Prairie Dogs and Harvester Ants as Ecosystem Engineers
on the Colorado Shortgrass Steppe

 

By Chrissy Alba-Lynn
Colorado State University, Department of Biology

 

 

      What do black-tailed prairie dogs  (Cynomys ludovicianus) and western harvester ants (Pogonomyrmex occidentalis) have in common? Cursorily, one might assume “nothing at all”: comparing furry, burrowing mammals to centimeter-sized insects (whose skeletons aren’t even located inside their bodies) is like comparing apples to oranges, right? Not if both species act as ecosystem engineers.

 

Ecosystem engineers are organisms that, by creating a disturbance, alter the resources available to other plants or animals living in the same area (Jones et al. 1997). A great example of an ecosystem engineer is the oft-described “industrious” beaver. By actively modifying habitat to meet their needs, they alter community composition, structure, and function, and they maintain these changes over time. Similar to beavers, prairie dogs and harvester ants actively engineer and maintain their dwellings, and in the process alter certain aspects of the plant and animal community.

 

Prairie dogs and harvester ants (photo at right by Dale Ward) are conspicuous denizens of the Colorado shortgrass steppe ecosystem. Prairie dog colonies can be expansive, and the vegetation growing on colonies is often different from the surrounding, uncolonized grassland (Whicker and Detling 1988). Through their continual grazing, prairie dogs directly affect vegetation by creating closely cropped “lawns.” Depending on colony age, changes in the plant community include replacement of mid-height grasses by grazing-tolerant shortgrasses; greater species richness of flowering dicots (forbs); and lower grass biomass and diversity. (As a result of these changes, competition for forage between prairie dogs and cattle has been the catalyst of much debate.) At the sub-colony scale, mound-shaped entrances to prairie dog burrows dot the landscape, and they are a good example of how disturbances create patchy (heterogeneous) habitat. During burrow excavation, prairie dogs churn up soil from deeper horizons. They subsequently pack down and maintain this unearthed soil, resulting in the highly disturbed, often denuded, mounds. Such microhabitat creates niche variety, which is essential to maintaining diverse communities over time.

 

Through casual observation, I noticed that prairie dog mounds and harvester ant nests might similarly contribute to habitat heterogeneity. harvester ant nests are typically composed of a central cone of pebbly soil surrounded by a distinct disk of cleared vegetation. The soil and vegetation of nests are highly disturbed because Ants, like prairie dogs, excavate underground living quarters and keep the entrances to their subterranean dwellings mostly cleared of vegetation.

 

Harvester ants can indirectly affect vegetation growing beyond the perimeter of nests because they are seed predators. These generalist foragers gather many types of seeds from the soil surface and return them to the nest for immediate consumption or to be stored in granaries. Seeds are significantly less abundant in heavily foraged areas surrounding nests, and P. occidentalis can remove up to 100% of the seed set of preferred species (Crist and MacMahon 1992). In other ecosystems (e.g., deserts and serpentine grasslands), harvester ant seed removal has been shown to affect plant community composition (Davidson et al. 1984, Hobbs 1985); however, there is little information available about how P. occidentalis’ seed harvesting affects shortgrass steppe vegetation at scales beyond their highly disturbed nests

 

I measured certain mound and nest characteristics, as well as the characteristics of associated vegetation, to determine whether these superficially comparable disturbances indeed similarly contribute to habitat patchiness. My research addressed 1) whether prairie dog colonies provide suitable habitat for harvester ants, 2) how mounds and nests are dispersed on the landscape, 3) whether mounds and nests are similar habitat in terms of bare ground and vegetation characteristics, and 4) whether, beyond mounds and nests, harvester ants alter vegetation at the sub-colony scale in a manner distinct from prairie dog–induced changes at the colony scale.

 

With much-welcomed partial support from the John Marr Fund, I collected data during the summer of 2005 on the Shortgrass Steppe Long-Term Ecological Research Site located about 60 km northeast of Fort Collins. I found several lines of evidence indicating that prairie dog colonies provide suitable habitat for P. occidentalis, despite possible differences in on-colony versus off-colony vegetation. Nest densities were the same on and off colonies (25 nests per hectare on and off colonies, suggesting that P. occidentalis populations on the shortgrass steppe achieve similar sizes in both habitat types (Figure 1a).

 

 

 

 

 

 

 

 

In addition, on-colony nests were significantly larger in area than off-colony nests (Figure1b).

 

 

 

 

 

 

 

 

 

 

 

 

Figure 1a and 1b. Summary statistics of prairie dog mound and harvester ant nest density and size. (a) Mean number (± SE) of mounds and nests per hectare. “Nests on” refers to nests located on colonies; “nests off” refers to nests located off colonies. (b) Mean area (± SE) denuded of vegetation by individual mounds and nests; mound and nest size is equivalent to the area denuded of vegetation.

 

This size difference might result in part from prairie dog–induced changes to colony vegetation. Low-growing grazing lawns maintained by prairie dogs, as well as highly disturbed and loosened soil on and around mounds, could facilitate ants’ clipping of vegetation surrounding their nests. Addtionally, despite all of the prairie dog activity that occurs on and around mounds, 22% of harvester ant nests located on colonies directly overlapped prairie dog mounds. This is rather striking considering that mounds only covered ~1.2% of the study area, and suggests that harvester ants might preferentially colonize mounds. Newly mated ant queens look for previously disturbed bare ground in which to begin excavating new nests (Terranella et al. 1999). Highly denuded prairie dog mounds could serve as a visual cue, attracting queens that prefer to excavate disturbed soil.

