Benjamin F. Tracy

The Ecology of the Earth's Grazing Ecosystems Profound functional similarities exist between the Serengeti and Yellowstone

A s recently as 150 years ago, most of Earths grasslands supported large migratory populations of hoofed herbivores belonging to the Artiodactyla, Perissodactyla, and Proboscidea-that is, ungulates. These herbivores included bison (Bison bison) on the North American plains, saiga antelope (Saiga tatarica) on the Eurasian steppe, wildebeest (Connochaetes taurinus) and zebra (Equus burchelli) on the African savanna, and the ecologically equivalent kangaroos (Macropodidae) on the Australian savanna. As a result of the post-industrial global expansion of cropland and cattle ranching, most grasslands grazed by Pleistocene megaherbivores were eliminated. Today, they are restricted to the worlds few large grassland reserves that protect all seasonal ranges of the animals. In this article, we describe profound functional similarities between two of the most celebrated of these remaining habitats, the Serengeti ecosystem of east Africa and Yellowstone National Park (YNP) of the North American intermountain west, which previously have been considered to

Herbivore influence on soil microbial biomass and nitrogen mineralization in a northern grassland ecosystem: Yellowstone National Park

Abstract Microorganisms are largely responsible for soil nutrient cycling and energy flow in terrestrial ecosystems. Although soil microorganisms are affected by topography and grazing, little is known about how these two variables may interact to influence microbial processes. Even less is known about how these variables influence microorganisms in systems that contain large populations of free-roaming ungulates. In this study, we compared microbial biomass size and activity, as measured by in situ net N mineralization, inside and outside 35- to 40-year exclosures across a topographic gradient in northern Yellowstone National Park. The objective was to determine the relative effect of topography and large grazers on microbial biomass and nitrogen mineralization. Microbial C and N varied by almost an order of magnitude across sites. Topographic depressions that contained high plant biomass and fine-textured soils supported the greatest microbial biomass. We found that plant biomass accurately predicted microbial biomass across our sites suggesting that carbon inputs from plants constrained microbial biomass. Chronic grazing neither depleted soil C nor reduced microbial biomass. We hypothesize that microbial populations in grazed grasslands are sustained mainly by inputs of labile C from dung deposition and increased root turnover or root exudation beneath grazed plants. Mineral N fluxes were affected more by grazing than topography. Net N mineralization rates were highest in grazed grassland and increased from dry, unproductive to mesic, highly productive communities. Overall, our results indicate that topography mainly influences microbial biomass size, while mineral N fluxes (microbial activity) are affected more by grazing in this grassland ecosystem.

The role of ammonia volatilization in controlling the natural 15N abundance of a grazed grassland

Although the variation in natural 15N abundance in plants and soils is well characterized, mechanisms controlling N isotopic composition of organic matter are still poorly understood. The primary goal of this study was to examine the role of NH3 volatilization from ungulate urine patches in determining 15N abundance in grassland plants and soil in Yellowstone National Park. We additionally used isotopic measurements to explore the pathways that plants in urine patches take up N. Plant, soil, and volatilized NH3δ15N were measured on grassland plots for 10 days following the addition of simulated urine. Simulated urine increased 15N of roots and soil and reduced 15N of shoots. Soil enrichment was due to the volatilization of isotopically light NH3. Acid-trapped NH3δ15N increased from −28‰ (day 1) to −0.3‰ (day 10), and was lighter than the original urea-N added (1.2‰). A mass balance analysis of urea-derived N assimilated by plants indicated that most of the N taken up by plants was in the form of ammonium through roots. However, isotope data also showed that shoots directly absorbed 15N – depleted NH3-N that was volatilized from simulated urine patches. These results indicate that NH3 volatilization from urine patches enriches grassland soil with 15N and shoots are a sink for volatilized NH3, which likely leads to accelerated cycling of excreted N back to herbivores.

SHIFTS IN GRASSLAND INVASIBILITY: EFFECTS OF SOIL RESOURCES, DISTURBANCE, COMPOSITION, AND INVADER SIZE

There is an emerging recognition that invasibility is not an intrinsic community trait, but is a condition that fluctuates from interactions between environmental forces and residential characters. Elucidating the spatiotemporal complexities of invasion requires inclusion of multiple, ecologically variable factors within communities of differing structure. Water and nutrient amendments, disturbance, and local composition affect grassland invasibility but no study has simultaneously integrated these, despite evidence that they frequently interact. Using a split-plot factorial design, we tested the effects of these factors on the invasibility of C3 pasture communities by smooth pigweed Amaranthus hybridus L., a problematic C4 forb. We sowed seeds and transplanted 3-week old seedlings of A. hybridus into plots containing monocultures and mixtures of varying composition, subjected plots to water, soil disturbance, and synthetic bovine urine (SBU) treatments, and measured A. hybridus emergence, recruitment, and growth rate. Following SBU addition, transplanted seedling growth increased in all plots but differed among legume and nonlegume monocultures and mixtures of these plant types. However, SBU decreased the number and recruitment rate of emerged seedlings because high residential growth reduced light availability. Nutrient pulses can therefore have strong but opposing effects on invasibility, depending on when they coincide with particular life history stages of an invader. Indeed, in SBU-treated plots, small differences in height of transplanted seedlings early on produced large differences in their final biomass. All facilitative effects of small-scale disturbance on invasion success diminished when productivity-promoting factors were present, suggesting that disturbance patch size is important. Precipitation-induced invasion resistance of C3 pastures by a C4 invader was partly supported. In grazed grasslands, these biotic and environmental factors vary across scales and interact in complex ways to affect invasibility, thus a dynamic patch mosaic of differential invasion resistance likely occurs in single fields. We propose that disturbance patch size, grazing intensity, soil resource availability, and resident composition are inextricably linked to grassland invasions and comment on the utility of community attributes as reliable predictors of invasibility. Lastly, we suggest temporal as well as spatial coincidences of multiple invasion facilitators dictate the window of opportunity for invasion.

