We're working with some homeowners near Auburn to try to control yellow starthistle using sheep. As part of this experiment, the homeowners are also spraying some of the starthistle patches. Several weeks ago, I heard second hand that the man who is applying the herbicide told the homeowners that grazing wouldn't work. I didn't feel too bad - I feel the same way about the long-term effectiveness of spraying!
His comment, however, made me think about the difference in perspectives that would lead one to favor grazing over herbicides (or vice versa). I'm not necessarily opposed to using herbicides as a tool (if they can be used safely). To me, though, herbicides are often used to treat a symptom rather than a disease. Let me explain.
Yellow starthistle is often a symptom of larger land management issues. Infestations seem to occur where there's been soil disturbance, un-managed grazing, or other problems. Starthistle, with it's deep tap-root, can often out-compete other rangeland plants for moisture - it's especially adept at using late season moisture.
By my observation, chemical applications will kill starthistle, reducing seed production and ultimately reducing the viability of a population of yellow starthistle. However, I've also observed that starthistle is typically replaced by equally undesirable invasive weeds when spraying is used. In our region, medusahead barley, another invasive from Eurasia, seems to be well-suited to take over once starthistle is sprayed.
By contrast, the type of grazing management that we practice is designed to address the underlying health of the soil. By managing the timing, duration and intensity of our grazing, we hope to cycle nutrients through the soil (including carbon). We also hope to create a micro-environment that favors the growth of more desirable plants (usually a mix of native grasses and beneficial introduced annual grasses). With the project I've described above, we started by putting the sheep through stands of last year's starthistle carcasses. The ewes ate last year's seeds (reducing the seedbank for this year's crop) and trampled the dead stalks. This trampling puts a substantial amount of carbon in contact with the soil, where bacteria and other soil organisms can break it down into accessible soil nutrients. As the current year's growth begins, we make a second pass - grazing the germinated plants will stress them enough to reduce seed production. When the plants begin to bolt (usually in May), we'll make one final pass.
We all enjoy instant gratification - spray a plant this week, and it will start to die immediately. A systems approach (like grazing) takes longer - our soils didn't become unhealthy overnight, so grazing can't fix the problem overnight, either.
Here's a video of us using herd effect to trample last year's starthistle.