As some readers of this blog may know, I'm currently working on a research project examining livestock guardian dog behavior. The back story is this: several years ago, I was invited to demonstrate electro-net and livestock guardian dogs at a workshop on livestock protection tools. The electro-net fencing was easy! However, since I was speaking at midday, the LGD demo was less than dynamic - the dog came over to the fence, barked half-heartedly at the people he didn't recognize, and resumed napping in the shade!

This experience got me thinking! How could I demonstrate the effectiveness of these dogs without dragging folks out to observe the sheep in the middle of the night (when the dogs are much more active)? Geographic positioning system (GPS) technology seemed like a possible answer - but commercial GPS collars were too expensive for my cooperative extension / sheepherder budget. While perusing Facebook one day, I ran across a post from Dr. Derrick Bailey at New Mexico State University. Dr. Bailey was using home-built GPS collars to track cattle distribution on New Mexico rangeland! At last, an affordable solution! Dr. Bailey was gracious enough to spend an hour on the phone with me talking about my project ideas - and he shared the technical details of the collars he was using.

Here's a quick photo guide to building the collars I'm using on LGDs (and on sheep). The materials include:

• LGD collars from Premier 1 Supplies (I like these extra-wide collars - I think they're comfortable for the dogs, and they seem to hold up in rangeland conditions). https://www.premier1supplies.com/p/guard-dog-collars?cat_id=164
• 3-1/2" x 2" threaded nipples and threaded caps (for the case)
• 1/2" x 5/32" pop rivets and #8 SAE flat washers (to attach the case to the collar)
• i-gotU GT-600 travel and sports logger (available on Amazon - https://www.amazon.com/i-gotU-USB-Travel-Sports-Logger/dp/B0035VESMC/ref=sr_1_2?crid=2OO21VMYVPBN2&keywords=i+got+u+tracker&qid=1564451973&s=gateway&sprefix=I+got+U%2Caps%2C194&sr=8-2

The collars take about 5 minutes to build. The i-gotU trackers can be programmed to collect GPS coordinates from every 5 seconds up to every 5 minutes. Set at 5 minute intervals, the batteries in the unit will last 10 days. Dr. Bailey also sent me plans for an auxiliary battery system - that will be my next project!

I've also experimented with an Optimus 2.0 tracker (https://www.amazon.com/Optimus-Tracker-6543857646-GPS-2-0/dp/B01C31X50K/ref=sr_1_4?crid=UR8F2VQBBT8M&keywords=optimus+tracker&qid=1564452200&s=gateway&sprefix=optimus+t%2Caps%2C199&sr=8-4) which sends a real-time signal to my cell phone with the position and speed of travel of the unit. These trackers don't record positions, but they are useful from a practical standpoint -  they will send an alarm to my phone if a guard dog is out of my pasture.

I'm hoping that we'll have some data to share from my project on the Tahoe National Forest north of Truckee in the next couple of weeks. Working with Talbott Sheep Company, I've collared 2 dogs in each of 2 bands of sheep. So far, the collars seem to be working great!

And on a humorous note, as you can see from the photos, I put UCCE (for University of California Cooperative Extension), along with my phone number, on the collars. I received a text yesterday that said:

"Hello, we found Ucce at the upper little truckee campground this morning. He still has his tracker around his neck and is just hanging out at the campsites."

I explained that we were doing a research project with the dogs and that someone would come by to get the dog soon.

That said, I think Ewecie (or maybe Ewechie) would be a great name for a guard dog, don't you!?

Here are some photos to walk you through building a collar.

One of the questions I'm asked most frequently when it comes to livestock guardian dogs is, "How many dogs do I need to protect my sheep/goats/cows?"  As you might imagine, the short answer is, "It depends." The long answer is more complex. From an economic perspective, the answer is, "As many as it takes to hold predator losses in your operation at an acceptable level, but no more than that." From a production perspective, I've found that the answer depends on operational characteristics, the environment, and the abilities of the specific dog(s).

