Posted on

Research Studies on Lost Pet Behavior and Recovery

Hiding cat

Unfortunately, there are few research studies on lost pet behavior.  This is a list of citations and abstracts of published research papers related to lost pet behavior and recovery.  There are links to a few of the papers.  You may be able to find other papers by visiting your local college or university and checking their online journal databases.

Huang, L., M. Coradini, J. Rand, J. Morton, K. Albrecht, B. Wasson, and D. Robertson.  2018.  Search methods used to locate missing cats and locations where missing cats are found.  Animals 8(1), 5.

Abstract:  Missing pet cats are often not found by their owners, with many being euthanized at shelters. This study aimed to describe times that lost cats were missing for, search methods associated with their recovery, locations where found and distances travelled. A retrospective case series was conducted where self-selected participants whose cat had gone missing provided data in an online questionnaire. Of the 1210 study cats, only 61% were found within one year, with 34% recovered alive by the owner within 7 days. Few cats were found alive after 90 days. There was evidence that physical searching increased the chance of finding the cat alive (p= 0.073), and 75% of cats were found within 500 m of the point of escape. Up to 75% of cats with outdoor access traveled 1609 m, further than the distance traveled by indoor-only cats (137 m; p < 0.001). Cats considered to be highly curious were more likely to be found inside someone else’s house compared to other personality types. These findings suggest that thorough physical searching is a useful strategy, and should be conducted within the first week after cats go missing. They also support further investigation into whether shelter, neuter and return programs improve the chance of owners recovering missing cats and decrease numbers of cats euthanized in shelters

Lord, L.K., W. Ingwerson, J.L. Gray and D.J. Wintz.  2009.  Characterization of animals with microchips entering animal shelters.  Journal of the American Veterinary Medical Association 235(2): 160-167.

Abstract:  Objective: To characterize animals with microchips entering animal shelters and the process used to find owners.  Design: Cross-sectional study.  7,704 microchipped animals entering 53 animal shelters between August 2007 and March 2008.  Procedures: Data for animals with microchips were recorded by participating animal shelters and reported monthly.  Results: Of 7,704 animals, strays accounted for slightly more than half (4,083 [53.0%]), with the remainder classified as owner-relinquished animals (3,225 [41.9%]) and other (396 [5.1%]). Of 3,425 stray animals for which animal shelters reported that the owner was found, a higher percentage of dog owners (2,191/2,956 [74.1%]) than cat owners (298/469 [63.5%]) was found. For 876 animals for which the owners could not be found, the main reasons were incorrect or disconnected telephone number (310 [35.4%]), owner did not return telephone calls or respond to a letter (213 [24.3%]), and animal was registered to another group (151 [17.2%]). Of 1,943 animals for which animal shelters contacted a microchip registry, 1,129 (58.1%) were registered in the database. Purebred neutered dogs whose owner information was in the shelter database registry or microchip registry had a higher likelihood that the owners would be found.  Conclusions and Clinical Relevance: The high rate for return of microchipped dogs and cats to their owners supported microchipping as a valuable permanent pet identification modality; however, issues related to registration undermined its overall potential. Bundling of microchip implantation and registration, point-of-implantation data registration, use of annual compliance and update reminders, and providing access to all registries are potential solutions.

Lord, L.K., T.E. Wittum, A.K. Ferketich, J.A. Funk, and P.J. Rajala-Schultz.  2007.  Search and identification methods that owners use to find a lost cat.  Journal of the American Veterinary Medical Association 230(2): 217-220.

Abstract:  Objective: To characterize the process by which owners search for lost cats and identify factors associated with time to recovery.  Design: Cross-sectional study.  Sample population:  Owners of 138 cats lost in Montgomery County, Ohio, between June 1 and September 30, 2005.  Procedures:  A telephone survey was conducted.  Results:  73 of the 138 (53%) cats were recovered; median time to recovery was 5 days (range, 0.5 to 81 days). Most cats (48 [66%]) that were recovered returned home on their own or were found in the neighborhood (5 [7%]); most other cats were recovered through posting of neighborhood signs (8 [11%]) or calling or visiting an animal agency (5 [7%]). The highest success rate for any of the search methods that were used was only 12% (posting neighborhood signs). Only 26 of the 138 (19%) cats had some type of identification at the time they were lost (ie, identification tag, rabies tag, or microchip). Owners allowed 82 (59%) cats to spend at least some time outdoors. The percentage of sexually intact cats recovered by their owners (4/16 [25%]) was significantly lower than the percentage of neutered cats recovered (69/122 [57%]).  Conclusions:  Results suggest that the percentage of lost cats recovered by their owners is low, possibly in part because of the lack of use of traditional identification methods and the general acceptance that cats may roam. Veterinarians can help educate owners about the importance of identification and the need to keep cats indoors.

Lord, L.K., T.E. Wittum, A.K. Ferketich, J.A. Funk, and P.J. Rajala-Schultz.  2007.  Search and identification methods that owners use to find a lost dog.  Journal of the American Veterinary Medical Association 230(2): 211-216.

Abstract: Objective: To characterize the process by which owners search for lost dogs and identify factors associated with time to recovery.  Design:  Cross-sectional study.  Sample Population:  Owners of 187 dogs lost in Montgomery County, Ohio, between June 1 and September 30, 2005.  Procedures:  A telephone survey was conducted.  Results:  132 of the 187 (71%) dogs were recovered; median time to recovery was 2 days (range, 0.5 to 21 days). Dogs were recovered primarily through a call or visit to an animal agency (46 [34.8%]), a dog license tag (24 [18.2%]), and posting of neighborhood signs (20 [15.2%]). Eighty-nine (48%) dogs had some type of identification at the time they were lost (ie, identification tag, dog license tag, rabies tag, or microchip). Owners had a higher likelihood of recovery when they called an animal agency (hazard ratio, 2.1), visited an animal agency (1.8), and posted neighborhood signs. Dogs that were wearing a dog license tag also had a higher likelihood of recovery (hazard ratio, 1.6). Owners were less likely to recover their dogs if they believed their dogs were stolen (hazard ratio, 0.3).  Conclusions:  Results suggest that various factors are associated with the likelihood that owners will recover a lost dog. Both animal agencies and veterinarians can play a role in educating dog owners on the importance of identification tags, licensing, and microchips and can help to emphasize the importance of having a search plan in case a dog is lost.

