Amy Clark and Samuel B. St. CLair, Plant and Wildlife Science Department
Introduction
Invasive species pose a serious threat to earth’s ecosystems. This is a problem especially in the arid west where millions of acres of Great Basin shrubland have become dominated by invasive weeds such as cheatgrass and halogeton following disturbances like fire. Both of these invasive species lower the quality of the range by making it difficult for more desirable, native plants to become established. It has been shown that ant communities can be used to evaluate ecological response to disturbances in rangelands (Hoffman 2010). While there have been studies on the effect of fire on harvester ant communities (Suazo et al. 2013) and how ants affect the establishment invasive plants (Ostoja, Schupp, & Sivy 2009), it is not well established how fire affects ant influence on invasive plant establishment. In this study we track the harvester ant population for one year following fire and measure their impact on plant establishment through seed preference and seedling fate studies. We predict ant populations to increase in burned relative to undisturbed areas and ant activity to reduce plant invasion.
Materials and Methods
Rush Valley is a sagebrush dominated area located west of Utah Lake. BYU has five established plots made up of four treatments: burned or unburned with or without the presence of small mammals. Each treatment is 30x30m and enclosed by 1m tall, ¼” hardware cloth fencing, flashing was used to exclude small mammals. To estimate ant abundance, pitfall traps were placed 10m in from each corner in each treatment for a total of 80 traps across the 5 plots. These traps were 10cm cups buried to ground level and set by filling the cups halfway with propylene glycol. The traps were left in place for one week each month while the ants were active (from May to October). Seed preference studies used 15cm petri dishes, buried to ground level and covered with hardware cloth to prevent small mammal or bird access. Six seed types were used; two native grasses: indian rice grass and squirreltail bottlebrush; two invasives: halogeton and cheatgrass; and two native forbs: penstemon and yarrow. A mound of similar size was selected in each treatment and dishes were placed 1m outside of the disk of each mound. One hundred seeds were placed in each trap and seeds were collected and counted after one day. For the seedling fate experiment, 55 seedlings of 6 different species were grown in the greenhouse on a sand-soil mix. A native, naturalized, and invasive species of both grass and forbs were used (squirreltail bottlebrush, wildrye, cheatgrass, sagebrush, penstemon and lewis flax –used in lieu of halogeton). When the seedling were roughly 2” tall they were transplanted 2.5m out from the center of 27 different ant mounds. Plants were randomly sorted and placed in two rows on opposite sides of the mound. The seedlings were watered daily and density was calculated after 1 and 2 weeks.
Results
Ant abundance reflected ant activity across seasons by month. The amount of ants caught in the burned treatments was almost always significantly higher when compared to unburned treatments (see Fig. 1). Overall abundance for the year was more than double in burned plots (Fig. 2). Many difficulties were encountered with the seedling study as high winds in the valley may have affected the results. For the seedling fate study it was found that invasive cheatgrass was removed the least and that the native forbs and woody species was removed the most –sagebrush and flax (Fig.3). Data was entered and analyzed using Excel Office and error bars show one standard deviation from the mean.
Discussion
Our hypothesis that ant abundance would be greater on burned sites was supported by our results. The difference between treatments may be even larger when distance to nearest mound is factored in. However, our expectation that ants contribute to biotic resistance by inhibiting invasive plant establishment is not supported by our findings. It appears that ants are more likely to defoliate and remove more robust seedlings before grass species. This may also have been affected by slight variations in initial seedling densities. Although we did not have ample repetition or significant results with the seed preference study we found that cheatgrass was often left untouched and the greatest proportion was taken from trays with larger, native seed as found by Ostoja (Ostoja et al. 2009). The harvester ants remove materials found directly on the disk but are more likely to select for larger seed.
Conclusion
While the current observable trend is that ant populations are rising in burned areas, this trend may shift as more invasive species dominate the disturbed areas. Ants create disks around their mounds to control hydrology and temperature of the nest, which is essential to survival. As invasive weeds overgrow the mounds faster than they can be removed by the ants it is possible that the colony will die. We are also tracking the number of mounds in each plot as new mounds are formed or as mature mounds become inactive. This information may help determine the general effect of fire on ant abundance overtime when combined with our trap data. The results of this study are important in determining the role of ants for plant invasion following disturbance in sagebrush systems.
References
Hoffmann, B. D.
Ostoja, S. M., E. W. Schupp, and K. Sivy. 2009. Ant assemblages in intact big sagebrush and converted cheatgrass-dominated habitats in Tooele County, Utah. Western North American Naturalist 69:223-234.
Suazo, A. A., D. J. Craig, C. H. Vanier, and S. R. Abella. 2013. Seed removal patterns in burned and unburned desert habitats: Implications for ecological restoration. Journal of Arid Environments 88:165-174.