While it is tempting to try to develop a rule of thumb recommendation (like one dog per 100 ewes), reality is usually more complicated. Wearing my sheepherder economics hat for a moment, the fundamental question comes down to comparing the costs of a dog versus the benefits the dog provides. On the cost side of the ledger, I must account for the cost of dog food, veterinary care, and depreciation. In our operation, these annual expenses amount to roughly $600 per dog. On the benefit side of the ledger, I need to know how many sheep DON'T get killed by predators to determine if my $600 in expenses are justified. Obviously, this is not an easy number to estimate - how can I measure something that doesn't happen? How do I quantify the sheep that might have died had I not had a livestock guardian dog with them? I suspect we'd lose more sheep if we didn't use dogs, but I'm not willing to leave the sheep unprotected to find out!

Operational characteristics, in my experience, play a significant role in determining the optimal number of dogs. Birthing seasons (spring vs. fall), other livestock protection tools (like electric fence, on-site herders, night penning, etc.), grazing management (set stocking versus rotational grazing), and the number of individual herds or flocks all factor into determining the right number. Using our operation as an example:

• We lamb in the late winter and early spring, when there is not a significant natural prey base for the wild predators in our environment. Our lambing paddocks are 7 miles from our home. This argues for more dogs.
• We use electro-net fencing, which definitely deters canine predators (dogs, coyotes and foxes) as well as bobcats. This allows us to get by with fewer dogs.
• We move the sheep frequently - they move to fresh pasture every few days, and graze different properties in spring/summer versus fall/winter. I suspect all of this movement keeps the predators off balance. This allows us to get by with fewer dogs.
• We rarely (if ever) have all of our sheep in one mob. This time of year, the mature ewes are in one flock; the feeder lambs and replacement ewe lambs are in a second flock; the rams in a third location. During breeding season, we have two separate breeding groups plus a group of lambs. This argues for more dogs.

Based on these factors, we feel that we need at least three dogs for our small, part-time operation. With three dogs, we can protect three different groups of sheep or place two dogs together during our most vulnerable time of year (lambing). During some parts of the year, we have more dogs than necessary, which provides flexibility if we begin to have problems with predation.

The environment where we're grazing, and the predators it contains, is a second critical consideration. Here in the Sierra foothills (at least at the moment) our main predators (in order of potential threat) are domestic dogs, coyotes, mountain lions, black bears, bobcats, foxes, and birds of prey. I've spoken with ranchers on the north coast who would add crows, ravens, and magpies to that list. And ranchers in northeastern California would add gray wolves. Predator density and prey base also come in to play. Are there several established packs of wolves in the region? Is there sufficient native prey? Are these particular wolves (or coyotes, or mountain lions) known to prey on livestock? Each of these questions are important to consider when determining how many dogs a particular operation might need.

Finally, every livestock guardian dog is an individual. Some are athletic and want to patrol a wide area; others want to stay with their livestock. Some dogs are more canine aggressive than others (an important trait in wolf habitat); others will harass bears. And these traits will change over time - a dog that was aggressive and athletic in his younger days might be content to stay with lambs on irrigated pasture in his later years. In my experience, there is more variation between individuals than there is between livestock guardian dog breeds (a subject for a future blog post!).

Finally, I started a new phase of my livestock guardian dog behavior study this week. I'll be tracking the movements of four dogs (2 each in separate 1000-ewe bands of sheep) grazing on the Tahoe National Forest in Nevada and Sierra Counties (in an area that a collared Oregon wolf has been known to visit in the last 12 months). This is a long-time producer with experienced herders operating on open range with no fences. They typically use two dogs with one band and three dogs with the other, and experience less than one percent death loss while the sheep are on Forest Service allotments. They also have additional dogs they can add to each band if predator problems begin to escalate.

I think this illuminates the "it depends" answer in my first paragraph! They have 1 dog per 400 sheep; we have 1 dog per 51 sheep. They are grazing mature ewes in a relatively wild environment for only 75 days - and at a time when the natural prey base is plentiful. We need more dogs to protect ewes and lambs at an especially vulnerable time of year (and I should note - the large operation needs more dogs at lambing as well). The common thread for each of these operations, however, is that we are constantly evaluating our need for predator protection against the cost of providing it. If we could get by with fewer dogs, we would; similarly, if the large operation needs more dogs this summer, they'll add dogs. In other words, it depends!