Lord, L.K., T.E. Wittum, A.K. Ferketich, J.A. Funk, and P.J. Rajala-Schultz.  2007. Search methods that people use to find owners of lost pets.  Journal of the American Veterinary Medical Association 230(12): 1835-1840.

Abstract:  Objective:  To characterize the process by which people who find lost pets search for the owners.  Design:  Cross-sectional study. Sample Population-188 individuals who found a lost pet in Dayton, Ohio, between March 1 and June 30, 2006. Procedures-Potential participants were identified as a result of contact with a local animal agency or placement of an advertisement in the local newspaper. A telephone survey was conducted to identify methods participants used to find the pets’ owners.  Results:  156 of 188 (83%) individuals completed the survey. Fifty-nine of the 156 (38%) pets were reunited with their owners; median time to reunification was 2 days (range, 0.5 to 45 days). Only 1 (3%) cat owner was found, compared with 58 (46%) dog owners. Pet owners were found as a result of information provided by an animal agency (25%), placement of a newspaper advertisement (24%), walking the neighborhood (19%), signs in the neighborhood (15%), information on a pet tag (10%), and other methods (7%). Most finders (87%) considered it extremely important to find the owner, yet only 13 (8%) initially surrendered the found pet to an animal agency. The primary reason people did not surrender found pets was fear of euthanasia (57%). Only 97 (62%) individuals were aware they could run a found-pet advertisement in the newspaper at no charge, and only 1 person who was unaware of the no-charge policy placed an advertisement.  Conclusions:  Veterinarians and shelters can help educate people who find lost pets about methods to search for the pets’ owners.

L.K. Lord, B. Griffin, M.R. Slater, and J.K. Levy.  2010.  Evaluation of collars and microchips for visual and permanent identification of pet cats.  Journal of the American Veterinary Medical Association. 

Abstract:  Objective:  To determine the percentage of pet cats still wearing collars and having functional microchips 6 months after application.  Design:  Randomized controlled clinical trial. Procedures:  538 client-owned cats were randomly assigned to wear 1 of 3 types of collars: plastic buckle, breakaway plastic buckle safety, and elastic stretch safety. Each cat was fitted with the assigned collar, and a microchip was inserted SC between the scapulae. Owners completed questionnaires about their experiences and expectations of collars at enrollment and at the conclusion of the study.  Results:  391 of the 538 (72.7%) cats successfully wore their collars for the entire 6-month study period. Owners’ initial expectations of the cats’ tolerance of the collar and the number of times the collar was reapplied on the cats’ necks were the most important factors predicting success. Type of collar likely influenced how often collars needed to be reapplied. Eighteen (3.3%) cats caught a forelimb in their collar or caught their collar on an object or in their mouth. Of the 478 microchips that were scanned at the conclusion of the study, 477 (99.8%) were functional.  Conclusions:  Most cats successfully wore their collars. Because even house cats can become lost, veterinarians should recommend that all cats wear identification collars since they are the most obvious means of identifying an owned pet. For some cats, collars may frequently come off and become lost; therefore, microchips are an important form of backup identification. Owners should select a collar that their cat will tolerate and should check it often to ensure a proper fit.

Weiss, E. M. Slater, and L. Lord.  2012.  Frequency of lost dogs and cats in the United States and the methods used to locate them.  Animals 2: 301-315.

Abstract:  A cross-sectional national random digit dial telephone interview was conducted between September and November 2010. There were 1,015 households that had owned a dog or cat within the past five years. Of these 817 households owned dogs and 506 owned cats. Fourteen percent of dogs (95% Confidence Interval (CI): 11–16%) and 15% (95% CI: 12–18%) of cats were lost in the past five years. No owner demographic variables were associated with losing a pet. Ninety three percent (95% CI: 86–97%) of dogs and 75% (95% CI: 64–85%) of cats were recovered. For dogs, searching the neighborhood and returning on their own were the most common methods of finding the dog; 14% were found through an identification tag. For cats, returning on their own was most common. Dogs were more likely than cats to be lost more than once. Cats were less likely than dogs to have any type of identification. Knowledge of the successful methods of finding dogs and cats can provide invaluable help for owners of lost pets. Since 25% of lost cats were not found, other methods of reuniting cats and their owners are needed. Collars and ID tags or humane trapping could be valuable approaches.

Posted on

Should We Catch this Dog?

I originally wrote this article as a response to a similarly titled article Should We Catch the Dog? by Mark Johnson, DVM, of Global Wildlife Resources.  The article and ensuing discussion were about whether we should always catch feral/roaming dogs or whether it is sometimes in the dog’s best interest to leave them free.  This is a rarely asked question and may not often be feasible, but it is an interesting discussion nonetheless.  I would suggest also checking out Mark’s article and the reader comments if this is an issue that you are facing.

I run a business where I help people find their missing cats and dogs, and sometimes this involves catching a frightened lost dog. It is actually surprising, especially to the panic-stricken and bewildered owner, how often a beloved pet dog will run in fear even from their owner once lost and in an unfamiliar environment. This is commonly referred to as the dog being in “survival mode,” and I have worked several cases (and heard of many more) where a lost dog acted feral until it was captured. Some revert to their normal friendly selves while still in the trap as soon as they see their owner, while others need to be brought home to relax in a familiar environment before acting friendly and social again. I have also seen similar behavior in lost or stray dogs where the owner was not present. While loose, they acted extremely fearful and untrusting of people, but once trapped and brought indoors, they turned out to be friendly, well-socialized dogs.