 

Because prairie dog mounds and harvester ant nests can last on the order of decades (Carlson and White 1987, Coffin and Lauenroth 1990), it is of interest to understand how these long-lasting microhabitats are dispersed relative to one another. Nearest neighbor analyses (Clark and Evans 1954), which are used to describe the dispersion of points on a landscape, showed that mounds and nests occur significantly closer to one another than is expected based on chance (Figure 2 at right).  This highlights the fact that where populations of prairie dogs and harvester ants co-occur, the spatial attributes of heterogeneity are altered (versus where each species occurs singly). This could have implications for many organisms that exploit bare ground (e.g., vipositing insects or weedy plants).

 

Figure 2. Ratio of the measured distance between nearest neighbors to the theoretical distance between nearest neighbors (M-M refers to a mound-to-mound nearest neighbor pairing type; M-N = mound-to-nest paring type; N-N on and N-N off are nest-to-nest pairing types on and off colonies, respectively.) Ratios <1 indicate aggregated dispersion. Ratios >1 indicate uniform dispersion. Asterisks denote a significant departure from 1 (or random dispersion).

 

Conversely, nests were uniformly dispersed in relation to other nests (i.e., nests occur farther from other nests than is expected by chance; Figure 2), a pattern that has been found previously for harvester ant nests (e.g., Ryti and Case 1986). This overdispersion results in part from territoriality and intraspecific competition for resources, most notably seeds. There was no difference in the dispersion pattern of nests located on versus off prairie dog colonies, which indicates that the factors contributing to nest spacing (e.g., resource competition) did not differ across habitat types.

 

Ants on colonies remove ~36 m²/ha of vegetation, approaching 30% of the total area denuded by prairie dogs (~118 m²/ha). Taken together, ants and prairie dogs denuded 153 m²/ha on colonies (~1.5% of vegetation in the sampled area) versus 26.5 m²/ha (~0.27% of vegetation) denuded by ants off colonies. Ant nests (both on and off colonies) had significantly less intact vegetation than did mounds. Mounds supported greater plant species richness and diversity than on-colony nests, but not off-colony nests. Two plants species significantly contributed to the relatively high vegetation cover on mounds: the native perennial forb, scarlet globemallow, and the exotic perennial forb, Russian Thistle. The greater cover of Russian Thistle on mounds versus nests (both on and off colonies) suggests that mound disturbances in particular allow establishment of this weedy exotic, possibly by releasing it from competition with native perennial grasses. And although Scarlet Globemallow is a preferred forage item of prairie dogs, the plants seem to be relatively tolerant of prairie dog activity on and near mounds. In some cases, Scarlet Globemallow forms near monocultures in areas highly disturbed by prairie dogs (photo at right).  In general, prairie dogs maintain less meticulously denuded patches of soil than do ants, and of the three patch types (mounds, and nests on and off colonies) on-colony nests support the lowest richness and diversity of plants. 

 

Beyond the perimeter of mounds and nests, prairie dogs are the main agent of vegetation change on colonies. Harvester ants’ contribution to vegetation heterogeneity was minimal and restricted to nests, which cover only ~1/3 of 1% of the study area. Vegetation sampled beyond the perimeter of mounds and on-colony nests was similar in height, cover, richness, and diversity. Thus, there is no evidence that ants alter vegetation in a manner that is distinct from the changes brought about by prairie dogs, except where ants are clipping nest vegetation. In contrast, there were several differences in vegetation sampled on versus off colonies, which supports the findings of previous researchers that prairie dog grazing induces changes in the plant community. The data gathered in this study show that prairie dogs create heterogeneity at both the colony (via grazing) and sub-colony scales (via mound-building), while harvester ants’ contribute additional heterogeneity at the nest scale.

 

Literature Cited

 

Carlson, D.C., and E.M. White. 1987. Effects of prairie dogs on mound soils. Soil Science Society of America Journal 51:389–393.

 

Clark, P.J., and F.C. Evans. 1954. Distance to nearest neighbor as a measure of spatial relationships in populations. Ecology 35:45–453.

 

Coffin, D.P., and W.K. Lauenroth. 1990. Vegetation associated with nest sites of western harvester ants (Pogonomyrmex occidentalis Cresson) in a semiarid grassland. American Midland Naturalist 123:226–235.

 

Crist, T.O., and J.A. MacMahon. 1992. Harvester ant foraging and shrub-steppe seeds: Interaction of seed resources and seed use. Ecology 73:1768–1779.

 

Davidson, D.W., Inouye, R.S., Brown, J.H. 1984. Granivory in a desert ecosystem: Experimental evidence for indirect facilitation of ants by rodents. Ecology 65:1780–1786.

 

Hobbs, R.J. 1985. Harvester ant foraging and plant species distribution in annual grassland. Oecologia 67:519–523.

 

Jones, C.G., Lawton, J.H., and M. Shackak. 1997. Positive and negative effects of organisms as physical ecosystem engineers. Ecology 78:1946–1956.

 

Rogers, L.E. 1974. Foraging activity of the western harvester ant in the shortgrass plains ecosystem. Environmental Entomology 3:420–424.

 

Ryti, R.T., and T.J. Case. 1986. Overdispersion of ant colonies: A test of hypotheses. Oecologia 69:446–453.

 

Terranella, A.C., Ganz, L., and J.J. Ebersole. 1999. Western harvester ants prefer nests sites near roads and trails. The Southwestern Naturalist 44:382–383.

 

Whicker, A.D. and J.K. Detling. 1988. Ecological consequences of prairie dog disturbances. BioScience 38:778–785.

 

 

Email questions to Chrissy Alba-Lynn.