As a sheep producer and an extension educator, I'm often asked which breed of livestock guardian dog (LGD) I prefer. Some have heard that Great Pyrenees dogs are more likely to wander. Others have heard that Akbash and Anatolian Shepherd dogs can be more aggressive with humans. My experience with LGDs over the last decade suggests that there is greater variation between individuals than between breeds - in other words, breed is not as important as matching the behavior of a specific dog with the ranch's needs and environment. New research published in Rangeland Ecology & Management sheds new light on this topic.

In "A Livestock Guardian Dog by Any Other Name: Similar Response to Wolves Across Livestock Guardian Dog Breeds," authors Dan Kinka of Utah State University and Julie Young of the Wildlife Services National Wildlife Research Center compare several new (to the United States) breeds of LGDs with the typical American "whitedog" (Great Pyrenees, Akbash, Anatolian, Maremma, and crosses of these breeds). Some producers believe that because the American whitedog breeds were initially selected to protect small ruminants from coyotes, they may not be well suited to deterring larger predators (especially gray wolves and grizzly bears). As part of the study, Kinka and Young imported kangals from Turkey, karakachans from Bulgaria, and cao de gado transmontanos from Portugal. These breeds were selected because they were typically used in their home countries to protect livestock from wolves and European brown bears - and they were considered to be human-friendly. The imported dogs were placed with sheep ranchers in Idaho, Montana, Oregon, Washington and Wyoming. The whitedogs used by these operations were considered to be a single control breed for the purposes of the study.

The study included direct observation of behavior during normal ranch operations, as well a decoy test designed to simulate mule deer and wolf encounters. Kinka and Young recorded five behavior components (activity, posture, vocalization, proximity to livestock, and out-of-view to the observer). Within these components a number of specific behaviors were documented (like scanning, investigation, vigilance, chasing, etc.).

While the authors noted some subtle differences in behavior and responses to simulated wildlife encounters between breeds, they noted "that kangals, karakachans, transmontanos and whitedogs spent equivalent proportions of time in most behaviors during both baseline sampling and simulated wolf encounters." They also found that LGD age and time of day influenced LGD behavior and that sex had no effect on any behavior - observations I've made with my own LGDs. For example, our LGDs always seem to be much more active and vigilant at dusk than during the middle of the day. Ultimately, the authors suggest, "the homogeneity of behavioral data for multiple LGD breeds suggests that regardless of breed, LGDs operate in much the same way. As such, breed may be a less important predictor of a 'good dog' than often suggested."

So what makes a good dog? Obviously, this definition varies from one operation to the next based on context. In our operation, a good dog needs to stay with our sheep, inside our electro-net fencing. A good dog shouldn't chew on or chase (or kill!) the livestock it is protecting. A good dog should be reasonably friendly with people but prefer the company of sheep. And good dog should deter coyotes, mountain lions, black bears and other minor predators in our environment.

If wolves continue to move south, I suspect my definition of a good dog might evolve. I know ranchers in the northern Rocky Mountains who are using larger dogs (including some of the breeds evaluated in this study). A large-scale targeted grazing contractor who has grazed sheep and goats in wolf territory in Montana and Idaho swears by intact male whitedogs. Last week, I visited a sheep permittee as they turned out onto a Tahoe National Forest grazing allotment approximately 12 miles northeast of where the California Department of Fish and Wildlife detected a GPS-collared wolf in early June. The ranch manager indicated he'd be adding more dogs to each of his two bands of sheep. For those of us in California, wolves are a wild card - our dogs have never had to contend with a large, pack-hunting predator.

While formal research on the behavioral attributes of successful LGDs is critical, we also need to share our on-the-ground experiences! What do you look for in an LGD in your operation? Do you use different dogs for different situations? I suspect each of us will have a slightly different answer to these questions! I hope you'll join in this conversation!

Reference

Kinka, D., Young, J.K., A Livestock Guardian Dog by Any Other Name: Similar Response to Wolves Across Livestock Guardian Dog Breeds, (2018), https://doi.org/10.1016/j.rama.2018.03.004

Multi-species grazing - grazing cattle, sheep, goats, or even poultry or pigs in the same pastures - is often cited in the popular small farm press as a beneficial practice. Each of these livestock species has different forage preferences and grazing behaviors. Each species is impacted by different internal parasites; grazing more than one species in a pasture can disrupt the life cycles of these parasites. Some producers have observed that grazing small ruminants with cattle can provide predator protection benefits as well. For the most part, however, livestock producers in the foothills practice multi-species grazing on a very small scale.

