Friday, August 7, 2009 - 8:00 AM

COS 120-1: Assessing Bromus tectorum in Rocky Mountain National Park using sampling and modeling approaches

James E. Bromberg, Colorado State University, Sunil Kumar, Colorado State University, and Cynthia S. Brown, Colorado State University.

Background/Question/Methods

Cheatgrass (Bromus tectorum L.), an invasive winter annual grass, may be increasing in extent and abundance at high elevations in the western United States. However, data to track this species in high elevation environments are limited. To address changes in the distribution and abundance of cheatgrass, we used traditional statistical methods and computer modeling. We re-sampled plots from two vegetation surveys established in 1993 and 1996 in Rocky Mountain National Park (RMNP) for occurrence and cover of cheatgrass. We assessed changes in the species’ presence and abundance over time. While the causes of a potential increasing distribution at high elevations were not evaluated, we wanted to know if detectable changes were occurring in this region and what factors were most responsible for its occurrence.
Results/Conclusions

While not all comparisons between years demonstrated significant changes in cheatgrass abundance, the mean cover of cheatgrass increased nearly five-fold from 1993 (0.73%) to 2007 (3.64%) in one of the two vegetation surveys (p=0.02). Cheatgrass was present in 50% more of the plots in 1999 than in 1993 (p=0.01). In the second survey, cheatgrass was present in 30% more of the plots in 2007 than in 1996, however this increase was only marginally significant (p=0.07). Maxent, a species environmental matching model, predicted similar distributions and probabilities of cheatgrass occurrence over three sampling time periods using the survey data. The results of the model demonstrate Maxent’s utility for accurately predicting the species’ potential range with limited data. The model found that distance to roads, elevation and vegetation community influenced the predictions most.  The response of this species to interannual environmental variability makes detecting change challenging. However, our results suggest that cheatgrass is likely increasing in frequency and abundance in RMNP. Continued sampling will be necessary to verify this trend.