Evaluating management strategies and recovery of an invasive grass ( Agropyron cristatum) using matrix population models

Abstract

The success of invasive species control depends on management type, duration and recovery following management. Population models were here used to compare response and recovery of clipped and herbicide-treated Agropyron cristatum populations invading native North American grasslands. Matrix models were produced for experimental populations of A. cristatum using probabilities of tussocks changing size class (based on tiller number), as well as average seed production, germination, juvenile survival (progression to size class 1), and survival of seeds in the seed bank. Tiller and seed production, as well as growth rate were compared among four management strategies: clipping (to simulate grazing) for one or two consecutive years, and selective herbicide application for one or two years. The responses of managed populations were compared with an unmanaged population. Elasticity analyses were used to identify life-history stages that contributed the most to survival and growth, thereby providing important insights into invasive species control. Herbicide application for one year slowed population growth, while application for two years eradicated the population. Clipping for one year slowed population growth, but did not cause the population to decrease in size, whereas the population clipped for two years was stable. Recruitment, especially seed production, contributed the most to the high growth rate of the unmanaged population, which was significantly higher than the growth rate of populations that were clipped or treated with herbicide for either one or two year periods. In summary, strategies to control this invasive grass need to target seed production. This study emphasizes the importance of long-term sustainable control strategies.