Background

Diffuse knapweed was most likely imported in alfalfa seed from the Caspian Sea region of Turkmenistan. It was first reported in Washington state in 1907. It is now widespread in all the states west of the Rocky Mountains and is estimated to be increasing at about 18% per year. The plant is allelopathic, meaning that it produces chemicals in the soil that prevent other plants from growing. Infested areas suffer from reduced biodiversity and soil quality and increased erosion. The plant is distributed throughout Boulder County, with especially dense stands in the southern part of the county.

Boulder County has used a variety of methods to control diffuse knapweed populations. Large-scale herbicide applications in the '90s helped set the weed back, but were controversial. Insects have shown success, though progress is slow. Grazing has also presented possibilities. Local permittees pointed out that cattle eat diffuse knapweed when it is bolting. Based on this discovery, a study done in cooperation with City of Boulder Open Space and Colorado State University found that grazing diffuse knapweed in the bolting stage reduced stand densities by about 50% and also reduced seed set by about 50%. This was significant. But since mature plants can produce as many as 18,000 seeds each, even a 50% reduction leaves managers looking for additional tools.

Plants that are grazed in the bolting stage generally resprout and set buds and seeds by the end of the growing season.   But if animals would graze the plant after it has set buds and flowered, it might be possible to reduce the plants ability to resprout and set seed. The purpose of this project was to determine if steps used to train cows to eat other weeds might be successfully used to train cows to eat late season diffuse knapweed so that animals already grazing on Boulder County Open Space might become an additional tool in managing the weed.

Process

Here is the three-step process as applied to this project:

Step 1) Know the target plant: What is its nutritional value? Does it contain any hazardous secondary compounds (toxins)? Will supplements aid animals in eating the plant?
      A review of the literature for diffuse knapweed indicates that there are no toxins of concern. Its nutritional value changes through the season, its crude protein dropping from a high of 18% in the rosette stage to 8% during the bud and flowering stage (Roche). This low nutritional value was of potential concern in this project's success.

Step 2) Choose the right animals: Research shows that younger animals learn more quickly than older animals and that mothers pass diet selection knowledge on to their young (Burritt 2000). Therefore by working with young females, we can eventually have a whole herd of weed eaters for a very small investment.
      For this project I worked with 50 bred, angus/limousin cross heifers owned by Babe and Leo Hogan. We trained in a large pasture where the animals had free choice forage and water. The number of animals and the training location presented two challenges. First, I normally only train 25 to 30 animals because of the logistics of hauling feed and harvesting weeds during the training period. However, since all the animals were already in pasture together we decided to experiment with training a larger number. The second challenge was that in order to conduct training we first had to train the animals to come to the feeding site. Fortunately, it took the heifers only one day to become familiar with grain bags and so they followed me or my volunteers whenever they saw us with a grain bag.  

Step 3) Build on how animals choose foods by creating positive experiences with new foods, and making the unfamiliar seem familiar.
      Basic behavioral research shows that the more positive experiences with new things a creature has the more willing it is to try other new things (Villalba 2004, Thorhallsdottir 1987, Burritt 1990, duToit). I built on this basic behavior by giving cows a series of good food experiences morning and afternoon from Tuesday, July 10 through the morning of Friday, July 13. Familiar tubs, people and flavors were incorporated in the feeding, because of research showing that animals are more likely to try something if it is associated with familiar forms and flavors (Launchbaugh 1997, Provenza 1995).   In this case the tubs used for training were the same tubs used for protein supplement through the winter, so the heifers were already familiar with them. The familiar flavors I took advantage of were ones the animals were already familiar with from their winter feeding: corn and molasses. When I used molasses as an additive, I diluted it to 25% molasses and 75% water so I could spray it on using a hand sprayer.
      On Friday afternoon, because the animals had been eating so well, I began with weeds. At this stage the plant is very stemmy with very few leaves. The only parts that seemed potentially edible were the buds and flowers. So I focused on gathering these parts of the plant for the training. It was quite hot, and picking the weed was difficult, so I didn't have a lot of the weed for the very first try. I put it in the tubs, sprayed it with the diluted molasses mixture, added red bran and some barley and molasses and set it out for the heifers. The heifers did as all animals I have trained so far. They ate some but not all of the weed. I repeated this process on Saturday and on Sunday fed the weed plain. The weather was quite hot and the heifers ate a little of the weed, but not all. When I returned in the morning, they ate the leftover weeds from the tubs and the ground around them. That afternoon I fed one last round of the plain weed, spraying the bottoms of the tubs with molasses to ensure that every animal at least stuck her head in a tub. They cleaned their plates.

Step 4) Practice in Pasture: Research has demonstrated that a small amount of pressure increases an animals willingness to try new foods in a new environment (Villalba 2004). Therefore when I transition animals to the pasture I begin by putting them in small trial pasture. A good trial pasture has only a few days of forage and includes a wide variety of foods along with the target weed. Below is the pasture:

View Larger Map