Lost dog at pre-baited trap in Vermont.
Lost dog at pre-baited trap in Vermont.

One such case stands out in my mind.  I received a call from a Tennessee rescue group that had lost a foster dog in Vermont.  He was a pug mix and had slipped his leash and run away upon arrival at his foster home.  I was not contacted until one month after he went missing, but I located him within three weeks using 30 large florescent posters put up over a six-mile radius from where he went missing.  It turned out that he was living near a small town about four miles from where he escaped, and a local woman had been putting out food for him but couldn’t get anywhere near him.  We used a humane trap baited with food and blankets mailed up from his original foster mom in Tennessee.  He appeared subdued, but not outwardly friendly once trapped.  However, when we carried the trap into the house and let him out, he transformed into a different dog.  He ran from person to person and jumped into laps and appeared ecstatic to be back in a home after seven long weeks living in the woods. Here’s example of a similar story where the dog was loose for years before being caught: Ghost Dog Finally Makes It Home.

Lost dog caught on wireless wildlife camera entering an enclosure trap.
Lost dog caught on wireless wildlife camera entering an enclosure trap.

On the other hand, I worked another case where the outcome was very different. I was assisting another lost pet tracker in a case where the lost dog was a former street dog from the south, and he escaped upon arrival at his new home in New England. He was actually caught a few weeks later, but then again escaped his new owners within one week. Even when indoors at his new home, this dog was extremely fearful of people and had a reported history of biting in fear. It took several months and many trapping attempts before he was finally captured, and despite being on his own living in the snow and cold, the Animal Control Officer thought he looked remarkably healthy. In a very unfortunate series of events, the dog ended up seriously biting someone shortly after his capture. Now he is being held indefinitely by animal control, his new owners did not reclaim him (for complicated reasons so please don’t judge), and his future is uncertain.  I don’t think there really was the option to leave him running loose where he was, but the outcome was really disheartening.  (Update: I originally wrote this article in 2014.  This dog lived for 1 ½ years as a foster dog before passing away due to an illness.  He was described as a productive shelter volunteer that helped evaluate and befriend new shelter dogs, and no additional incidents of biting are known.)

The real question is probably not whether this dog should have been captured once lost in New England, but whether it was in his best interest to be caught down south in the first place and transported up here as an adoptable dog. I don’t know enough about these programs to provide a critical argument, but it does raise some interesting questions.  Do feral or severely under-socialized dogs that are rescued down south and then transported up north really benefit from the situation?  It is not uncommon for these dogs to run off shortly after being adopted, and if found and caught, they remain at a high risk of taking off again at the first opportunity.  Are there other options for these dogs besides killing them or adopting them out?  Are trap-neuter-return (TNR) programs for dogs ever a viable option in the United States?

Obviously this is a complicated issue that would need to be addressed on a case-by-case basis.  Personally, I think in some cases (if the option exists) that an extremely fearful or feral dog might be better off left alone to live on their own similar to TNR colonies of feral cats. However, I would caution people that it is not possible to assess the temperament of a loose dog until it has been trapped and allowed to feel safe indoors. Unfortunately, this leads to the quandary that in order to determine if you should catch a dog, especially one where the owners cannot be located and you don’t know the dog’s history, you really need to catch the dog first.

– Danielle Robertson, Lost Pet Research & Recovery


Posted on

Kyon Smart Collar: Never Lose Your Pet Again

Kyon Collar Photo 1

Owning a pet is a great pleasure. If only we knew what they are thinking and feeling! And if only we could ensure that they would never get lost! Pets deserve our attention as they have needs and feelings. Wouldn’t it be ideal if they could communicate with us? So, what if we told you there is a pet collar which gives your pet a voice?

KYON smart collar is a multifunctional pet tracker that helps you communicate with your pet and simultaneously protects it from getting lost.

KYON is a stylish wearable device of highly advanced technology supporting a GPS and a GSM technology accompanied by an LED display where you can project the messages you want. Along with a Basestation and an easy-to-use mobile application that audits and receives alerts or info regarding your pets’ condition, KYON brings you closer to your best friend.

KYON smart collar tells you the exact location of your pet while it notifies you if your pet runs away from a specific distance approximately 300ft, a “safe zone” which can be extended by adding a SIM card for an extra $4.99 per month.

Kyon Collar Photo 2

With its Sense technology, the KYON collar lets pet owners feel their pets’ mood. For example, there is an embedded 9- axis accelerometer which updates you about your pet’s activity levels (if feeling happy or sleepy) when you are not at home. Moreover, KYON collar has a heat sensor which informs you whether your pet is feeling too hot or too cold. It holds as well a water sensor that operates like a “lifeguard” by notifying the owner when it detects water.

KYON smart collar, offers some additional exciting features which along with the aforementioned guarantee that it is the smartest collar there is. The collar helps you avoid dangerous dog fights with its pacifying technology that uses a high frequency sound that calms your pet. Tip: When two pets of contradictory breeds are both wearing a KYON collar, then the pacifier feature is activated automatically. Barking can be sometimes risky and annoying. KYON prevents your beloved “yeller” from barking with its “Shhh…!” feature.

In addition, with KYON you can also in-flight monitor you pet and thanks to KYON’s vet appointment and annual vaccination as well as walk reminder your pet’s needs won’t be forgotten.

KYON has a 30 day life rechargeable battery (depends on usage).  It is designed for pets with 25 to 50cm neck perimeter and its leash clip can support over 30 kg and it operates in 120 countries worldwide. Details, finally, can make the difference, hence KYON smart collar is offered in a variety of colors in order to make the choice you prefer.

KYON pet collar is a KICKSTARTER project which has launched its campaign in March and aims to be funded by the 15th of May 2016 (5:35 PM +03:00). If you want to be part of this initiative, then, don’t hesitate: Make your pet the perfect gift!

