Date of Award
Doctor of Philosophy
Past grassland restoration efforts in the Prairie Pothole Region (PPR) of North America often aimed to restore nesting habitat for waterfowl species. A low-diversity non-native grass and forb seeding mix, known as Dense Nesting Cover (DNC), was frequently used in restoration, and was believed to benefit the broad diversity of wildlife dependent upon restored grasslands. However, grasslands restored with DNC often experience high rates of invasion by non-native vegetation species, and maintenance of these restorations is expensive and requires intensive management. More recently, high-diversity native seeding has been explored as a restoration strategy to decrease the cost of DNC maintenance and provide resources for a greater number of grassland dependent fauna. However, there are gaps in the information concerning how grassland-dependent fauna respond to these restorations, and as such, there is hesitation among land managers to invest in the initial higher cost of high-diversity native seeding mixes until faunal responses are known. The goals of this research were to (1) understand how vegetation diversity, richness, and cover type, variables directly influenced in reseeding restoration, impact grassland-dependent faunal communities; (2) understand how vegetation structure influences the grassland dependent faunal communities to inform post-restoration management activities; and (3) understand how waterfowl and grassland passerines respond to grassland vegetation, structure and landscape variables, in efforts to understand which restoration strategies may be most beneficial to the broader grassland bird community. To perform this study, I selected 26 study sites representing a gradient of vegetation species richness and three cover types; unseeded native grassland (Native), low-diversity and non-native Dense Nesting Cover (DNC), and high-diversity native seed mix (HDM). At each study site, I documented the responses of grassland arthropods, small mammals, grassland birds to vegetation cover type, richness, diversity and structure. I also assessed how vegetation cover type influenced differences in community structure of each of the taxonomic groups. Additionally, because I suspected that bird species are also likely to respond to landscape-level and nest-site level habitat characteristics, I accounted for these variables in our bird models. In Chapter One of this dissertation, I studied the effects of vegetation species richness, diversity and structure on grassland arthropod communities at 23 of our study sites. Arthropod sampling was conducted during July of 2016 using pan traps and sweep net surveys. The goal of using two methods of evaluation was to account for a greater proportion of species in the arthropod community and target taxa that inhabit different parts of the vegetation (i.e., pan traps are better for surveying pollinators and ground dwelling arthropods, whereas sweep net surveys better target arthropods that live higher in the vegetation such as leafhoppers). A total of 25,521 arthropods representing 107 taxonomic families were collected. Vegetation richness and diversity, cover type, percent live vegetation and native cover, and litter depth were important predictors of arthropod community measures, and multivariate analysis of the arthropod community indicated significant differences between Native and DNC sites, which was explained by differences in vegetation richness, percent forb cover, and litter depth. Results suggest that species richness of grassland restoration seeding mixes likely impacts arthropod richness and diversity, and DNC does not produce arthropod communities similar to native grassland. Additionally, because vegetation structural variables were important determinants of arthropod community measures, grassland management practices will influence the resulting arthropod community and influence the success of grassland restoration seed mixes. Results demonstrate the potential for successful restoration outcomes using high-diversity seed mixes, and indicate that low-diversity, non-native seed mixes, such as DNC, do not restore native grassland arthropod communities. In Chapter Two, I assessed the responses of the grassland small mammal community to vegetation cover type, as well as richness, diversity and structure of the vegetation community. During July 2014 to 2016, I used Sherman live traps to survey grassland small mammal communities on 24 study of my study sites. Small mammal abundance was highest at low-diversity DNC sites, and lowest in native grassland. Small mammal diversity was highest at HDM sites and lowest at DNC restoration sites. I selected three focal species to investigate small mammal responses to grassland vegetation including Peromyscus spp., Microtus spp., and Ictidomys tridecemlineatus. Abundances of the different focal taxa were influenced by different vegetation structural variables. Peromyscus spp. abundance was negatively influenced by percent native vegetation cover, Microtus spp. abundance showed yearly variation and was impacted positively by litter depth and negatively by vegetation richness, and Ictidomys tridecemlineatus abundance was influenced by cover type. Small mammal communities of DNC sites differed from Native sites, but HDM was not different from Native or DNC. Ictidomys tridecemlineatus abundance was higher at Native and HDM sites, while DNC sites had higher Peromyscus spp. abundance. Results indicate species-specific management is required to meet small mammal management goals, and diversity of the restoration seed mix is likely to influence grassland small mammal communities. Additionally, DNC is not supporting small mammal communities similar to what is observed at Native grassland