training for ethical depravity elsewhere,” Leopold wrote.
Assuming that recent studies into wounding loss and illegal kills are accurate, hunters today have taken Leopold’s scolding to heart.
Feature image via Captured Creative.