When we started in the commercial sheep business over 15 years ago, we knew we wanted to use livestock guardian dogs (LGDs) to protect our sheep from predators. LGDs were not as common then in the foothills as they are today, so our choices were somewhat limited - and my knowledge of these dogs was even more limited. One LGD puppy looked much like another (white and fuzzy) - and while I knew enough to pick a pup from working stock, I didn't know the questions I should be asking - or even what I should be looking for in terms of behavior. While we were lucky enough to pick up an older dog who turned out to be a decent protector, our record of success in our early years was mixed at best.
As we gained more experience using LGDs, we started to look for specific traits in new dogs. And we started to realize the importance of appropriate bonding and early-life "training" (I use the word "training" here differently than I might use it with respect to a herding dog - training a LGD doesn't necessarily involve teach a dog specific commands). As I've gained more experience and insight, our record of success has improved.
Over the last six months, I've been collaborating with extension colleagues in California and elsewhere to increase our understanding about what makes a solid livestock guardian dog. Carolyn Whitesell, who is the new human-wildlife interactions advisor with UC Cooperative Extension in the Bay Area, has experience working with LGDs in Africa. Bill Costanzo, who comes from a California sheep background, is a LGD extension specialist with Texas A&M. We've worked to come up with a new fact sheet on selecting the right LGD puppy. You can download it here.
Over the coming months, we hope to produce a series of fact sheets on caring for your LGD, as well as on bonding techniques and problem-solving. Carolyn and I will also be surveying producers about their techniques for bonding LGDs with livestock. Stay tuned!
In the meantime, you may want to check out my Flying Mule Dogs channel on Instagram - you can follow me at @flyingmule. I'll be posting additional videos about the LGDs we use in our operation!
When we first started using livestock guardian dogs nearly 15 years ago, I often joked that we our sheep operation was "predator-friendly" because our dogs were not. In other words, the protective behavior of our dogs allowed us to avoid using lethal control on coyotes, mountain lions, and other predators in our environment. And in those 15 years, knock on wood, we've never had to kill a predator. This isn't to say we wouldn't use lethal control if we came upon a predator killing our sheep, but our dogs (for the most part) seem to be doing their jobs.
California now has a predator (the gray wolf) that is protected under both the State and Federal Endangered Species Acts. Under both laws, it is illegal to "harm or harass" a wolf - and lethal control (at least in California) is not an option (even if a wolf is killing livestock). Consequently, the potential interaction between LGDs and wolves takes on new implications - could a rancher whose dog injured (or even killed) a gray wolf be held liable for "taking" an endangered species? Would the California Department of Fish and Wildlife, which recommends LGDs as a non-lethal livestock protection tool in wolf country, actually prosecute such a case?
More than two-thirds of these dogs had no reported interactions with wildlife. Of the 71 dogs that did interact with wildlife, 48% interacted with herbivores, while 73% interacted with predators. Just 9% had lethal interactions with herbivores (where the herbivore was killed); 10% had lethal interactions with predators. All of the documented interactions between LGDs and predators (a total of 44 instances) were defensive in nature (in other words, the predator approached the herd and the dog responded). Ultimately, the authors conclude, "Overall, the conservation benefit of LGDs does not appear to be outweighed by ethical implications of their use; LGDs were shown to be highly targeted and discriminatory towards predators attempting to predate on livestock." In other words, the LGDs in this study were effective at deterring predation while minimizing impacts on non-threatening wildlife.
I found this research especially interesting in light of my own observations of our LGDs and the LGDs used by other California operations. I've never witnessed a direct interaction between our dogs and a predator (which is why we're using GPS collars and trail cameras to try to document these encounters). In my 15 years of using these dogs, I've encountered 2 dead raccoons and 2-3 dead jackrabbits in our sheep paddocks, which I assume were killed by our dogs. I have also observed two of our dogs going after an otter (which is a much more vicious predator than I'd realized). I know that our dogs will respond when they hear coyotes howling in the distance, but I've never seen a direct confrontation.