I realized this morning that it's been some time (nearly a year, in fact) since I posted an update on my progress with our youngest livestock guardian dog, Elko. In case you missed the first five installments of this journal, Elko is a Pyr x Akbash dog we acquired from Fred Groverman in Petaluma in September 2017. I've been tracking our progress in his development. This weekend marked a significant step in Elko's development.

During the last week of March, I traveled through northeastern California talking about (and more importantly, learning about) protecting livestock from predators. Thanks to a Renewable Resources Extension Act grant, a number of my UC Cooperative Extension colleagues and I were able to bring some folks with experience dealing with wolves and grizzly bears in Wyoming and Montana to share their perspectives with California ranchers. During our formal workshops - and during the 700+ miles we traveled together - I learned a tremendous amount!

George Edwards is the executive director of the Montana Livestock Loss Board, a state program that compensates ranchers for losses to wolves, mountain lions, and grizzly bears. George provided invaluable insights into why a compensation program makes sense - and into the types of tools that ranchers are using in Montana to avoid these losses. He shared this outstanding video about an innovative carcass disposal program in western Montana.

I first met Cat Urbigkit nearly 10 years ago at an American Sheep Industry conference in Reno. I've since had the chance to get to know Cat as a fellow producer and friend. Her experiences using livestock guardian dogs to protect both sheep and cattle from large carnivores - including gray wolves and grizzlies - was especially enlightening.

Now that I've had a few weeks to reflect on what I learned, my own experiences with livestock guardian dogs have come into clearer focus. Cat emphasized that no two dogs are alike - just as no two livestock operations are alike. A dog that will work for Cat in western Wyoming may not be a good fit for me in the foothills of the Sierra Nevada; similarly, my dogs may not work well in Cat's extensive rangeland environment. In other words, a dog "trained" by someone else - that is, bonded with someone else's livestock in their environment - won't necessarily stay with my sheep. I have not been successful in outsourcing the critical bonding period. My dogs have to bond with my sheep.

Livestock guardian dogs, as Cat says, are "dogs of nature." Rather than "training" them, we must give them the opportunity to fully express their own instincts. They should smell, hear, and see the livestock they'll spend their lives protecting from the earliest possible moment. I've written previously that I think this means that my dogs need to bond with my sheep and my system (electric fenced paddocks in the urban-rural interface). I think that's not quite right, however. After visiting with Cat, I think the bond between dog and my sheep is the most critical component. My dogs stay with my sheep because of their bond - not because of my fence. The dogs I used early in my shepherding career roamed because I hadn't allowed them to bond properly, I suspect. And this bonding process has to happen on my operation - a livestock guardian dog "specialist" who isn't also a rancher can't duplicate my specific conditions.

Elko spent last summer with our dry ewes in the company of another LGD, Bodie. When we split our sheep into separate breeding groups in late September, Elko (who was then just over a year old) went with his own group of ewes. About a week into our breeding season, we suspect we lost a ewe in Bodie's group to a coyote attack (in a paddock with low visibility due to vegetation). We put Elko with Bodie at that point, with good results. Bodie, who we know we can trust with lambing ewes, stayed with the ewes through the winter and during lambing. Once we separated the rams from the ewes in mid-November, Elko stayed with the rams until this past Saturday. The rams went home to be shorn; the ewes and lambs went to irrigated pasture - and Elko joined Bodie as part of his "final exam."

Last year, we put Bodie with an older dog at the beginning of lambing season (here's a video link). The older dog, Reno, chased Bodie away from lambing ewes, which helped Bodie understand what was expected of him. By coincidence, our last ewe lambed on Saturday shortly after we put Elko with Bodie and the ewe flock - and I observed Bodie provide similar training. Bodie would not let Elko get too close to the ewe and her new lambs. I suspect that this was at least partly because Bodie wanted to eat the afterbirth; regardless, the result was that Elko learned to respect a lambing ewe's "personal space."

