Repository logo

Graduate Degree Program in Ecology

Permanent URI for this community

https://hdl.handle.net/10217/100395
These digital collections include theses, dissertations, faculty publications, photographs, and datasets from the Graduate Degree Program in Ecology.

Browse

Browsing Graduate Degree Program in Ecology by Subject "adaptive management"

Now showing 1 - 2 of 2
  • Thumbnail Image
    ItemOpen Access
    Grazing for grassland birds: assessing how management and environmental conditions affect abundance in Colorado's shortgrass steppe
    (Colorado State University. Libraries, 2018) Davis, Kristin Petersilia, author; Aldridge, Cameron, advisor; Augustine, David, advisor; Skagen, Susan, committee member; Noon, Barry, committee member
    Rangelands are temporally and spatially complex socio-ecological systems on which the predominant land use is livestock production. In North America, rangelands also contain approximately 80% of remaining habitat for grassland birds, a guild of species that has experienced precipitous declines since the 1970s. Some evidence suggests livestock grazing can be managed to benefit certain grassland bird species by generating the vegetation structure/density they prefer. These benefits, however, appear to be ecosystem-specific, are equivocal even for species predicted to benefit from grazing (e.g., those that prefer short, sparse vegetation), are rarely considered in conjunction with the full suite of environmental factors known to influence grassland birds (e.g., precipitation, vegetation composition), and are poorly understood for species breeding in the shortgrass steppe. To address these research gaps, I evaluated how two grazing management systems – continuous, season-long grazing and adaptive, rest-rotational grazing – and environmental characteristics (e.g., vegetation structure, vegetation composition, precipitation and topography) affected grassland bird abundance in Colorado’s shortgrass steppe. I fit hierarchical distance sampling models that accounted for temporary emigration in a Bayesian framework to five years of point count data (2013-2017) for five focal grassland bird species collected from an ongoing grazing experiment on the Central Plains Experimental Range, a USDA experimental range site, in northeastern Colorado. I first examined grazing impacts on grassland bird abundance in conjunction with ecological sites, which represent local soil and plant characteristics. When grazing management was evaluated in conjunction with spatial variation in ecological sites, I found two of our five focal bird species responded to grazing management: McCown’s longspur abundance decreased and grasshopper sparrow abundance increased in pastures rested from grazing for the entire previous year. In addition, abundances of all focal species varied across ecological sites. To evaluate environmental influences on grassland bird abundance, I first used model selection (deviance information criterion; DIC) to identify temporal scales of precipitation and spatial scales of topography that best predicted grassland bird abundance. I then fit two environmental models – 1) a full environmental model with the best topographic and precipitation scale and vegetation structure and composition covariates for each species, and 2) a full environmental model where I replaced the precipitation scale with a categorical effect of year. Finally, I used model selection (DIC) to evaluate the predictive capacity of my grazing models compared to my environmental models for each species. I found precipitation had the largest magnitude of effect on the abundance of lark bunting and grasshopper sparrow. Vegetation structure had the largest magnitude of effect on the abundance of McCown’s longspur, and vegetation composition had the largest magnitude of effect on the abundance of western meadowlark. Vegetation structure and precipitation had the largest magnitude of effect on horned lark abundance. Precipitation strongly and positively affected the abundance of all focal species except western meadowlark, where the effect was strong but negative. Vegetation structure strongly affected the abundance of all species except grasshopper sparrow, and characterized species by their preferred vegetation structure/density (e.g., sparse grass-preferring species’ abundances decreased and dense grass-preferring species’ abundances increased with vegetation structure). Responses to vegetation composition were generally species specific, but cover of standing dead vegetation and shortgrass affected the abundance of two of my five focal species. Only McCown’s longspur responded to topography. Although my focal species responded to multiple environmental characteristics considered in my environmental models, the grazing models had greater predictive capacity than the environmental models for some of my focal species that prefer more moderate to tall/dense vegetation on the landscape – lark bunting and western meadowlark – and the grazing and environmental models had equal predictive capacity for grasshopper sparrow. This study suggests the factors influencing grassland bird abundance in the shortgrass steppe are complex and diverse. Grazing management alone can predict patterns in grassland bird abundance, but these species also responded to specific components of vegetation composition, vegetation structure and precipitation. Thus, grazing impacts on grassland birds may be context-dependent and managers should consider local environmental conditions (e.g., ecological sites, precipitation conditions, vegetation composition) when developing grazing management for grassland birds. My study occurred during historically wet and average precipitation years, so repeating these analyses in drought would reveal additional and important insight into drivers of grassland bird abundance in the shortgrass steppe. Ultimately, my results suggest cattle production on rangelands can continue to support human economic needs while also supporting grassland bird populations in the shortgrass steppe.
  • Thumbnail Image
    ItemOpen Access
    Improving state-and-transition models for management of sagebrush steppe ecosystems
    (Colorado State University. Libraries, 2015) Tipton, Crystal Yates-White, author; Fernandez-Gimenez, Maria, advisor; Ocheltree, Troy, advisor; Aldridge, Cameron, committee member
    The sagebrush biome was once the most widely-distributed in North America, but has recently experienced range reductions of up to 45% and has been considered one of the most endangered ecosystems in the United States (West 1983, Noss et al. 1995, Miller et al. 2011). Management for multiple land-use goals in this biome is complex, requiring an intricate understanding of biotic and abiotic interactions, their responses to disturbance, and the potential for catastrophic ecosystem shifts in response to stress. State-and-transition models (STMs) illustrate the complex relationships between ecosystem components and convey both equilibrial and non-equilibrial dynamics, in a conceptual, visual framework (Westoby et al. 1989, Walker and Westoby 2011). Recognizing their potential to guide both research and management decision-making, the Natural Resource Conservation Service, U.S. Forest Service, and Bureau of Land Management recently signed an interagency agreement to develop and use STMs to guide rangeland management decision-making nation-wide (Caudle et al. 2013). The growing popularity of STMs has brought them under increased scrutiny (Knapp et al. 2011, Tidwell et al. 2013). Common criticisms of STMs include: 1) reliance on insufficient empirical datasets or knowledge-based data prone to bias; 2) failure to explicitly identify the spatial and temporal scale of the model and the limitations of its generalizability; 3) dependency on assumptions of linear, reversible succession toward a climax reference community while ignoring the roles of non-equilibrial change, multiple disturbance types and abiotic gradients in shaping system resilience; 4) focus on the practices associated with structural change, while overlooking the ecological process feedbacks that influence disturbance response; 5) failure to validate STMs by testing model predictions. Motivated by the need for improved sagebrush-steppe management tools, my thesis addresses these criticisms and challenges by exploring new approaches to build and refine STMs. The first chapter provides a review of sagebrush-steppe ecosystem dynamics, paradigms of vegetation change, and the application of STMs to natural resource management. The second chapter presents work to apply a collaborative, iterative approach proposed by Kachergis et al. (2013c) that integrates knowledge-based and empirical datasets to develop an STM for a Wyoming big sagebrush-steppe ecosystem in Moffat County, Colorado. The third chapter presents a pilot project to revise an existing STM by incorporating the role of specific ecological processes (nitrogen cycling) into a state transition. I conclude that the approaches employed here can address many of the challenges and criticisms of current STMs, but should be coupled with rigorous experimental testing of model assumptions and uncertainties and long-term monitoring of experimental outcomes. In addition, collaborative approaches should take care to carefully balance resource limitations with the desire to include a broad base of stakeholders and research interests, carefully manage stakeholder expectations, and explicitly define success in terms of both the collaborative process and the scientific outcomes.