A number of larger ranching operations in the Sacramento Delta have been grazing sheep and cattle together on a large scale with promising results. Typically, a rangeland manager (like me) would suggest that one cow equals five sheep in terms of stocking rate; in other words, for every five sheep a rancher adds to a pasture, he or she would have to remove one cow. The ranchers in the Delta, however, are finding that this ratio doesn't necessarily hold. Ryan Mahoney of Emigh Livestock, for example, reports that he can increase the stocking rate on his pastures by grazing sheep and cattle - without over-utilizing the pastures.

But what about animal performance? Obviously, the point of any commercial grazing operation is to put pounds on livestock or to produce offspring. I recently came across a 2001 paper in Proceedings, Western Section, American Society of Animal Science by B.C. Glidewell (of the USDA Agricultural Research Service in Oklahoma), J.C. Mosely (of Montana State University) and J.W. Walker (of the Texas Agricultural Experiment Station).  In an experiment conducted at the U.S. Sheep Experiment Station near Dubois, Idaho, the authors compared the diets and average daily gains (ADG) of yearling Columbia ewes and yearling crossbred steers in four treatments:

• Cattle alone
• Sheep alone
• 75% cattle / 25% sheep
• 50% cattle / 50% sheep

Paddock size varied to keep stocking rate constant between treatments. Fecal samples were used to determine the botanical composition of sheep and cattle diets during the 28-day grazing periods.

Keep in mind that the experiment was conducted on a vastly different range type than the annual rangelands and irrigated pastures in our foothill and Sacramento Valley regions. However, the analysis of dietary preferences suggests more dietary overlap that I would have expected. The diets of both were grass-dominated (see Table 1).

Livestock	% Grass	% Forbs	% Shrubs
Cattle	80%	19%	1%
Sheep	76%	22%	2%

Table 1: Dietary Preferences

When sheep and cattle grazed together, dietary overlap averaged 86%. These findings were consistent with other research conducted on similar forage types in eastern Oregon (Vavra and Sneva 1978), which found dietary overlap of 78-86%. They differed from results in western North Dakota (Kirby et al. 1988), which found 30-35% overlap. The nutritive quality of sheep and cattle diets (as measured by the percent of crude protein (CP) and neutral detergent fiber(NDF)) did vary by treatment. Cattle diets were less fibrous when they grazed with sheep (that is, they were lower in %NDF), likely because cattle ate less grass due to competition with the sheep.

Steer performance (measured in ADG) did not vary significantly among treatments, but did trend higher as the proportion of sheep increased. Similarly, the performance of the yearling ewes in the trial trended higher when sheep grazed with cattle.

The authors also compared total gain per hectare (roughly 2.5 acres). As you might expect, total gain per hectare was related to growing-season precipitation in eastern Idaho. Interestingly, total gain by hectare also seemed to be related to the combination of sheep and cattle. In a wet year, multi-species grazing produced more pounds of livestock per hectare than cattle grazing alone. In a dry year, multi-species grazing produced as much gain per hectare as cattle grazing alone. The authors suggest that over the long term, at least in eastern Idaho, gain per unit of rangeland should increase by grazing cattle and sheep together.

Obviously, multi-species grazing won't fit every operation. Sheep and cattle often require different working facilities, fences, and management skills. However, as ranchers in the Delta are finding anecdotally, grazing cattle and sheep together may actually increase production from a given pasture. We should consider doing a similar study on annual rangelands and irrigated pasture!

Here's the complete reference to the paper:

Glidewell, B.C., J.C. Mosley, and J.W. Walker. 2001. Sheep and cattle response when grazed together on sagebrush-grass rangeland. Proceedings, Western Section, American Society of Animal Science. 52: 156-159.

Since I haven't posted an update to my Livestock Guardian Dog Journal for four months, I thought an update on this project might be timely! We've been training a new dog to work during lambing (with some interesting observations about behavior). We've trained the new dog to respect 3-wire temporary fencing (as opposed to electro-net). And we've been collecting GPS and trail camera data on predator interactions. Lots to report!

