Coyotes

For a new fence line there is 2 by 4 inch welded wire, standard or hot-dip galvanized after welding, and a heavier 100-foot roll with 4 by 4 inch mesh. For a wood dog-ear fence, the Coyote Roller 8-foot kit mounts spinning rollers on top, and a shorter 4-foot kit also fits chain-link.

What Was Checked, and What Is Named Here

One figure in that specification is not something any product can carry, and it is worth separating before the cards appear. UC IPM’s minimum fence height of 5 1/2 feet is an instruction to whoever builds the fence. It is not a property of a roll of wire, and no product below is described here as meeting it.

What the Sources Describe

UC IPM opens on the normal case: “Normally, coyotes are elusive animals that avoid contact with humans. Most active after dusk and before daylight, they are typically seen only at a distance.” It then describes the change that produces the problem: “Within urban and suburban areas in California, however, some coyotes have adapted to residential neighborhoods, parks, and open spaces, and seemingly have lost their fear of humans.” and adds that “Suburban coyotes can reach densities far greater than they do on rangeland or undeveloped wildlands.”

UC Cooperative Extension in San Diego County describes the animal as primarily nocturnal and omnivorous, calls them “not picky eaters”, and notes they have been known to attack and eat domestic animals, especially cats.

Attractants: Where the Sources Put the Cause

UC IPM does not treat the loss of fear as something that happens on its own. It names a cause, and the sentence is the spine of its whole page: “Never intentionally feed or provide water to coyotes, as this causes them to quickly lose their fear of people and become aggressive.”

Everything the same source says about exclusion, hazing and removal sits downstream of that sentence. It is worth reading before the parts of this page that involve buying or building anything.

Hazing: a Live Disagreement

Two sources, nineteen years apart, and this page does not pick between them.

UC Cooperative Extension in San Diego County defines the term plainly: “Hazing is pretty much what it sounds like” — shouting, arm waving and noise-making directed at coyotes to change their behavior.

The 2007 position: devices habituate, and are impractical in suburbs

UC IPM, in text stamped March 2007: “Using sound or visual stimuli to keep coyotes away from livestock or other resources will provide only temporary effectiveness, if any.” The same source states that “In the absence of any real threat, coyotes quickly adapt or habituate to sounds, flashing lights, propane cannons, scarecrows, and so on.” It reports that a strobe-siren device developed for sheep flocks “was effective only for several weeks to a few months at most locations where it was tested”, and concludes: “Because of the disturbance such devices cause, they are impractical for use in suburban areas.”

The same 2007 source nonetheless recommends hazing by people rather than by devices: “Suburban residents who see a coyote in their neighborhood should attempt to frighten it away by shouting, throwing rocks, squirting it with a water hose, blowing portable air horns, or otherwise acting aggressively in order to reinforce its fear of people.”

The 2025 study: retreat observed, durable change not observed

A peer-reviewed study published in April 2025 followed community-based hazing in Edmonton by 120 trained volunteers over 1,598 patrols. It reports: “Coyotes retreated before volunteers were within 40 m during 124 (71%) of the observations and retreated immediately from 22 (96%) of the hazing events.”

The paper’s own null result is the part that bears on the disagreement: “We found no evidence that hazing changed subsequent measures of overt reaction or flight initiation distances by coyotes” — which the authors attribute in part to how rarely it was implemented. Their conclusion is carefully qualified: “Our study supports the use of community-based hazing to reassure members of the public while potentially promoting wariness in coyotes.”

These two sources are not simply opposed, and it is worth saying how. The 2007 source is about devices; the 2025 study is about people. Both are skeptical that the effect lasts — one on principle, the other on its own measurements. Where they differ is on whether the practice is worth doing anyway. This page reports that difference and does not settle it.

UC Cooperative Extension in San Diego County describes a parallel Southern California study — twenty GPS-collared coyotes, ten hazed and ten as controls, with a crossover after eighteen months. Its results are not in the material we read, so this page reports that the work exists and says nothing about what it found.

Exclusion: the Specification, to the Inch

This is the most precisely specified thing in any source consulted for this page. UC IPM gives it as one design with a stated limitation attached, and the limitation comes first in its own text: “While it is expensive and difficult to construct a completely coyote-proof fence, a fence that discourages coyotes will have the following design characteristics.”

