Every human-wildlife conflict has a file. A date, an address, a species, a category, and an officer's name at the bottom. The file is accurate.
It is also the last page of a story that started months earlier and a few thousand feet uphill, in a berry crop that did not set.
This post is for the people who receive those files — wildlife agencies, town and county staff, watershed and landscape partnerships, land trusts, and the carriers who eventually see the claim. The definition first, in the language your own forms use. Then the part the form has no field for.
what human-wildlife conflict means in agency language
The IUCN Species Survival Commission's Human-Wildlife Conflict & Coexistence Specialist Group defines human-wildlife conflict as "struggles that emerge when the presence or behaviour of wildlife poses an actual or perceived, direct and recurring threat to human interests or needs, leading to disagreements between groups of people and negative impacts on people and/or wildlife."
That is the working definition behind the IUCN SSC guidelines most agencies now reference. Two phrases in it do real work, and both get skipped in casual use.
"Actual or perceived." A conflict does not require a loss. A report about an animal someone believes is likely to cause a problem sits inside the definition. That is not sloppiness — it is an accurate description of what agencies actually manage, which is a caseload of calls, not a caseload of carcasses.
"Between groups of people." The definition is explicit that the conflict is usually human-to-human as much as human-to-animal: the neighbor who feeds, the neighbor who calls, the HOA board, the rancher, the outfitter, the officer who has to decide. That is why the practitioner field now writes it as conflict and coexistence, and why resolution is so rarely about the animal.
Operationally, the categories are narrower than the definition. California's Wildlife Incident Reporting system, run by the Department of Fish and Wildlife, sorts reports into four types: depredation (the animal caused property damage), potential human conflict (someone observed an animal they expect will cause a problem), general nuisance (a negative interaction with no property damage), and sightings. Between 2017 and 2023 it logged 31,904 reports across 63 species. Depredation accounted for 18,190 of them.
Hold those four categories. They are about to say something the definition cannot.
the turn: the file is a food-year record
In November 2025, Kendall Calhoun and colleagues published an analysis of that full incident database in Science Advances, matched against twelve-month rolling precipitation at four-kilometer resolution. The headline result is unusually easy to quote.
For every 25 millimeters — roughly one inch — of decrease in annual precipitation, reported conflicts rose:
| species | increase in reported conflicts per ~1 inch less annual rain |
|---|---|
| mountain lion | 2.11% |
| coyote | 2.21% |
| American black bear | 2.56% |
| bobcat | 2.97% |
Take the asymmetry inside that result into your next meeting, because it matters more than the percentages. Depredation, general nuisance, and potential-human-conflict reports all rose as precipitation fell. Sightings did not. Precipitation had a neutral effect on people simply reporting that they had seen an animal.
That functions as an internal control. If dry years only made people more watchful or more irritable, sighting reports would have moved too. They did not. What moved were the three categories that describe a negative interaction — a loss, a nuisance, or a threat someone expected — while the one category with no negative valence at all, "I saw an animal," stayed flat. The authors had expected potential-human-conflict reports to be weak; they were not. Perception under drought is part of the signal, which is exactly why the caveat below matters.
The honest caveat comes from the lead author, not from us. Calhoun has said plainly that it is unclear whether reports rise because there are objectively more conflicts, or because people perceive wildlife more negatively when their own resources are stressed. Both can be true at once. Keep the caveat — and then notice that it does not change the operational fact: the file is what the town, the agency, and the carrier actually pay for, whichever mechanism produced it.
Colorado's 2026 season is the same relationship at a harder scale. Colorado Parks and Wildlife released preliminary figures through 31 August 2026: 8,672 bear-related sightings and incident reports, against 3,680 through August 2025; 98 relocations against 33; 294 bears euthanized against 83. CPW labels these numbers preliminary and will finalize them in its 2027 report.
Those are not scoreboard numbers. They are the arithmetic of a year in which critically low snowpack, a late spring frost, widespread drought, and a heavy fire season took out the mast (acorn and nut) and berry crop before the animals needed it most. Travis Black, CPW's northwest regional manager, put the constraint on the record: "While relocation is an option, with bears often it's not the best option, especially when you're having a bad bear year with mass crop and berry failures out there." Relocated bears have turned up back in the same block within days of a forty-mile move.
Read the sequence rather than the total. Snowpack fails. Frost takes the bloom. Gambel oak, serviceberry, and chokecherry produce nothing worth the walk. Hyperphagia — the pre-den feeding surge, when a bear works to put down roughly 20,000 calories a day — arrives on schedule in August whether or not the food did. Then the reports arrive. The animal did not change its behavior — its pantry did.
Human-wildlife conflict is what a failed food year looks like once it reaches a form.
what each file is actually recording
These are not California's four report categories. They are the ladder a file climbs — two ways a report gets opened, two ways it gets closed — and the labels that show up in ordinances, insurance notes, and news coverage. Each records something real. None of them has a field for the hillside.
| the file | what the form records | what it has no field for | who absorbs the cost |
|---|---|---|---|
| sighting | A person saw an animal and chose to call. In the California data, sighting frequency does not track precipitation — this is the category that stays flat in a dry year. | Whether anything was actually lost. | Dispatcher and officer time. |
| property damage | A bin, coop, shed, vehicle, orchard, or livestock loss, with a date and an address. The largest category by far. | The berry crop that failed in July, four thousand feet upslope and two months earlier. | Homeowner deductible, HOA, rancher, carrier. |
| habituation | A pattern across repeat reports — same animal, same block, a learned route to reliable calories. It is a management judgment made by officers, not a checkbox on a form. | That the learned route was the only route still producing. | Agency staff hours, relocation attempts, and the animal's remaining options. |
| euthanasia | The end of the management ladder, and a decision a person had to make in the field. | Everything upstream of it. | The animal. Then the town again next year, when the next one arrives. |
Habituation is the hinge. It is the only row where the file stops describing an event and starts describing a relationship — and it is the row where management options collapse from many to two.
a raid and a collision are different invoices
If you arrived here from the movement side of this problem, keep the two straight, because they get merged constantly and they do not share a fix.
