Associating plant community composition and forage quality with soil fertility and seedbank dynamics in managed Kansas rangelands

Abstract

This thesis investigates population and community-level ecology in eastern Kansas rangelands, emphasizing the interconnected effects of fire, grazing, and herbicide treatments on aboveground plant community composition, forage quality, soil fertility, and the soil seedbank. Chapter 1 provides a literature review exploring how fire and grazing shape plant community composition and seedbank dynamics. Chapter 2 examines relationships among aboveground plant community composition, forage quality, soil nutrients, and seedbank composition in a two-season field study comparing areas dominated by native perennial grasses ("grass-dominated") with areas dominated by early-successional forbs, woody plants, or bare ground ("disturbed"). Disturbed areas exhibited higher levels of soil nutrients (nitrate, phosphorus, potassium), higher soil organic matter and carbon-to-nitrogen ratio, greater seedbank abundance and diversity, higher crude protein, and greater residue cover. Analyses of covariance were also performed to evaluate associations between soil and plant characteristics without differentiating grass-dominated from disturbed areas. These analyses were mostly non-significant. These findings illustrate the potential for qualitative vegetation surveys to provide some insight into complex ecological processes. Chapter 3 used Uncrewed Aerial Vehicles (UAVs) to apply herbicide treatments and monitor herbicide efficacy. We evaluated impacts on mature plants, including Baldwin’s ironweed (Vernonia baldwinii), goldenrod (Solidago spp.), leadplant (Amorpha canescens), western ragweed (Ambrosia psilostachya), and white sagebrush (Artemisia ludoviciana) as well as potted seedlings, including alyssum (Lobularia maritima), Baldwin’s ironweed, common evening primrose (Oenothera biennis), common milkweed (Asclepias syriaca), and sericea lespedeza (Lespedeza cuneata). A lower rate of PastureGard®HL (0.21 kg ae ha⁻¹ triclopyr + 0.067 kg ae ha⁻¹ fluroxypyr) effectively suppressed some weeds at the seedling stage while minimizing non-target impacts. The recommended rate (0.63 kg ae ha⁻¹ triclopyr + 0.21 kg ae ha⁻¹ fluroxypyr) controlled both mature and seedling plants. Chapter 4 is a pilot study aiming to develop a method for seedbank quantification in sericea lespedeza. The seeds of this species exhibit physical dormancy, which must be broken prior to greenhouse emergence assays. We tested several seed coat scarification methods including heating to 79 C, 107 C, or 204 C; 2 or 20 freeze-thaw cycles; and mechanical abrasion with sandpaper. This pilot study suggested that 20 freeze-thaw cycles would be the preferred method, although further replication is needed. Collectively, these findings emphasize the importance of integrating aboveground with belowground data, and management practice with ecological process. This research supports more informed, integrated vegetation and weed management strategies for sustaining rangeland productivity and resilience.