Typescript. Includes vita and abstract. Includes bibliographical references (leaves 146-166). Also available online in Scholars' Bank; and in ProQuest, free to University of Oregon users.

Thesis (M.S.)--Virginia Polytechnic Institute and State University, 1993. / Vita. Abstract. Includes bibliographical references (leaves 182-218). Also available via the Internet.

Thesis (M.S.)--Wisconsin. / Includes bibliographical references (leaves 101-114).

Thesis (M.S.)--Miami University, Dept. of Zoology, 2007. / Title from first page of PDF document. Includes bibliographical references (p. 20-26).

Tallgrass prairies have some of the deepest and most fertile topsoil on earth. Widespread conversion of these grasslands to agriculture has decreased soil Carbon (C) storage by exacerbating erosion and disrupting aggregates that protect C from decomposition, coupled with lower plant C inputs. Thus, a primary goal of some grassland restorations is to improve soil structure and functioning. Conversion of cultivated systems to perennial grasslands often increases soil C, microbial biomass, and soil aggregate size and stability. A few studies have documented changes in soil microbial community structure after restoration. The objective of this work was two fold: 1) to explore the importance of soil texture and 2) determine plant diversity effects on recovery of soil biotic and abiotic properties. In the first study changes in soil microbial phospholipid fatty acid (PLFA) profiles and soil aggregates were examined in two 0-19 year chronosequences of restored grasslands in Nebraska on soils differing in texture, silty clay loam (SCL) and loamy fine sand (LFS), and compared them to native prairie. Soil was collected from the 0-10 cm soil depth at each site in May of 2007 and 2008. The SCL chronosequence exhibited increases in total PLFA biomass P<0.05, r2=0.29), PLFA richness (P<0.0001, r2=0.25), fungi (P<0.0001, r2=0.65), fungal:bacterial ratio (P<0.0001, r2=0.67), Gram (+) bacteria (P=0.02, r2=0.22), Gram (-) bacteria (P=0.05, r2=0.16), and actinomycetes (P=0.02, r2=0.23). Average soil aggregate diameter also increased (p=0.0002, r2=0.52). However, LFS sites showed no change across the chronosequence for any parameter. Total PLFA biomass (ANOVA, P<0.0001), richness (P<0.0001), and fungi (P=0.005) were greater on SCL restorations than LFS, but LFS had greater fungal:bacterial ratios (P=0.02). Soil microbial groups and soil aggregates were highly correlated, especially in the LFS choronosequence indicating that structural recovery is key to microbial community recovery. The second study investigated high diversity restorations with low diversity restorations on silty clay loam. In this study, high diversity and low diversity restorations in southeast Nebraska, aged 4 and 8 years were compared. The quantity of forbs seeded was too low and high diversity communities were a mixture of dominant C4 grasses (Andropogon gerardii Vitman, Schizachyrium scoparium (Michx.) Nash, Panicum virgatum L., Bouteloua curtipendula (Michx.) Torr. and Sorghastrum nutans (L.) Nash) and subdominant C3 grasses (Elymus canadensis L., Pascopyrum smithii (Rybd.) A. Löve, and Elymus virginicus L.). Eight year old plantings had greater root biomass, root C storage, root C:N ratio (P<0.05 for all), microbial biomass (low diversity only, PC<0.1, PN<0.05), PLFA richness (low diversity only, P<0.05), mycorrhizal fungi (P<0.05), and C mineralization (low diversity only, P<0.05) than 4 year old plantings. Low diversity plantings, which contained almost exclusively dominant C4 prairie grasses, had greater root C storage (P<0.1), mycorrhizal fungi (8 years only, P<0.1), and C mineralization (8 years only, P<0.05). Thus, C4 grasses and their associated arbuscular mycorrhizal fungi seem to drive recovery of soil C, soil respiration, and soil microbial communities over time. Overall, this work indicates that rates and success of belowground recovery are dependent on both abiotic and biotic factors in restoration. Restored plant communities affected soil recovery as dominant C4 grasses appeared to drive belowground recovery, but recovery depended on soil texture.

Prairies of the Pacific Northwest are threatened systems, with only ~2% of historic land remaining. The combined risk of global climate change and land use change make these systems a high conservation priority. However, little attention has been paid to the microbiota. Fungal endophytes are ubiquitous in plants and are important in ecosystem functioning and host dynamics. To understand fungal community assembly, we used high-throughput sequencing to investigate the composition of fungal foliar endophyte communities in two native, cool-season (C3) bunchgrasses along a natural latitudinal gradient. We quantified the importance of host, host traits, climate, edaphic factors, and spatial distance in microbial community composition. We found that spatial distance was the strongest predictors of endophyte community, while host traits (e.g., plant size, density) and abiotic environment were less important for community structure. These findings underline the importance of dispersal in shaping microbial communities. This thesis includes previously unpublished, co-authored material.