A wildlife-vehicle collision is the invoice for a severed route. The animal was going where it has always gone, and pavement was in the way. A dumpster raid is the invoice for a failed pantry. The animal was going somewhere new, because the old place stopped producing.
Different failure, different remedy, different budget line. A crossing structure will not put fruit back on a hillside. A funded berry crop will not stop an elk from stepping onto a highway at dusk. Both are habitat-condition problems; they land on different desks with different engineering.
We keep that work filed separately on purpose: what a wildlife crossing actually is covers the structure and the four components that make one work, and the carcass is the invoice covers what collision data does and does not count. Neither of those is about the pantry. This one is.
treat the file and fund the pantry
Here is the part of Calhoun et al. that rarely makes the news write-up. The mitigation discussion does not end at attractant management. It ends at supply.
The authors point to artificial water catchments that give wildlife an alternative to human water sources in drought years, and to prescribed burning that enhances habitat and increases the abundance of natural prey. Their phrase is promoting nonanthropogenic resources for conflict-prone species. Out of journal register: give the animal a reason to stay where it was.
Two limits, stated plainly. This is not supplemental feeding — Colorado Parks and Wildlife has refused that since a 2013 commission policy, because piles of food concentrate solitary, territorial bears, spark fights, and teach them people mean meals. And it is not a same-season fix. Water that holds through a dry August, fire that keeps oak and shrubland producing mast, riparian ground that stays green, less water drawn out of the wild range: these take years to build and pay off in the next bad year, not this one.
That is a habitat-condition investment, and it is the one line item that almost never has a funding source attached. Trash enforcement has a budget. Officer response has a budget. Relocation has a budget, a truck, and a protocol. The berry crop has volunteers and weather.
This is the gap ensurance is built for. The mechanics in one sentence: ecological condition is valued as a producing asset, and money routes to a named place through an account that holds and spends it in the open — funding forage, water, and habitat before the failure rather than compensating after it. habitat.ensurance is the purpose account for the habitat flow, one of nineteen ecosystem service flows we account for. It is live and the volumes are small; we would rather say that than imply a market that does not yet exist.
For carriers, the useful implication is narrower and more immediate: this exposure has a weather variable attached to it. Precipitation is gridded, modeled, and forecast. Calhoun measured reports, not claims. But depredation — property damage — is the largest report category, and if claims track it even loosely, a wildlife-damage book that follows a twelve-month rolling precipitation deficit is a book you can see coming — and one whose severity depends partly on whether anyone funded forage upstream while the deficit was building. The coverage questions themselves — dwelling, other structures, comprehensive auto, and the attractant denial — are their own post, coming in this series.
None of this replaces the file. Somebody still has to take the call, secure the bin, run the hazing, and make the hard decision in the dark. The claim is narrower than "conflict management is broken," because it isn't. The claim is this: as long as the only funded response sits downstream of the report, the report count is the only number that can move.
frequently asked questions
what is human-wildlife conflict?
Human-wildlife conflict is the struggle that emerges when the presence or behaviour of wildlife poses an actual or perceived, direct and recurring threat to human interests or needs — leading to disagreements between groups of people and negative impacts on people, wildlife, or both. That is the IUCN Species Survival Commission definition, and it is the one most wildlife agencies work from. In practice it shows up as four kinds of report: sightings, property damage, nuisance interactions, and anticipated conflict.
why does drought increase human-wildlife conflict?
Because drought removes calories and water from the wild range before it removes them from town. Berry, acorn, and forage crops fail; prey biomass drops; surface water disappears. Animals follow what is left, which is irrigated, watered, and stored — in yards, orchards, livestock pens, and bins. Calhoun et al. measured the relationship across 31,904 California incident reports: every 25 millimeters (about one inch) of decline in annual precipitation was associated with a 2.1% to 3.0% increase in reported conflicts for mountain lions, coyotes, black bears, and bobcats. Reported sightings, by contrast, did not track precipitation at all.
is human-wildlife conflict the same as a nuisance animal?
No, and the difference matters legally and operationally. "Nuisance" is a status assigned to an individual animal under a local ordinance or state code, usually triggering a defined response ladder. Human-wildlife conflict is the broader condition — a relationship between a food year, a landscape, a set of human attractants, and often a disagreement between neighbors about what to do. One animal can be designated a nuisance. A watershed cannot. Treating a landscape-scale food failure as a series of individual nuisance cases is exactly how agencies end up with a caseload that grows faster than the staff.
what to do with this
If you receive these files — agency, county, town, or watershed partnership — the near-term move is to put a funding line next to the habitat side of the ledger, not just the response side. Start at habitat.ensurance.
If your work is movement, not food — corridors, crossings, collision data — start with what a wildlife crossing actually is and the carcass is the invoice. Different invoice, adjacent problem.
Next in this series: the same raid, different species — bears in Aspen, elephants in maize, baboons at the bins, and the one mechanism underneath all three.