Height. “Fence height should be a minimum of 5 1/2 feet and should be built higher on sloping terrain.”

Mesh aperture. “Net wire-mesh should be no larger than 6 inches between stays.”

The buried apron. “To deter digging under, bury a galvanized wire-mesh apron, attached securely to the bottom of the fence, 4 to 6 inches below the soil and extending outward at least 15 inches.”

The mesh figure, and what sellers state

Of the four figures above, the mesh aperture is the one retail listings state plainly. All three rolls below give it in their own text, and all three are inside UC IPM’s limit of no larger than 6 inches.

The three rolls are in no order of preference and none is called better than another, because no source consulted for this page compares one product against another. What each entry records is which part of the specification the seller states, and which it does not.

The rollers below are named because the sources on this page describe a rolling fence topper as the exclusion method, and each listing states the length it covers.

The buried apron: nothing is named, and the number is the reason

UC IPM asks for an apron extending outward at least 15 inches. One product we read is built for exactly that job — its own description calls it a physical barrier of galvanized welded wire on the ground leading up to the fence — and it publishes what it achieves:

Its own text states the kit allows for 8 to 12 inches of coverage on the ground. Against a specification of at least 15 inches, that is short, and the seller is the one supplying the number. So nothing is named for this class. That is a finding about a published figure rather than a doubt about a product, and it is the cleanest kind of result this method produces: the criterion was answerable, and the answer was no.

The same description mentions a larger version of the same system with a greater ground coverage, which would clear the 15 inches. We have not read that listing, and this page names products it has read. It is recorded as something to check rather than as something to buy.

Rollers and overhangs: additions to that fence, not alternatives to it

UC IPM gives both in a single sentence, and the opening words govern everything after them:

“An extra degree of protection against coyotes scaling a fence can be obtained by installing a wire-mesh overhang of at least 18 inches, slanted outward, or roller-type devices designed to be attached to the top of a fence, which prevent coyotes from getting a foothold in their attempts to climb or jump over.”

Read as the source writes it, neither the overhang nor the roller is a measure you take instead of the height, the mesh and the apron. Both are an extra degree of protection added to a fence that already has those three. A roller fitted to a four-foot fence with an eight-inch mesh and no apron is not what this sentence describes, and this page will not present it as though it were.

Rollers: three listings state the mechanism

Two of the three are the same kit in two fence-type fitments and are named here as one product, with both fitments identified so a reader picks the one matching their fence. The third is a different manufacturer and a different length.

The outward overhang: a class that exists, and nothing named in it

This is the one place where what is sold and what the source specifies come apart, and it is worth setting out rather than passing over.

UC IPM asks for a wire-mesh overhang of at least 18 inches slanted outward. Outward matters: the point of the angle is to defeat an animal climbing in from the far side.

There is a healthy retail category of fence-top extension arms with mesh, and several listings state an outreach figure comfortably past 18 inches. Almost all of them are containment toppers. They are sold to stop a cat or a dog getting out, and the arm therefore turns back over the owner’s own garden. A stated outreach of twenty-four inches is a real measurement of a real arm; it says nothing whatever about which side of the fence that arm leans over.

A figure that matches half a criterion does not satisfy it. The direction is not a detail attached to the specification — it is the part that does the work, and no listing we read states it. So nothing is named for this class, the criterion stands as UC IPM writes it, and no product is faulted by name for being built to do a different job perfectly well.

UC IPM closes the section with a caveat that belongs with every figure above it: “In general, truly coyote-proof fences are so expensive to build and maintain that they are economically viable only to protect very valuable commodities.” That is the source’s own assessment of its own specification, and it is not softened here.

On electric fencing, UC IPM says designs of various kinds have been effective in excluding coyotes and that retrofitting existing fences with electrified wires may add effectiveness, then adds: “Additionally, electrical fencing may be inappropriate for use or illegal in residential or suburban areas.” Since residential and suburban readers are most of this page’s audience, that caveat is given here rather than the recommendation it qualifies.

Pets and Personal Safety

UC IPM records something that bears directly on how much protection a large dog is assumed to have: “In suburban areas, there have been instances of groups of coyotes attacking large dogs such as Labrador retrievers, even in the presence of their owners.”