Working with several range sheep operations this summer, I suspect that wildlife encounters may be more common in these unfenced, herded systems. The herders I've worked with report that the dogs are very active at night (as some of my trail cameras and the GPS data suggest).
I'm interested in learning from other producers - how do your LGDs respond to non-threatening wildlife versus predators? What kinds of things do you do to correct inappropriate behaviors? I hope you'll share your insights in the comments to this blog!
Warning: the video link in this blog post includes images of a sheep killed by a coyote.
This story begins several weeks back. We split our ewe flock into two breeding groups in late September, and kept a third group (of replacement ewe lambs, who won't be bred until next fall) separate. With three groups of sheep, we felt like we needed to put our youngest livestock guardian dog (10-month-old Dillon) with the lambs. He had been with a handful of lambs at our home place, and seemed to be fine - a bit exuberant (as puppies can be), but fine.
A week into this situation, we noticed that one of the lambs had a chewed ear. We've had young dogs that occasionally chewed on ears, so we weren't too worried. In the next several days, four more lambs were injured, several seriously. We decided to put Dillon with a group of older ewes and rams.
His inappropriate behavior continued - and escalated. He began chasing sheep, which culminated with a ewe that became tangled in the electronet and died. We brought Dillon home (and put a "dangle stick" on his collar to make chasing sheep uncomfortable). The older dog went back with the breeding group, and we left the ewe lambs protected only by electric fence.
Fast-forward to this week. On Wednesday, we moved the lambs to a new paddock partly enclosed by electro-net fencing, partly by hard-wire sheep fence. On Thursday morning, we found a dead lamb in the paddock.
The rancher part of me was upset - we would expect this ewe lamb to grow up to produce five or six sets of lambs. The farm advisor part of me decided that this was an educational opportunity - I wanted to learn how to tell what kind of predator had killed the lamb.
I made a quick call to our local wildlife specialist (in Placer County, these folks work for the county - in other areas, county trappers are employed by USDA Wildlife Services). He looked a a few photos and said, "That looks like a coyote." He also told me how to investigate the carcass to know for sure.
You might wonder, why would this matter? The lamb was dead. As a rancher, I wanted to know what I was dealing with! Coyotes can get through a 4x6 inch hole; mountain lions can go over most fences. Mountain lions are protected by the State of California; coyotes, at the moment, are not. As a scientist, I was inherently curious. I wanted to know the differences between coyotes and mountain lions in terms of predatory behaviors.
Our trapper told me, "A mountain lion will usually kill its prey by crushing the base of the skull from above and behind; a coyote will kill by crushing the trachea from below. A lion will not usually eat the digestive tract; a coyote will eat everything. A lion will bury what it doesn't eat; a coyote will eat in the open and leave the rest." He also told me that skinning the neck of the lamb would confirm the predator involved: "Hemmoraghing on the throat would indicate a coyote; wounds on the top of the neck at the base of the skull would suggest a mountain lion."
Fortunately, I had my hunting pack in the pack seat of my truck (including rubber gloves and a sharp knife). I skinned the neck and found lots of trauma around the trachea - and no wounds on the top of the neck. We were dealing with a coyote, as the video below indicates.
We resolved the issue (hopefully) by bringing Dillon back to guard the lambs. We left his dangle stick on with the hope that he wouldn't chase the sheep. We also moved the sheep to a new paddock that was more secure. Our landlords told me this morning that they'd heard coyotes - and Dillon barking - most of the night. And we did not lose any more lambs.
As a shepherd, the premature death of any animal feels like a failure. I hate to put a young dog in a position where we have to rely on him before he's ready; I also hate to subject our sheep to depredation. But I also recognize that there are economic considerations involved. Treating the lambs that Dillon injured earlier in the month has cost us money; losing a ewe lamb to a coyote cost us more. These kinds of trade-offs are part of ranching, I suppose; my job as a farm advisor is to help others evaluate these choices objectively.
In the next several weeks, I hope to offer a tool to help others compare the cost of using a livestock guardian dog against the benefits. Stay tuned!