Elko's training is not completed - he's still in the canine equivalent of his late teen years. At the risk of anthropomorphizing our LGDs, I know that I didn't always make the best decisions in my late teen years - we'll keep a close eye on Elko while he's with the lambs and ewes. That said, this weekend he passed a significant test!

A final note on costs - Elko cost us $525.35 to acquire (cost of puppy + mileage). Through Saturday, we've spent $526.44 on vaccinations and dog food.

Also - here's a link to a new fact sheet on LGDs we developed as part of our grant-funded project.

As you probably know if you've read previous posts on Ranching in the Sierra Foothills, my go-to tool for protecting sheep on our foothill rangeland and irrigated pasture is my livestock guardian dogs. Since we're trying to protect our sheep from carnivores, it makes sense to protect them with a carnivore! The right dog (or dogs) will protect our sheep from stray dogs, coyotes, mountain lions, black bears, foxes, and other predators. Check these previous posts to learn more about our livestock guardian dogs:

• Which Livestock Guardian Dog Breed?
• The LGD Puppy Journal - #1
• Livestock Protection Tools - Part 1

But livestock guardian dogs are not the only tool available! Some recent research suggests that an ever-changing mix of livestock protection tools may be necessary. Predators can adapt, and so must livestock producers. A 2017 paper in the Journal of Mammalogy tracked the adaptive use of a suite of nonlethal livestock protection tools in Idaho (see Stone et al. 2017, "Adaptive use of nonlethal strategies for minimizing wolf-sheep conflict in Idaho"). Other research acknowledges the difficulty designing experimental studies to analyze the effectiveness of specific tools (see Eklund et al. 2017, "Limited evidence on the effectiveness of interventions to reduce livestock predation by large carnivores").

But formal research isn't the only approach to determining whether a given tool works in a specific environment. On-the-ground use of these tools by ranchers in real-world production settings is invaluable. In my experience, the success or failure of a particular livestock protection tool depends, at least in part, on the paradigm and know-how of the person using the tool. I expect the livestock guardian dogs I've "trained" to work in my environment - and when they don't, I try to learn from my mistakes. My own hands-on experience with these dogs gives me a greater level of comfort in using them in a variety of settings. Similarly, my familiarity with electro-net fencing helps me adapt this tool to our environment.

Recently, a group of colleagues and I received grant funding to put on a series of Livestock Protection Tool Field Days in a number of northern California counties that have been visited by gray wolves (Nevada, Sierra, Plumas, Lassen, Modoc and Siskiyou). These field days will be held in late March 2019 (check the calendar on my website for exact dates - go to https://ucanr.edu/sites/Livestock/). In addition to providing information about using livestock guardian dogs in rangeland environments with both cattle and sheep, these field days will give producers an opportunity to set up and use some additional tools, including electro-net fencing, turbo fladry (an electrified fence with flapping red flags that apparently deters wolves), and FoxLights (a device that emits random flashes of light to mimic a person with a flashlight).

This week, we are testing the FoxLight devices in small foothill paddocks. I'll be setting up trail cameras to detect whether the coyotes and foxes are deterred by the lights - but the ultimate test will be a lack of predation! One of the lights is in a hard-wire paddock that has had coyote incursions in the past; the other is in an electro-net paddock. Stay tuned!

We've also purchased a half-mile of turbo fladry. Sometime in the next month, we'll set up a fladry barrier to familiarize ourselves with setting it up and taking it down. Stay tuned for information on this, as well!

As with any tools, there will be a learning curve for these new techniques. I've built lots of temporary electric fence, but laying out a quarter-mile of poly-wire with red vinyl flags will take a different technique. I'll need to learn to place the wire at the proper height to deter wolves. I'll need to figure out how to re-wind the fladry. As always, I'll track the cost of using these tools, as well. With livestock guardian dogs, the costs include dog food and vet bills. With fladry, the cost will mostly involve my own time.

Finally, I know that I learn best by doing. Reading about a technique or a tool is a helpful introduction, but I need to use these tools in a real-world setting to gain any sense about their effectiveness - and about my willingness to use them. I suspect many of you feel the same way! I hope you'll be able to make one of our field days in March!