GPS Collaring / Remote Sensing Project

We have been putting GPS sensors on two livestock guardian dogs that are with a flock of 82 sheep (bred ewes and open yearling ewes) west of Auburn. One of these dogs is a 10-year-old Anatolian shepherd neutered male; the other is a 2-year-old Anatolian x Maremma intact male. These collars record location every 5 minutes. We've also deployed seven trail cameras on the parameter of the sheep paddock to document wildlife, domestic animal and human activity in the proximity of the sheep. Our hope is that when we compare the time stamp on photos with the GPS locations of the dogs, we'll begin to understand what kinds of interactions the dogs have with predators and non-predators.

The habitat where the sheep are grazing is foothill oak woodland and open grassland. To date, the cameras have detected coyotes, foxes, deer, jackrabbits, skunks, raccoons, owls, and small birds - along with domestic dogs, walking/jogging/cycling humans and horseback riders. We're in the process of going through the GPS data to determine what the dogs were doing when these animals and people showed up in the cameras. Here are a few of the most interesting photos:

  

Learning to be a Lambing Dog

Our oldest dog, Reno, has been an outstanding dog at lambing. He keeps his distance from lambing ewes, is very patient with rambunctious lambs, and keeps afterbirth cleaned up (which can attract scavengers and predators). Since he's ten years old, we decided we need to try 2-year-old Bodie with the lambing ewes this year. We also hoped that Reno would teach him manners and respect - Bodie is still a bit immature behaviorally.

Our first lamb was born on February 22, and I was fortunate to arrive shortly after the birth. As has been typical, Reno was lying about 20 yards away from the ewe and lamb. Bodie met me at the pasture fence well away from them. After I had been there about 10 minutes watching the new lamb, Bodie joined us. I shot video of his interaction with the ewe and with Reno - you can view it at this link:

Training a New Lambing Dog (YouTube)

I suspect that some of Reno's protectiveness has to do with his love for eating afterbirth! That said, in the weeks since this interaction, Reno has enforced Bodie's respect for the sheep even when there isn't afterbirth available. And Bodie seems to have matured. He's more respectful of the sheep, less rambuctious in his behavior, and a better guardian dog in general.

Developing an LGD Puppy

Finally, an update on the Pyrenees x Akbash puppy we picked up in September. Elko is going to be a big dog - he's already as big as Bodie. Since pulling the rams from the flock in November, Elko has been with the rams learning manners. For several months, we kept him with Reno (which also helped on the manners front). Since we moved Reno to the lambing flock, Elko has been on his own. He's still definitely a puppy - we're not expecting him to provide much protection at this point, but he is learning to stay with his sheep.

Several weeks ago, we tried an experiment using a different type of fencing. We have found that most of our dogs will stay in 42-inch electro-net. However, we wanted to try training the sheep (and the dogs) to 3-strand poly-wire fencing. I installed a short stretch of fence at our home place and have watched Elko check it out and decide to stay on the proper side. Success!

Here's a relatively recent photo of Elko:

Stay tuned for more information on these topics! And just a note: last week, Reno became lame on a back leg. Our small animal veterinarian thinks he probably tore his ACL. At the moment, he's recuperating in the barn and watching over a trio of very annoying bottle lambs. Given the seriousness of his injury, he's probably permanently retired. He's been a great dog!

 

Modern electric fencing systems can be incredibly useful in a variety of settings. Single-wire portable systems can help producers manage grazing on irrigated pastures or crop stubble. Multiple-wire fences and electro-net systems are used to control sheep and goats in targeted grazing situations or on properties without permanent fencing. Multiple-wire systems can also be used to temporarily fence riparian areas and other key sites in rangeland settings. Finally, these systems can also help protect livestock from predators.

Unlike the physical barrier of a barbed wire or woven wire fence, electric fence is a psychological barrier. As such, animals need to be trained to respect electric fence - and we humans have to be trained to install, maintain, and use it correctly! And as with any management tool, efficiency is critical to making portable electric fencing systems work from a labor and cost perspective. At one time or another, we've used electric fencing in our own operation to contain sheep, goats, cattle, horses, mules, chickens and hogs.

On Thursday, November 9, we will be holding our first Electric Fencing Field Day at Robinson Ranch in Penn Valley, California, beginning at 8:30. You'll learn about the principles of electric fencing, get hands-on experience with a variety of fencing systems, and learn how to troubleshoot problems. Thanks to our co-sponsor, LiveWire Products, you'll have a chance to learn about the latest fencing technology!

For more information, or to register for this free workshop, click here! Or email me at dmacon@ucanr.edu for more information!