BASAL RESOURCE COMPOSITION AND MACROINVERTEBRATE COMMUNITY STRUCTURE IN TALLGRASS, MIXED-GRASS, AND SHORTGRASS PRAIRIE HEADWATER STREAMS

North American prairie headwater streams are highly threatened and relatively poorly studied. Most studies on prairie streams have occurred at the Konza Prairie Biological Station, a tallgrass prairie Long Term Ecological Research site in the Flint Hills ecoregion near Manhattan, KS. According to the Stream Biome Gradient Concept, several ecosystem factors vary along a gradient from more allochthonous forested streams to more autochthonous desert streams, with grassland streams often intermediate in several key ecological factors including litter inputs, primary production, and invertebrate abundance and biomass. However, few studies have examined the degree of variation that exists within prairie headwater streams, and whether this variation occurs along a longitudinal gradient as well, with more mesic tallgrass prairie streams differing from more xeric shortgrass prairie streams, and mixed-grass sites intermediate between the two. I examined thirteen prairie headwater stream sites in the central United States from 2014 to 2017. My objective was to determine whether basal resource composition – including standing stocks of coarse particulate organic matter (CPOM), fine particulate organic matter (FPOM), and very fine particulate organic matter (VFPOM), sestonic and benthic chlorophyll-a levels, and sources of CPOM – differed significantly among streams in tallgrass, mixed-grass, and shortgrass prairie regions. In addition, I examined whether invertebrate communities differed among tallgrass, mixed-grass, and shortgrass prairie regions, and whether this was reflected in the functional feeding group composition, habit composition, voltinism, and dispersal ability of invertebrate communities. There were no significant differences in total CPOM, FPOM, and VFPOM standing stocks among regions. However, CPOM composition did differ with region, with tallgrass sites having higher standing stocks of leaf litter, but lower standing stocks of grass litter and macrophyte litter than the other regions. Benthic chlorophyll-a did not differ significantly among regions, but there were lower sestonic chlorophyll-a levels in tallgrass sites. Given higher light availability and nutrient levels in shortgrass and mixed-grass streams, lack of stable substrata may be limiting benthic algae in these regions. Invertebrate abundance and biomass were highest in mixed-grass sites and lowest in tallgrass sites, with shortgrass sites intermediate. Mixed-grass sites also had significantly higher Shannon diversity and taxa richness than tallgrass sites. A NMDS revealed that sites differed in overall community structure. Functional feeding group composition did not differ significantly across regions, with collector-gatherers, followed by predators, dominating in all sites. High Predator-Prey Index (predator biomass: other invertebrate biomass) values in each region indicates strong top-down pressure and high turnover rates of prey taxa. While there was a weak correlation between leaf litter AFDM and invertebrate community structure, the correlation between latitude and longitude and invertebrate community structure was much higher, indicating that basal resources may not be the main drivers in these systems. Invertebrate habit composition did not differ with region; all regions were dominated by taxa preferring fine substrata (burrowers and sprawlers). All regions showed selection against semivoltine invertebrates and were dominated by high dispersing insect taxa, though the proportion of the insect community that consisted of high dispersers did not differ significantly with region. My results suggest that generalizing about prairie streams based on studies from one or a handful of sites may not be prudent, at least for some aspects. The higher proportion of autochthonous inputs in shortgrass and mixed-grass regions may drive increased invertebrate abundance, biomass, richness, and diversity, but the relative hydrologic stability of the mixed-grass sites might also explain these results or have an interactive or additive relationship with primary production. Overall, the link between basal resources and communities across prairie types was somewhat weak, and all regions were dominated by collector-gatherers with rapid life-cycles and high dispersal abilities, indicating that disturbance may be a more important community filter than basal resource composition. Streams in all three regions have highly variable hydrology, and this may be an overriding factor that results in similarity in communities