Guest blog article written by Fenny Chroni

Posted on

When NOT to Use a Tracking Dog to Find a Lost Dog

Running Dog - Morgue FileThe idea of using a tracking dog to find a lost dog is very compelling, but most people who pursue this option do not have a good understanding of how a tracking (or trailing) dog works.  In some cases a tracking dog CAN provide useful information for locating a lost dog such as confirming sightings or establishing a direction of travel.  However, very few lost dogs are actually found and captured during the search (i.e. a “walk-up find”), which is what most people are hoping for when they hire a tracking dog team.

What many people do not consider is that there are actually some cases when you should NOT try to use a tracking dog to find a lost dog.  In these situations a tracking dog is not only a waste of money, but they can actually be detrimental to finding and catching the lost dog.  The situations where you should not use a tracking dog to find a lost dog include most cases where there are multiple sightings of the lost dog in a general area, and the dog is running in fear from everyone.  This most often occurs with newly adopted dogs and skittish lost dogs.  However, even an otherwise friendly dog can enter what is known as “survival mode” (where they run from all people including those that they know) if they are lost in a frightening situation (such as a car crash) or if they are on the run for several days, especially if people attempt to chase or capture them.  Sometimes these lost dogs will run for several miles (1-5 is common and 10 or more miles is not unheard of), but in most cases the lost dog will eventually settle down in a place where they feel safe.  Generally this safe place is somewhere with food, water, shelter, and (very importantly) where people are not attempting to approach or catch them.  In some cases the lost dog will actually circle around and come back to close to where they went missing.

If you you get multiple sighting (even 2-3) of the lost dog in a general area (hopefully less than 1 mile apart), then the lost dog has likely found a safe place to hide out.  The last thing that you want to do in this situation is chase the dog out of his newly found haven.  If you use a tracking dog, they may help you find out where your dog has been taking shelter and getting food, but in the process you may scare your dog out of the safe place.  Likewise, it is a very bad idea to have human search teams go into this area and look for the lost dog, especially if it is a wooded area.  Even if they see the dog, they are most likely going to scare him out of the area.  In either of these situations, the lost dog may feel pressured to leave the area and find a new safe place, perhaps miles away.

In these types of cases, it is very important to leave the dog alone and encourage others to report sightings, but not to approach or attempt to catch the dog.  Most of these dogs are ultimately caught using lure and capture techniques such as feeding stations, calming signals, surveillance cameras and/or humane traps.   If your lost dogs fits this profile, you may still want professional advice and/or assistance in catching them from a trained pet detective.  If your dog does not fit this profile, then read this accompanying article on Search Dogs to determine if a tracking dog team could help you find your missing dog.

Lost Pet Research & Recovery offers phone consultations throughout the United States and on-site services in Massachusetts, Connecticut, Rhode Island and parts of Vermont, New Hampshire, Maine and New York.  For more information, visit: Find a Lost Dog.  If you are looking for on-site assistance outside of this area, then check the Missing Pet Partnership Pet Detective Directory.

Posted on

Regional Risk of Coyote Attacks on Pets

I have written several articles on coyotes killing cats and one on coyotes attacking dogs.

Coyote Pack in Yellowstone National Park
Coyote Pack in Yellowstone National Park in a parking lot near Tower Ranger Station

Most of these articles dispute the pervasive belief that the majority of missing cats and small dogs have become food for a coyote.  I believe that far more pets, especially cats, are never found because their owners do not know how to effectively search or give up too soon (often because their friends and neighbors convince them that their cat was killed by a predator).  In many locations a missing cat is much more likely to have become displaced (wandered or scared away from its home), become trapped (such as in a crawl space) or be hit by a car.  For more information see:

This is not to say that I believe that coyotes never kill cats.  Coyotes are an opportunistic and highly adaptable predator, and they will certainly kill cats on occasion.  In fact, in some parts of their geographic range, coyotes kill and probably eat a lot of cats.  

There is unfortunately very little research available on the occurrence of coyote attacks on pets, especially on a national level.  However, there is a fair amount of research on coyote attacks on humans, and the results from these studies suggest that far more cats and dogs are killed by coyotes in California than in any other state/province (Baker and Timm 1998; Timm et al. 2004; Timm 2006; White and Gehrt 2009).  The largest number of incidents occurred in Los Angeles, Orange, and San Diego Counties.  Other states/provinces that may show higher rates of coyote attacks on pets include Arizona, Nevada, British Columbia and Alberta.

The results from several studies suggest that far more cats and dogs are killed by coyotes in California than in any other state/province.  Other states/provinces that may show higher rates of coyote attacks on pets include Arizona, Nevada, British Columbia and Alberta.

Baker and Timm (1998) found that there is a predictable sequence of observed changes in coyote behavior that indicate an increasing risk to human safety.  Based on additional research, Timm et al. (2004) defined these changes in order of their pattern of occurrence as follows:

  1. An increase in observing coyotes on streets and in yards at night.
  2. An increase in coyotes approaching adults and/or taking pets at night.
  3. Early morning and late afternoon daylight observance of coyotes on streets and in parks and yards.
  4. Daylight observance of coyotes chasing or taking pets.
  5. Coyotes attacking and taking pets on leash or in close proximity to their owners; coyotes chasing joggers, bicyclists, and other adults.
  6. Coyotes seen in and around children’s play areas, school grounds, and parks in mid-day.
  7. Coyotes acting aggressively towards adults during mid-day.

Since several of these predictors of coyote attacks on humans include attacks on pets, we should be able to safely conclude that states/provinces with more attacks on humans will likely have many more attacks on pets.  For example, Baker and Timm (1998) found that the following incidents preceded attacks on humans in southern California.