sites. In Chapter Three of this dissertation, I investigated the responses of grassland bird communities to vegetation richness, diversity, structure on all study 26 sites. I conducted nest searches for waterfowl during May to July of 2014 to 2016, and searched for grassland passerines and other non-waterfowl bird species during May to July of 2015 and 2016. In 2016, I conducted additional point count surveys to detect bird species that may have been unaccounted for in nest searches. A total of 998 waterfowl nests of nine species, 282 passerine nests of nine species, and 32 nests of 10 other bird species were located. Five hundred and nineteen birds of 20 species were encountered in point count surveys. Richness and diversity of bird nesting and point count communities differed among cover types, and sites restored with HDM seeding had lower bird richness and diversity. Richness and diversity of the vegetation positively influenced the richness and diversity of both the bird nesting and point count communities. Multivariate analysis of the nesting communities indicated significant difference among HDM, Native, and DNC sites, and this was best explained by percent native vegetation cover, vegetation species richness, and vegetation density. Multivariate analysis of the point count communities did not reveal significant difference among cover types, but the structural variables vegetation density and litter depth were important explanatory variables for the ordination. Results study indicate that manipulation of vegetation species richness and diversity that occurs in high-diversity restoration has the potential to influence the grassland bird community, whereas greater vegetation richness and diversity are likely to support more diverse bird communities, but management of vegetation density will be an important management consideration. In Chapter Four, I investigated species-specific responses to vegetation diversity, richness, structure, and landscape variables in order to determine whether management efforts anticipated to benefit a particular taxonomic group of grassland bird species (i.e., waterfowl) are likely to meet the needs of other grassland-dependent birds (i.e., passerines) and to inform management efforts aimed at conserving and creating nesting habitat for grassland waterfowl and passerines. I used generalized linear mixed-models (GLMM) to determine which variables at the landscape, patch, and nest-site level were most influential to the nesting density and daily survival rates (DSR) of our primary focal species. Focal species included five waterfowl species: Blue-winged Teal (Spatula discors; N=365), Gadwall (Mareca strepera; N=173), Mallard (Anas platyrhynchos; N=302), Northern Pintail (Anas acuta; N=64), and Northern Shoveler (Spatula clypeata, N=61), and three species of grassland passerines: Bobolink (Dolichonyx oryzivorus; N=31), Clay-colored Sparrow (Spizella pallida; N=190), and Savannah Sparrow (Passerculus sandwichensis; N=20). Cover type only impacted two of our focal species, Blue-winged Teal and Clay-colored Sparrow, and density of both species was highest at HDM sites and lowest at DNC sites. Vegetation diversity and richness did not impact the nesting density of the majority of our focal species, but Northern Shoveler nesting density was negatively associated with vegetation species richness. DSRs of Northern Pintail and Gadwall were positively associated with vegetation species richness, and Bobolink DSR was negatively influenced by vegetation diversity. Responses to the other landscape-level and vegetation covariates of interest were mixed among species, and little consistency was observed across waterfowl or passerine species. However, at the landscape-level, several waterfowl and passerine species had lower DSRs associated with landscape-level components that contribute to the fragmentation of grasslands including wooded edges, crop fields, and developed areas (i.e., roads and buildings). Results indicated that waterfowl and passerine species respond to different parts of grassland vegetation, and successful management and restoration for nesting grassland birds will require the development of heterogenous habitat that provides resources for the diversity of grassland birds nesting at restored sites. Findings suggest that restoration that involves manipulating diversity, richness, or origin of the vegetation in grasslands through reseeding should be expected to directly influence grassland faunal communities. Because the taxa investigated in this study are responding to different aspects of the vegetation structure, management efforts should be multifaceted, and account for the differing needs of the diversity of grassland fauna dependent upon restored sites. Results also indicated that greater vegetation species richness and diversity results in a more rich and diverse wildlife community, and high-diversity native reseeding should be expected to promote more rich and diverse faunal communities, but successful implementation of high-diversity native reseedings will necessitate proper management of vegetation structure. Finally, all taxonomic groups studied demonstrated positive responses to native grassland habitat, regardless of vegetation richness and diversity. Therefore, preservation of native grassland and promotion of habitat heterogeneity should be prioritized in grassland conservation and restoration efforts.
This dissertation is only available for download to the SIUC community. Current SIUC affiliates may also access this paper off campus by searching Dissertations & Theses @ Southern Illinois University Carbondale from ProQuest. Others should contact the interlibrary loan department of your local library or contact ProQuest's Dissertation Express service.