On guard animals it gives a qualified positive: certain guard dog breeds, llamas and donkeys may exclude coyotes from pastures and are most effective when bonded to the stock, and then “there are occasions when guard animals are of limited effectiveness, and multiple coyotes or mountain lions may even attack guard animals.”

UC Cooperative Extension in San Diego County reports a dated figure on pet predation: “In a 2019 study, cats were found in 19.8% of coyote scat.” One study, one year, one region, and it is given with those limits attached.

Attack Risk, and How Old the Numbers Are

Both figures below come from a source whose text is stamped March 2007. They are reported, not assessed here, and neither is a current count.

UC IPM, in text stamped 2007: “More than 160 such attacks have occurred in California since the 1970s, and they are becoming more frequent, particularly in suburban areas of Southern California.” And: “While only one attack has been fatal (to a 3-year-old girl, attacked in her front yard in 1981), a number of attacks have resulted in serious injuries.”

A nineteen-year-old total, of something the same sentence says was becoming more frequent, is not a figure for today. We could not reach a state agency source to find a current one. What that number supports is that attacks have happened and have injured people; it does not support any statement about how many there have been by now.

UC IPM also publishes a staged behavioral progression — it describes “a predictable pattern of change in coyote behavior in these environments” set out as a table. That table is a graded risk-assessment ladder and it is named here rather than reproduced. What this page will report from it is the source’s own threshold: the stage at which pets are attacked is where UC IPM says area-wide action is recommended.

Bites, Scratches and Rabies

UC IPM gives one complete instruction, and it is quoted here in full rather than summarized: “If you are bitten or scratched by a coyote, wash the affected area thoroughly with soap and water and then seek immediate medical attention.”

On the risk itself: “Although most problem coyotes are healthy, the risk of rabies is always present.” The same source describes transmission by bite or scrape, and by handling a pet after an attack where saliva contacts broken skin or mucous membranes.

UC IPM goes on to describe what medical authorities typically recommend after contact with a wild coyote during an attack. That material is clinical, and this page names it rather than repeating it. What we will report is that UC IPM addresses post-exposure immunization and attributes the recommendation to medical authorities. What treatment is appropriate, and when, is a matter for a clinician and not for this page. The instruction above — wash the area, seek immediate medical attention — is the source’s own and is complete as it stands; nothing has been added to it.

On pets, UC IPM says to “vaccinate all pets for rabies, distemper, parvovirus, and other diseases, as recommended by a veterinarian”. On reporting, it says aggression or an attack should be reported to local authorities including animal control and the state wildlife agency, and livestock attacks to the county agricultural commissioner.

The Legal Frame, and a Dating Problem

What UC IPM stated in 2007, about California

UC IPM, in text stamped March 2007, referring to the California Department of Fish and Game: “Coyotes have no special protection in California and may be killed by any method that is not prohibited by federal, state, or local statutes.”

UC IPM, in text stamped March 2007, on leghold traps in California: “Since the passage of a state ballot initiative measure in November 1998, leghold traps cannot be used to capture coyotes except in situations where a human health and safety emergency has been declared by designated officials, or in selected situations where the existence of an endangered species is threatened by predation.”

UC IPM, in text stamped March 2007, on toxicants in California: “Toxicants or poisons used to control coyotes are illegal, with the exception of fumigant cartridges available only to predator control specialists to asphyxiate coyote pups in their dens.” The closing words are part of the exception rather than a detail after it: as that source states it, the exception covers one purpose, not general professional use.

UC IPM, in text stamped March 2007, on repellents: “No chemical repellents are registered for use in repelling coyotes from property or from livestock.”

UC IPM, in text stamped March 2007, on relocation, and naming the California Department of Fish and Game: “Captured coyotes must be destroyed, as relocating problem coyotes would incur unacceptable liability and risk on the part of the agency involved, and relocation is illegal without prior approval of the California Department of Fish & Game.”

Colorado is a different state with a different rule

A recommendation written for one state is not advice for another, and this page keeps them apart rather than averaging them. Everything in this subsection is Colorado’s and applies to Colorado.

Colorado State University Extension, in a publication updated 6 March 2018 citing Statute 33-6-107(9) and Wildlife Commission Regulations, states that where wildlife is causing damage to crops, real or personal property or livestock, “a person (or any employee or agent of the landowner) may hunt, trap, or take the following wildlife on lands owned or leased by the person without securing a license to do so” — and coyotes are named in that list.