  • In the three weeks prior to three children being approached and bitten by coyotes (in separate incidents), USDA-ADC personnel received 30-40 complaints of coyotes attacking or killing household pets, or approaching people during daylight hours.  (San Diego County 1998)
  • Prior to an attack on a poodle snatched from his owner’s arms, coyotes had been seen in early and late mornings chasing and killing cats and rabbits in the neighborhood.  (Orange County 1991)
  • The attack on a child was preceded by 3-4 weeks of coyote attacks on two dogs and six house cats as reported to animal control.  (Orange County 1992)
  • Two to three months prior to the first attack on children, cat remains were found numerous times on the college campus.  Coyotes were also seen chasing and carrying off cats at night and early in the morning.  (Riverside County 1995)

There is some disagreement on what constitutes a coyote attack (Fox 2006; Webster 2007), so data from different studies do not report the same number of attacks per location, but all studies found high numbers of attacks in California. White and Gehrt (2009) found reports of 70 (49%) attacks in California followed distantly by 18 attacks (13%) in Arizona from 1960-2006 (Figure 1).

Figure 1. Geographic distribution of coyote attacks in the United States and Canada during 1960-2006. White and Gehrt. 2009. Coyote Attacks on Humans in the United States and Canada. Human Dimensions of Wildlife 14: 419-432.
Figure 1. Geographic distribution of coyote attacks in the United States and Canada during 1960-2006. (White and Gehrt. 2009. Coyote Attacks on Humans in the United States and Canada. Human Dimensions of Wildlife 14: 419-432.)
Table 1. Number of coyote attacks by state or province, in which physical contact occurred from 1988-2006. (Timm. 2006. Coyotes Nipping At Our Heels: A New Suburban Dilemma. 11th Triennial National Wildlife & Fisheries Extension Specialists Conference. University of Nebraska - Lincoln. 139-145.
Table 1. Number of coyote attacks by state or province, in which physical contact occurred from 1988-2006. (Adapted from Timm. 2006. Coyotes Nipping At Our Heels: A New Suburban Dilemma. 11th Triennial National Wildlife & Fisheries Extension Specialists Conference. University of Nebraska – Lincoln. 139-145.

Timm and Baker (2004) reported 89 coyote incidents in California from 1978-2003, and Timm (2006) reported more than 160 human safety incidents in California involving coyotes since the early 1970s (Table 1).  Incidents of coyotes with rabies were excluded from this analysis.  Like White and Gehrt (2009), Timm (2006) also found that Arizona had a higher incidence of attacks, but found that Nevada and British Columbia were higher rather than Alberta.

One study on coyote predation on cats in southern California also supports a higher incidence of attacks on cats in this region of the country (Crooks and Soule 1999).  They found that 21% of 219 coyote scats contained cat remains.  Most studies on coyote diets found that cat remains were only present in 1-2% of coyote scats (including other studies in southern California and Arizona), with some as high as 6.7% (Schaumburg, Illinois) to 13.1% (western Washington) (see “Coyotes Don’t Eat Cats Very Often“).  Crooks and Soule (1999) also found that 25% of radio-collared cats in the study were killed by coyotes, and 42% of cat owners reported that coyotes had attacked or killed their cats.  Another study in Arizona found that cats constituted 42% of the diet of one pack of coyotes (Grubbs and Krausman 2009).

Even in states that don’t show a large number of human attacks (and therefore pet attacks), there may be smaller regions (e.g. even a single section of a town) where more cats and/or small dogs are being attacked by coyotes.  You may be able to determine the prevalence of coyote attacks in your area by doing an internet search for news reports or police logs of coyote attacks on pets or by calling your local Animal Control Officer or even talking to your neighbors.  However, be skeptical of all reports where a coyote attack is assumed just because a cat or small dog went missing and the event wasn’t witnessed and no remains were found.  According to Timm’s et al. (2004) sequence of changes in coyote behavior that indicate an increased risk to human safety, frequent reports of daylight observations of coyotes may be another warning sign.  See  “Understanding Coyote Behavior in Urban/Suburban Areas and Assessing Risks to Cats” to assess the risk of a missing cat being killed by coyotes.

Baker and Timm (1998) speculate that the coyotes’ “recent adaptation to urban and suburban habitats in places such as southern California has taken place over several generations, and such adaptation may involve learned behaviors passed from parent to offspring.” Based on dietary studies of coyotes, cats are not generally a common prey item, and preying on cats may be a specialized or learned behavior.  If this were the case, then we would only expect to see high rates of attacks on cats in certain regions.  In a small area such as town, this could be the result of a single coyote that has learned to specialize in killing cats.  However, it is possible for a solitary coyote or pack to have a home range that encompasses several towns (Fox 2006), which could lead to a larger area of attacks by a single coyote.  This type of issue would be similar to “problem individuals” of wolves or coyotes killing livestock (Linnell et al. 1999).  When seen in a larger area such as regions of southern California, the cat killing behavior is likely an adaption to urban life and is probably being taught to successive generations of coyotes.

Literature Researched

Baker, R.O. and R.M. Timm.  1998.  Management of Conflicts Between Urban Coyotes and Humans in Southern California.  Proceedings of the 18th Annual Vertebrate Pest Conference.  University of Nebraska – Lincoln.  299-312.

Crooks, K.R. and M.E. Soule.  1999.  Mesopredator release and avifaunal extinctions in a fragmented system.  Nature 400: 563-566.

Fox, C.H.  2006.  Coyotes and Humans: Can We Coexist?  Proceedings of the 22nd Annual Vertebrate Pest Conference.  University of California – Davis.  287-293.

Gehrt, Stanley and Seth P.D. Riley.  2010.  Coyotes (Canis latrans)” in Urban Carnivores: Ecology, Conflict, and Conservation. Gehrt, Stanley, D., Seth P.D. Riley, and Brian L. Cypher, editors.  The John Hopkins University Press.

Gehrt, S.D., C. Anchor, and L.A. White.  2009.  Home Range and Landscape Use of Coyotes in a Metropolitan Landscape: Conflict or Coexistence?  Journal of Mammalogy 90(5): 1045-1057.

Gehrt, Stanley.  2006.  Urban coyote ecology and management – The Cook County,Illinois coyote project.  Ohio State University Extension Bulletin 929.  32 pp.

