Plague and the black-tailed prairie dog: an introduced disease mediates the effects of an herbivore on ecosystem structure and function

Abstract

I investigated the effects of black-tailed prairie dogs (Cynomys ludovicianus) on plant communities and nutrient cycling on the shortgrass steppe of Colorado, and how colony age and activity (active or plague-extirpated) influence the effects of prairie dogs in this system. Plague, caused by the bacterium Yersinia pestis, was first noted in prairie dog colonies in the 1940s. I suggest that this introduced disease has altered the role of prairie dogs in shaping ecosystem structure and function on the shortgrass steppe. Plague causes periodic die-offs of entire colonies. During periods of inactivity, vegetation may recover from the effects of heavy grazing. Using 25 years of size and activity data about each colony on the Pawnee National Grassland, I found that most colonies (98%) experienced a plague-outbreak within 15 years of activity, and that over half (55%) of all colonies were inactive for at least five years after an outbreak. Furthermore, the probability of a colony attaining its pre-outbreak size within 5 years was low (28%). I first studied five colonies (5-17 years of age) both before (1997-1999) and after (2000-2001) a plague-outbreak. Mean total plant canopy cover and cover of functional groups (forb, graminoid, shrub, cactus) did not differ significantly (P ≤ 0.05) between on- and off-colony sites either before or after the plague-outbreak. However, mean total live plant biomass and graminoid biomass were lower on than off of colonies before, but not after the plague-outbreak, while forb biomass did not differ significantly between on- and off-colony sites either before or after the outbreak. Using herbivore exclosure cages on the same five colonies, I estimated the relative amounts of plant biomass removed by cattle and prairie dogs on the shortgrass steppe in 1998 and 1999. I found that cattle and prairie dogs had similar and additive effects on plant biomass. Cattle consumed about the same amount of plant biomass on and off of colonies. Also, total aboveground plant production was not significantly different between on- and off-colony sites, but graminoid production decreased and forb production increased with prairie dog grazing. To further elucidate how colony age and activity influence the role of prairie dogs on the shortgrass steppe, I studied plant communities on and off of three young (3-7 years since colonization), three old (~ 20 years since colonization), and three plague-extirpated (7-12 years old when extirpated) colonies from 2002 through 2004. I found that both young and old colonies, but not plague-extirpated colonies, had a significantly shorter canopy and lower graminoid cover and biomass than paired uncolonized sites. Although all three colony types showed the same trend, bare ground cover and forb cover and biomass were significantly greater on- than off-colony for only the old colony type. Biomass of standing dead plants (previous year's growth) and litter were lower only on the old colonies compared to off-colonies. Root biomass was not significantly different on than off of colonies for any colony type. Total plant species richness was greater only on old and plague-extirpated colonies, and greater species richness was due to greater numbers of forb species rather than graminoid species. To understand how aboveground effects of prairie dogs influence belowground processes, I studied differences in N cycling on and off the same young, old, and plague-extirpated colonies in 2003 and 2004. Shoot N concentrations of the dominant graminoid (Bouteloua gracilis) and forb (Sphaeralcea coccinea) were significantly greater on than off of active, but not plague-extirpated colonies. Despite greater shoot N concentrations, total standing crop N was lower on than off of active colonies for B. gracilis because the increase in shoot N was not great enough to offset the decrease in aboveground biomass. Although there was a trend toward higher root N concentration on than off of colonies, it was statistically significant when colonies were considered collectively, but not when considered by colony type. Intermound spaces on prairie dog colonies did not have significantly higher rates of net N mineralization than uncolonized grassland for any colony type, but prairie dog mounds on active colonies had greater rates of N mineralization compared to off-colony sites or intermound spaces on colonies, suggesting that mounds may be sites of high nutrient turnover. Soil organic matter pools (total C and N, particulate organic matter C and N) appear to be relatively unaffected by prairie dogs, with differences seen only in the older colonies. Both mound and intermound areas on old colonies had lower total organic C than off of old colonies. My results suggest that the above- and belowground effects of prairie dogs on the shortgrass steppe are generally similar to, but of lower magnitude than, their effects on the more mesic mixed-grass prairie where they have been most studied. My research also supports the hypothesis that the introduction of plague to North America may be altering the role of prairie dogs in shaping ecosystem structure and function because plague-extirpated colonies were rarely different from their associated off-colony sites. Also, effects of prairie dogs appear to increase with colony age, but because of plague, few colonies now persist for multiple decades.