Seedbank diversity in grazing lands of the Northeast United States.

We evaluated the species composition of soil seed banks from 9 farms (36 pastures total) located in the northeast United States. Our objective was to quantify the soil seed bank composition of pastures managed for intensive grazing and hay production. Seeds from pasture soils were allowed to germinate in a greenhouse under natural light conditions. Seedlings were identified as they germinated, and the experiment was concluded after 4 months. Germinable seed was dominated by annual (40%) and perennial (23%) forbes most of which contributed little useful forage for cattle. Perennial grasses (11%), except for bluegrass (Poa pratensis L.), were largely absent from the terminable seed bank, while legumes (19%) were more abundant. Seed bank species composition showed little similarity (44%) to the existing vegetation. Exceptions were bluegrass, white clover (Trifolium repens L.), and common dandelion (Taraxacum officinale Weber ex Wiggers). These species were abundant in both the germinable seed bank and existing vegetation on most pastures. Overall, our study suggests that seed banks in these northeast pastures support abundant white clover and bluegrass seed, both of which are important forages for cattle. Soil seed banks, however, will not supply a diverse assemblage of useful forages. If a manager seeks to establish diverse, mixed-species pasture, then re-seeding pastures with desired mixes may be the best option. DOI:10.2458/azu_jrm_v53i1_tracy

Disturbance persistence in managed grasslands: shifts in aboveground community structure and the weed seed bank

The length of time and form in which disturbances persist in systems depends on the intensity and frequency of disturbance and on the abilities of resident species to recover from such events. In grazed grasslands, trampling by large mammalian herbivores can periodically facilitate weed establishment by exposing patches of bare ground but whether an intense soil disturbance event results in a temporary increase in weed abundance or a persistent weed problem remains unclear. In May 2002, cattle trampling following heavy rain caused severe damage to nine-month old, rotationally grazed, cool-season pastures (Midwest USA). In September 2002, we compared the aboveground composition of paddocks (i.e., fenced pasture sections) that were heavily disturbed to those that received no damage. Relative to undisturbed paddocks, forage species relative cover was 17% lower in disturbed paddocks, and weed species and bare ground relative cover was 61% and 100% higher, respectively. By September 2004, paddock types did not differ in all aboveground community components. However, the abundance and species richness of weed seeds in the soil seed bank averaged respectively 82% and 30% higher in disturbed paddocks between 2003 and 2004. These findings indicate that a spatially extensive, intense soil disturbance event may soon become undetectable in components of aboveground pasture structure but can persist as an augmented weed seed bank. Because of high weed seed bank longevity, disturbances to formerly disturbed pastures would likely result in higher weed recruitment, with more species represented, than in those which lack previous disturbance. Disturbance history may thus be a useful predictor of weed community composition following subsequent disturbance. Based on empirical data supporting this proposition, we recommend that grassland managers explicitly incorporate disturbance history into dynamic management planning and do not rely exclusively on aboveground characters to evaluate the invasion status or colonization potential of an area by undesirable plants. We emphasize that the ecological legacies of past soil disturbance events cannot only influence the contemporary patterns and processes of grasslands, but importantly, affect their compositional trajectories following subsequent perturbation.

ELK GRAZING AND VEGETATION RESPONSES FOLLOWING A LATE SEASON FIRE IN YELLOWSTONE NATIONAL PARK

We studied responses of aboveground production, grazing by elk and the availability of eight elements (Ca, Fe, K, Mg, N, Na, P, Zn) in a Yellowstone National Park sagebrush grassland following a fire in 1992. We compared four areas of differing fire history: (1) an area burned in 1992, (2) an area burned in 1992 and 1988, (3) an area burned in 1988 and (4) an area with no recent fire history. The year after burning, graminoids produced more aboveground biomass on burned areas compared with unburned areas. Forages growing on burned soils were highly concentrated with all nutrients, except Ca, when elk grazed the site. Despite nutrient-rich forage on burned areas, elk consumed little forage in these areas. We hypothesized that elk ate less than expected in burned areas because of a large bloom of lupine (Lupinus sericeus), which may be unpalatable to elk.