Gompper, M.E.  2002.   Top Carnivores in the Suburbs?  Ecological and Conservation Issues Raised by the Colonization of North-eastern North America by Coyotes.  BioScience 52(2): 185-190.

Grubbs, Shannon E. and Paul R. Krausman.  2009.  Observations of Coyote – Cat Interactions.  Journal of Wildlife Management 73(5): 683-685.

Howell, R.G.  1982.  The Urban Coyote Problem in Los Angeles County.  Proceedings of the 10th Annual Vertebrate Pest Conference.  University of Nebraska – Lincoln.  21-23.

Linnell, J.D.C., J. Odden, M.E. Smith, R. Aanes, and J.E. Swenson.  1999.  Large carnivores that kill livestock: do “problem individuals” really exist?  Wildlife Society Bulletin 27(3): 698-705.

Timm, R.M.  2006.  Coyotes Nipping At Our Heels: A New Suburban Dilemma.  11th Triennial National Wildlife & Fisheries Extension Specialists Conference.  University of Nebraska – Lincoln.  139-145.

Timm, R.M. and R.O. Baker.  2004.  Coyote Attacks: An Increasing Suburban Problem.  Proceedings of the 21st Vertebrate Pest Conference.  University of Nebraska – Lincoln.  47-57.

Webster, J.C.  2007.  Missing Cats, Stray Coyotes: One Citizen’s Perspective.  Wildlife Damage Management Conferences Proceedings.  University of Nebraska – Lincoln.  74-116.

White, L.A. and S.D. Gehrt.  2009.  Coyote Attacks on Humans in the United States and Canada.  Human Dimensions of Wildlife 14: 419-432.


Posted on

Comparison of Dog GPS Trackers

Tagg the pet tracker (

There are basically two different kinds of GPS trackers for dogs.  The more popular design uses a GPS tracker, which is attached to your dog’s collar, and requires a monthly subscription plan to actually track your dog’s location.  Many of these allow you to place a virtual boundary around your yard, and if your dog leaves this area, a text or email alert is sent to you.  If your dog goes missing, you can track their location with texts to your cell phone or maps on a smartphone or computer.

Dog GPS collars available in the US

The other design uses a GPS tracker, which is usually built into a specially designed collar, and has a portable receiver.  The receiver provides you with either a map (e.g. Garmin Astro) or just an arrow with distance (e.g. RoamEO), and you use this to track down your dog.  The receivers have a maximum distance of 7-9 miles and perhaps a minimum distance of 1 mile.  No activation fee or subscription is necessary.  The shortcomings of this design are the limited range and short battery life, and you always have to carry the receiver on you to find your dog.

There are a few companies in the UK such as PawTrax Snooper and Loc8tor Pet that have developed a pay per use GPS tracker.  These are similar in design to the GPS trackers that require a monthly fee, but instead you prepay for a certain number of locations.  PawTrax may now be available for purchase in the US.  Email for more information.

For a more complete list of all the GPS trackers currently available for dogs and cats, check out this Comparison of GPS Pet Trackers.  If you know of any that I have missed, please let me know.  In February I purchased Tagg the Pet Tracker for my own dog, and I plan to write a review on this particular tracker.  Overall, I am happy with the purchase and it seems to work quite well though I have fortunately never had the chance to try it out for real.

Posted on

Drop Net Trap Designs

A drop net consists of a large net, which is often hung from poles, and suspended over a baited target area.  The trap must be monitored and when the target animal is under the net, it is either manually or remotely triggered, and drops on the animal.  Drop nets were first used to capture game birds and are now widely used by wildlife biologists to catch deer and other ungulates.  They have also been used to try and capture extremely trap-shy lost or stray dogs.

Here is a video of a drop net available from Wildlife Their drop net uses a radio-controlled electromagnetic trigger, and a 25′ x 25′ drop net costs $3,950.  They do also offer rentals, but prices are not provided.

Commercial drop net costs are prohibitive to most pet detectives, rescue workers, and owners who might need one to capture a trap-shy lost or stray dog.  However, I was able to find several drop net trap designs in scientific research journals that are relatively inexpensive and might not be that hard to build.

Manual Trigger Drop Net Designs

Deer drop nets where originally built with an explosive tripping mechanism (Ramsey 1968), and some more recent drop nets use radio-controlled release of electromagnets.  However, I wanted to start by looking at some simple manually triggered drop nets, which might be easier to build for the average person.  Honestly, I found most of the instructions at least somewhat confusing, and I may attempt to contact the researchers for more detailed instructions.  However, these instructions should at least provide useful brainstorming ideas for designing your own drop net.

Here’s a comparison of three manually triggered drop net designs.

A comparison of manually triggered drop net designs

Lopez, R.R., N.J. Silvy, J.D. Sebesta, S.D. Higgs, M.W. Salazar.  1998.  A Portable Drop Net for Capturing Urban DeerProceedings Annual Conference of Southeastern Association of Fish and Wildlife Agencies 52: 206-209.

Lopez et al.’s (1998) drop net design is the least expensive and easiest to set up.  The trap frame is portable and constructed of chain link fence top rail tubing with welded corner posts.  The net is 5.2 m x 5.2 m, which is held up by tension using a perimeter rope and washers.  The trigger mechanism is a piece of bailing twine tied to a pull-pin and washer mechanism.  For construction directions, see Lopez et al.’s (1998) Methods.

Lopez et al. 1998. A Portable Drop Net for Capturing Urban Deer. Proc. Annu. Conf. Southeast. Assoc. Fish and Wildl. Agencies 52: 206-209

D’Eon, R. G., G. Pavan, and P. Lindgren.  2003. A Small Drop-Net versus Clover Trap for Capturing Mule Deer in Southeastern British Columbia.  Northwest Science 77(2): 178-181.