On method, the same 2018 source is explicit: “If trapping, the ONLY types of traps that may be used are LIVE (cage or box traps).” On relocation: “Small game and furbearers captured in live traps cannot be moved from the capture site and must be killed or released on site when the trap is checked.” It records that tree squirrels, cottontail rabbits and raccoons may be relocated without a permit under stated conditions, and that “If you want to relocate any other species, you must first obtain a Relocation Permit from CPW.”

Two states, two rules, and neither is a national rule. California’s 2007 source describes professionals using padded leghold traps and snares with leghold traps barred to the public since 1998; Colorado’s 2018 source says a landowner may take a coyote causing damage without a license but may use live cage or box traps only. This page has read two states. It has not read the other forty-eight and says nothing about them.

A tension between two texts, recorded and not resolved

Our California trapping rules page quotes Title 14 section 465.5(e): it is unlawful for any person to use a steel-jawed leg-hold trap, padded or otherwise, to capture a nongame mammal, with an exception only for government employees or their duly authorized agents in an extraordinary case.

UC IPM, in text stamped March 2007, states that urban coyote removal is done by professionals who capture them in “padded leghold traps or snares”.

Both texts are set out here and neither is adjusted to fit the other. Whether a particular professional falls inside the exception in section 465.5(e) is a question about a licensing regime, and answering it would mean reading statutes this page has not read. Our California trapping rules page also records that section 465.5 itself contains no provision on relocation or transport, and warns in its own words that this is not a statement that moving an animal is lawful. The same caution applies to the relocation sentence quoted above: it is UC IPM’s statement, dated 2007, and it is not statute.

Removal, and Who Does It

This is the half of the subject that the sources do not give to a householder.

UC IPM: “Typically, coyote removal in urban or suburban areas is conducted by predator control professionals who shoot coyotes or capture them in padded leghold traps or snares.” And: “Professional assistance is usually required to remove the responsible coyotes as selectively and as efficiently as possible.” And the sentence that settles the shape of this page: “They can also use certain tools and methods that are not generally available to the public.”

Does removal work? The same source states both halves, one after the other

“Management experience has shown that removal of only a few problem coyotes from a population will reinstill fear of humans in the remaining population, often solving coyote problems in that locality for months or even years.”

“Because other coyotes quickly move in to occupy vacant territories, removal of several animals has no long-term impact on coyote numbers.”

Those two sentences are consecutive in the source, and they are given together here for that reason. The first is about behavior in one locality; the second is about population numbers. They are routinely quoted against each other as though one refuted the other. This page renders both, says what each is measuring, and leaves the source’s own framing intact.

Who the sources tell a reader to contact

The destinations UC IPM names are agencies rather than pest control companies: the county agricultural commissioner, the county Cooperative Extension office, and USDA Wildlife Services by telephone and website. It notes that private nuisance wildlife control firms also operate in some areas. For reporting, it names local authorities including animal control and the state wildlife agency for aggression or an attack, and the county agricultural commissioner for livestock attacks.

Those are the destinations this page passes on, in the sources’ own terms. We attempted to read USDA’s own published material on predator damage management and could not: the requests failed at the transport level on every attempt across three separate rounds. Nothing on this page is drawn from USDA material, because we did not manage to read any. The agency is named here because UC IPM names it, not because we have read what it says.

What the Sources Say Does Not Work

Chemical repellents. UC IPM, in text stamped March 2007: “No chemical repellents are registered for use in repelling coyotes from property or from livestock.”

Sound and visual deterrent devices. The quotations are in the hazing section above. It is worth being precise about what that source claims: it does not say such devices never work. It says the effect is temporary, and that the disturbance makes them impractical in suburban areas.

Strobe-siren devices, propane cannons and scarecrows are named in the same passage as things coyotes adapt to in the absence of any real threat.

Questions

Sources

Every factual statement on this page traces to one of the four publications below. Where a source carries its own date, that date is given, and where a statement depends on it the date is repeated in the sentence itself. Three attempts to reach a California state wildlife agency page timed out, and every attempt to reach USDA material failed, so neither is cited here.