Invertebrates influence primary productivity and nutrient cycling in ecosystems. They are also important links between producer and higher trophic levels. Despite their important role in terrestrial ecosystem function, invertebrates are frequently overlooked in ecological restorations. Thus, the objective of this research was to quantify how belowground macroinvertebrate ecosystem engineers and communities change over time following ecological restoration and examine whether the source of dominant plant species and the composition of non-dominant plant species influence aboveground macroinvertebrates community structure in restored prairie. A chronosequence design (space for time substitution) was used to determine the role of restoration age, plant community, and soil structure on the recovery of two belowground macroinvertebrate ecosystem engineers (Chapter 2). Ants and earthworms were sampled from cultivated fields, grasslands restored for 1-21 years, and native prairie. Earthworm abundance increased with time since cessation of cultivation, concomitant with prairie establishment. The abundance and biomass of ants were more related to the structure of restored plant communities than time since restoration. The dominance of exotic earthworms, and a generalist ant species in these restorations, coupled with their known capacity to alter soil properties and processes, may represent novel conditions for grassland development. The same chronosequence of agricultural fields, restored prairies, and prairies that were never cultivated was used to address the second objective of this research, which was to quantify how the belowground macroinvertebrate community composition changed in response to ecological restoration and whether the communities became representative of undisturbed (“target”) communities. Macroinvertebrate communities in the two remnant prairie sampled were distinct from restorations and continuouslycultivated fields. The macroinvertebrate communities in prairie that had never been cultivated were also distince from each other, indicating a “target” community is hard to define. Belowground macroinvertebrates changed in a trajectory that was not representative of either remnant prairie, but was representative of the an average of both remnant prairies. Thus, if you reconstruct prairie from cultivated soil conditions (“build it”), macroinvertebrates will colonize (“come”), but attaining a community representative of a specific target may require introduction from that target. Colonization of macroinvertebrates in restorations aboveground are most likely to be influenced by aspects of the plant community. A third objective of this research was to quantify whether variation in dominant species (cultivars vs. local ecotypes) and composition of subordinate species (local species pools) influence the composition of aboveground macroinvertebrates. Macronvertebrate abundance, richness, diversity, trophic groups and community composition in late summer did not vary between prairie restored with cultivar and local ecotypes of the dominant grasses. This was observed in two field experiments. The species pool treatment did influence the macroinvertebrate community, as one species pool had slightly higher morphospecies diversity and hymenopterans that the other two species pools. This was likely due to the presence of an ant-tended legume, Chamaecrista facsciculata Michx., in one species pool. Overall, this research demonstrates that time since the cessation of disturbance (cultivatation) and plant communities influence macroinvertebrate communities in restored prairie. Restored prairies in the Midwest are likely to be colonized by exotic earthworms and cosmopolitan ants. More research is needed to reveal how they influence ecosystem functioning.Belowground, macroinvertebrate communities may not represent restoration “targets” and these “targets” may be hard to define if remnants are rare or there is a high degree of spatial variation on the landscape. Variation in plant communities above ground appears to influence the structure of aboveground macroinvertebrate communities more than variation within dominant species. Whether this aboveground variation is reflected is reflected belowground deserves further investigation.

Woody vegetation encroachment has become a major threat to tallgrass prairie streams mainly because of fire suppression. This process converts prairie streams from open to closed canopy systems. The effects of these riparian changes are poorly understood, but the relative importance of basal resources presumably shifts from primarily autochthonous to allochthonous with increasing canopy cover, potentially altering macroinvertebrate functional structure and production. To assess the effects of woody vegetation encroachment on stream ecosystem structure and function, riparian trees were removed from two headwater stream reaches on the Konza Prairie Biological Station (KPBS) in eastern Kansas. Experimental stream reaches were compared to streams with naturally open and closed canopies before and after the manipulation. Benthic organic matter and macroinvertebrates were collected monthly from each reach for one year before and one year after woody vegetation removal. Total community production in canopy removal reaches ranged from 8.9-10.2 g AFDM m-2 y-1 before riparian removal, and this increased significantly to 13.4-14.5 g AFDM m-2 y-1 after riparian removal. Scraper production in canopy removal reaches was 2.8-3.9 g AFDM m-2 y-1 before riparian removal, and increased significantly to 6.0-8.7 g AFDM m-2 y-1 after riparian removal, presumably due to enhanced food availability. Total community production in naturally open reaches ranged from 7.6-12.6 g AFDM m-2 y-1 before riparian removal and decreased to 6.5-9.8 g AFDM m-2 y-1 after riparian removal. Riparian forest removal altered macroinvertebrate production and functional structure, but higher macroinvertebrate production in canopy removal reaches compared to naturally open reaches suggested natural conditions were not restored one year after riparian removal. However, macroinvertebrate communities in naturally open and canopy removal reaches became more similar after riparian removal. Functional structure, based on production, in naturally open and canopy removal reaches after riparian removal was dominated by scrapers (45-60% of total production), with similar proportions of collector-gatherers (12-26%) and predators (15-25%). Collector-filterers and shredders contributed < 9% of total production in naturally open and canopy removal reaches after riparian removal. Results demonstrate that woody vegetation encroachment and riparian forest removal significantly influence tallgrass prairie stream structure and function. Information from this study can help inform and guide management, restoration, and conservation of remaining tallgrass prairie streams.

In one of the first reintroductions of a semi-wild population of bison (Bison bison) east of the Mississippi River, The Nature Conservancy reintroduced bison to the Nachusa Grasslands in northern Illinois in October 2014. Given the novelty of such efforts, questions remain regarding how human activity and prairie management affect bison habitat selection. My objective was to quantify bison habitat selection during 2014-16 to address literature gaps while investigating seasonal and annual changes. In October 2014, The Nature Conservancy collared 7 female bison with Lotek Iridium TrackM 3D and 4D collars programmed to take hourly locations. I randomly selected 1 bison location each hour to represent herd location and divided the resulting locations by season. Using resource selection functions, I compared the influence of land cover type, fire management, and concentrated human activity on habitat selection across seasons and at 2 scales: patch and enclosure. Land cover was consistently the best-fit model across seasons and scales with the exception of Winter 2016 at the enclosure scale. Bison strongly selected for partial restorations but showed no strong seasonal land cover selection trends. Bison selected for recently-burned prairie with the exception of Summer 2015 at the enclosure scale. At the patch scale, bison selected for areas closer to disturbed areas but did not show a strong trend at the enclosure scale. With a better understanding of how bison grazing is influenced by restoration management, wildlife managers can make better-informed decisions regarding bison restoration and public use.

Typescript (photocopy). / Digitized by Kansas Correctional Industries