D’Eon et al.’s (2003) design has the advantage that it can be constructed and set up by a single person.  The trap is meant to be suspended 1-2 m from the ground between two large trees.  The net is 6.1 m x 6.1 m with four sections of galvanized pipe attached to either end for weight.   A 60 m release cord is used to trigger the net by pivoting two additional sections of galvanized pipe.  For design instructions, see D’Eon et al.’s (2003) Methods.

Figure 1. D'Eon et al. 2003. A Small Drop-Net versus Clover Traps for Capturing Mule Deer in Southeastern British Columbia. Northwest Science 77(2): 178-181.

D’Eon et al.’s (2003) success rate (75%) was considerably lower than the other two designs.  However, they only attempted to catch four deer, and the first one escaped when they suspended the net 2 m above the ground.  When they moved the net to 1 m above the ground, they caught three additional deer in two trapping attempts.

Jedrzejewski, W. and J.F. Kamler.  2004.  From the Field: modified drop-net for capturing ungulatesWildlife Society Bulletin 32(4): 1305-1308.

Jedrzejewski and Kamler’s (2004) drop net design is more labor intensive to set up and costs significantly more than the other designs (though even this net is far cheaper than the commercially available ones).  However, this net is also three to four times the size of the other nets since it was being used to capture the much larger red deer.  Modifications of the net set up and trigger mechanism could probably be used on a much smaller net though the success rate would most likely decrease.

Jedrzejewski, W. and J.F. Kamler. 2004. From the Field: modified drop-net for capturing ungulates. Wildlife Society Bulletin 32(4): 1305-1308.

This drop net was built on a semi-permanent frame constructed of wooden poles sunk into the ground and braced with metal cables to support the weight of the net.  A pulley was attached to the cable at the top of each wooden pole.  Nylon ropes were then attached to the net near each pole, and the ropes were run through the nearest pulley.  All the ropes were tied to a single metal ring at the trigger mechanism with ropes on the far end coming across the top of the net.  The trigger mechanism is simply a hook through a metal ring, which releases when another rope is pulled.  This might sound complicated, but it will make a lot more sense when you look at the diagrams.  For complete instructions, see Jedrzejewski and Kamler’s (2004) Trap description and construction. 

Jedrzejewski, W. and J.F. Kamler. 2004. From the Field: modified drop-net for capturing ungulates. Wildlife Society Bulletin 32(4): 1305-1308.

Remote Trigger Drop Net Designs

For the more tech savvy people out there, a remote controlled drop net is another option.  Lockowandt (1993) wrote a paper on how to build “An Electromagnetic Trigger for Drop-Nets.” Unfortunately, I haven’t been able to find a copy of this paper available on the internet, but you could get a copy from your local University (like I did). Lockowandt (1993) estimated that the electromagnetic trigger system cost $980 and it took three days to build. Missing Pet Partnership was able to get instructions on building an electromagnetic drop net from their local Fish & Wildlife Agency.  See their case on Catching Sophie.

A somewhat different remote net trap was designed by Dematteis et al. (2010). This net trap is also activated by remote radio-controlled electromagnets, but it is an up-net rather than a drop net. With an up-net, the sides of the net enclosure lie on the ground, and when the net is triggered, the sides of the net are raised up creating an enclosure trap. Dematteis et al. (2010) reports that the up-net had a 96.2% (50/52) success rate when used to capture chamois (a goat-antelope species native to mountains in Europe) in their study. The estimated cost of their up-net enclosure trap was $4,617, making it cost prohibitive to duplicate. However, modifications of their design could be used to create a smaller, less expensive net trap.

If anyone has already built a small portable drop net that might be used to help capture skittish lost dogs, I would love to hear about it, especially if you are willing to share you designs with others.

Literature Researched

D’Eon, R. G., G. Pavan, and P. Lindgren.  2003. A Small Drop-Net versus Clover Trap for Capturing Mule Deer in Southeastern British Columbia.  Northwest Science 77(2): 178-181.

Dematteis, A., M. Giovo, F. Rostagno, O. Giordano, D. Finn, A. Menzano, P. Tizzani, G. Ficetto, L. Rossi, and P.G. Menguz.  2010.  Radio-controlled up-net enclosure to capture free-ranging Alpine chamois Rupicapra rupicapra.  European Journal of Wildlife Research 56: 535-539.

Jedrzejewski, W. and J.F. Kamler.  2004.  From the Field: modified drop-net for capturing ungulatesWildlife Society Bulletin 32(4): 1305-1308.

Lockowandt, S.P.E.  1993.  An Electromagnetic Trigger for Drop-Nets.  Wildlife Society Bulletin 21: 140-142.

Lopez, R.R., N.J. Silvy, J.D. Sebesta, S.D. Higgs, M.W. Salazar.  1998.  A Portable Drop Net for Capturing Urban DeerProceedings Annual Conference of Southeastern Association of Fish and Wildlife Agencies 52: 206-209.

Ramsey, C. W.  1968.  A Drop-Net Deer Trap.  Journal of Wildlife Management 32(1): 187-190.

Posted on

Deer Traps to Catch Skittish Lost Dogs?

Most people are probably not even aware that there are such a thing as deer traps, but wildlife biologists use a variety of traps to capture deer including box traps, Clover traps (a netted-box trap), drop nets, and drive nets..  Deer, like many trap-shy dogs, are generally not comfortable entering enclosed spaces, so the design of some of these traps might be useful for capturing shy dogs as well.  These are mostly just ideas at this point since I don’t yet have experience using these traps and some may not prove to be practical.  However, I wanted to share the variety of trap designs available in scientific research journals.  In this article, I’m going to focus on Clover traps and box traps, and next I’ll look at designs for net traps.

Clover Trap

The Clover trap is what first caught my attention when I was trying to think of a way to improve dog enclosure traps.  The Clover trap, which is named after its designer, is a netted-box trap.  Here is a demo video of a Clover trap.  You really only need to watch the first 2 minutes of the video, and I would not suggest wrestling any trapped dogs to subdue them.

