The objective of this study was to use fecal near-infrared reflectance spectroscopy (NIRS) to estimate the percent spotted knapweed in the diet of grazing ewes to fine tune grazing prescriptions. In trial 1, five ewes were randomly selected from a band of 900 ewes to estimate weekly variation in percent spotted knapweed in the diet. Fecal samples from these ewes were collected weekly from June 22 through August 17 (nine collections per ewe for a total of 45 observations). In trial 2, fecal samples were collected from 89 ewes in the band to estimate variability in percent spotted knapweed in the diet among sheep at two sampling dates corresponding to the bud (July 13) and postflowering (August 15) stages of spotted knapweed phenology. In trial 1, the percent spotted knapweed in the diet was similar (P > 0.05) on June 22, June 29, July 13, and July 20 but was greater (P < 0.05) on July 6 than June 22 and 29. Percent spotted knapweed in the diet was also greater (P < 0.05) from July 27 through August 17 than from June 22 through July 20. On July 13 in trial 2, 55% of ewes had 0 to 5% spotted knapweed in their diets, whereas 44% of ewes had 5 to 20% spotted knapweed in their diets. On August 15, only 1% of ewes had < 10% spotted knapweed in their diets, whereas 44% of ewes had 20 to 25% spotted knapweed in their diets. The estimated percent spotted knapweed in the diet of the 89 ewes in trial 2 was similar to that of the five ewes during the same time period in sampling trial 1 (5.3% compared with 5.0% for sampling trials 1 and 2, respectively, on July 13 and 22.0% compared with 20.7% on August 15 and 17 for trials 1 and 2, respectively). This is the first study to estimate percent spotted knapweed in the diet of individual grazing ewes. These data suggest that the best time to graze spotted knapweed–infested pastures would be in late July or early August, when spotted knapweed is flowering, but before viable seed production.

Predicting site vulnerability to nonnative plant establishment remains a difficult goal. Seedling survival is an important component of population dynamics and can affect the success of control strategies. Field manipulations allow potential causal mechanisms of site vulnerability to be evaluated under realistic environmental conditions. We conducted field studies to determine the effects of plant competition and differing precipitation regimes on spotted knapweed seedling survival. We also examined the effect of herbivory on rosette survival and growth. Seeds were sown into plots with vegetation intact or removed at three sites. Seeds were also sown into plots where plant competition and precipitation were manipulated in a factorial design at a single site. Field studies demonstrated that site accounted for much of the variation in emergence rate, while herbivory and plant competition affected seedling survival rates. We observed a wide range in emergence rates, with site averages ranging from 13.1 to 42.5%. Survival the following year ranged from 0.5 to 9.4% of sown seeds. Rosette survival was significantly higher when herbivores were excluded from plots. Below average precipitation reduced seedling survival; however, even with supplemental water, dry-down of exposed sites resulted in low seedling survival. Of the 8,000 seeds added to plots in one study, by autumn, only eight plants resulted, seven of which survived in watered plots with intact vegetation. Collectively, these results show that seedling survival is a critical phase in spotted knapweed population dynamics and can vary among habitats on the basis of plant competition and precipitation. Furthermore, herbivory affects all stages of the lifecycle from the seedling onward. The observed differences help explain the reported variability in seedling survival in the literature and inform efforts to control spotted knapweed using plant competition and biological controls.