Clover Trap
Close up of guillotine door mechanism

I do think this trap would require some modifications to securely trap a dog, but it does offer some interesting possibilities.  The netting makes the trap appear much more open than the traditional wire cage trap.  On the downside you probably would need to supervise the trap closely because a dog could likely chew through the netting.  The trip-wire mechanism might also need to be modified.  However, a very useful feature of this trap is that it doesn’t require the dog to duck under the door to enter the trap.  Many dogs (even non-trap-shy dogs) are somewhat leery of the overhanging door on many cage traps.  The Clover trap has a netted guillotine door that rolls up at the top of the trap.

Enclosure Trap designed by On Track Pet Tracking (

This type of netted-guillotine door might also be modified to attach to a larger dog enclosure trap like this one.  This enclosure trap is meant to be monitored and manually triggered when the dog enters, so chewing through the netting wouldn’t be a concern.  This particular trap uses a large wire-fence door, which makes the trap very comfortable to enter, but the door takes up more space within the trap and is slower to close than a guillotine door.  Here’s a video of my dog being caught in the trap.  Other enclosure traps often use a solid guillotine door, which can be intimidating for the dog to walk under.

For design instructions for the Clover Trap see VerCauteren’s (1999) research paper.  Clover traps can also be bought from online suppliers such as Wildlife (sold as parts or as complete traps).

Stephenson Box Trap

Stephenson box trap

The original deer box trap is the Stephenson box trap, which is basically a large wooden box with two guillotine type doors so it looks like a tunnel.  While it is an enclosed space, the sheer size of the trap (approximately 10 ft L x 4 ft W x 9 ft H) might make them comfortable for many dogs to enter.  These traps would be useful in a location where constant monitoring is not possible since they are very secure and offer some protection from adverse weather.  Anderson and Nielsen (2002) provide an updated design and instructions for anyone who might be interested in exploring this option further.

Literature Researched

Anderson, R. G. and C. K. Nielsen.  2002.  Modified Stephenson trap for capturing deer.  Wildlife Society Bulletin 30(2): 606-608.

Clover, M.R.  1954.  A portable deer trap and catch-net.  California Fish & Game 40: 367-373.

Thompson, M.J.K., R.E. Henderson, R.O. Lemke, and B.A. Sterling.  1989.  Evaluation of a collapsible clover trap for elk.  Wildlife Society Bulletin 17: 287-290.

VerCauteren, K., J. Beringer, and S. Hygnstrom.  1997.  Use of netted-cage traps in population management and research of urban white-tailed deerGreat Plains Wildlife Damage Control Workshop Proceedings.  Paper 380.

VerCauteren, K., J. Beringer, and S.Hygnstrom.  1999.  Use of netted cage traps for capturing white-tailed deerUSDA National Wildlife Research Center – Staff Publications. Paper 833.

Posted on

Trends in Coyote Attacks on Dogs

The Cooke County, Illinois, Coyote Project provides some interesting research on observations of coyote attacks on dogs in the Chicago metropolitan area.  Using a search of newspaper databases from 1990-2004, they found 70 articles of coyote attacks on dogs.  The average number of attacks per year increased from 0-2/yr in 1990 up to 6-14/yr in 2005.  Most attacks (60%) were on small breed dogs such as Yorkshire Terriers, Shih Tzus and Jack Russell Terriers, and these attacks were frequently fatal.  Coyote attacks on dogs peaked in December – February during the coyote mating season and again during April when the pups are usually born (Gehrt 2010).  They concluded that these patterns are consistent with resident, territorial coyote attacks rather than solitary individuals.  Attacks on larger dogs usually involved two or more coyotes and were most likely to occur during the mating season, when coyotes are most territorial.  In contrast, attacks on small dogs were likely to occur at any time of the year (Gehrt 2010).  Gehrt and Riley (2010) conclude that this pattern indicates that these attacks result from competition as well as predation.  In interspecific (between species) competition, larger carnivores such as wolves will opportunistically kill smaller carnivores such as coyotes and foxes.

Table 1. Coyote attacks on dogs based on size of the dog (from newspaper accounts of 70 attacks in the Chicago area from 1990-2005).
Table 2. Coyote attacks on dogs based on breed (from newspaper accounts of 70 attacks in the Chicago area from 1990-2005).

Since these results are only from one location and were based on newspaper articles, some care should be taken in generalizing the results.  For instance, the larger number of Yorkshire Terriers and Shih Tzus being attacked, does not necessarily mean that coyotes are more likely to attack these particular small breeds.  There may just be more of these breeds present in the Chicago area.

I would also like to point out that none of these results were actually published in the scientific literature.  In fact, I wasn’t able to find a single research study on coyotes attacking dogs, although I did (somewhat ironically) find one article published on dogs attacking a coyote.  I compiled these results from the Cooke County, Illinois, Coyote Project website, their research bulletin (Gerht 2006), and from a brief summary of the results in Urban Carnivores (Gehrt and Riley 2010).  However, in doing so, I did find some inconsistencies in the data.  For example, the website (where I got most of my numbers) indicated that they found 70 attacks from 1990-2004, but Gehrt and Riley (2010) state that there were 60 attacks from 1990-2007 (obviously they can’t both be right).  I also found that some of the percents in the breed pie chart (on the website) did not match the numbers in the text.  Therefore, Table 1 & 2 are my best interpretation of what I found and may not be entirely accurate.

Literature Cited

The Cooke County, Illinois, Coyote Project website

Gehrt, Stanley.  2006.  Urban coyote ecology and management – The Cook County,Illinois coyote project.  Ohio State University Extension Bulletin 929.  32 pp.

Gehrt, Stanley and Seth P.D. Riley.  2010.  Coyotes (Canis latrans)” in Urban Carnivores: Ecology, Conflict, and Conservation. Gehrt, Stanley, D., Seth P.D. Riley, and Brian L. Cypher, editors.  The John Hopkins University Press.

Updated: January 11, 2014