Next generation of land system science: Integrating meso-scale analysis and UAS remote sensing in changing plant communities of the United States' Southern Great Plains

As concerns about desertification, climate change, economic pressures, and reductions in global biodiversity become more prevalent, so too does the importance of research in the socio-ecological resilience of dryland agricultural communities. Cimarron County, Oklahoma and Union County, New Mexico have historically faced cyclical drought and land degradation as a challenge to agriculture. More recently, the expansion of woody plant species in rangelands threatens to destabilize grassland communities, limit forage for cattle, and reduce water resources. This thesis uses a Land System Science (LSS) approach to study how vegetation communities have changed in the two counties across multiple scales. By utilizing household surveys with ranchers, ground-level biodiversity and rangeland health assessments, unmanned aerial systems (UAS/drone) imagery, and satellite-based remote sensing, a holistic, integrative picture of the interrelated factors affecting woody plant encroachment (WPE) is presented. Household surveys indicated that agriculturalists are keenly aware of nuisance species on their property, with many taking actions to reduce their abundance. Woody species cited by landowners including one-seed juniper, Great Plains yucca, broom snakeweed, and cane cholla were reliably detected and identified to species level through ground and UAS observations. While there was strong agreement in land-cover estimates between the two methods, ground observations were more useful in measuring herbaceous species biodiversity while UAS was advantageous in surveying woody species across larger areas. WPE was affected by land-use factors, and was more severe in grazed pastures. By contrast, WPE was less severe on Conservation Reserve Program lands and plots with prior herbicide applications. Environmental factors also played a role, with greater WPE vulnerability in areas of rugged terrain, sandy soils, and cooler, drier climates. While LSS research has a history of using both field observations and satellite remote sensing to connect "people to pixels," the present investigation demonstrates of the utility of UAS imagery as a powerful bridge in scale. As studies in land systems evolve and become more important, UAS is poised to serve as a potential rapid assessment method which can provide context and additional detail to coarser- or finer-scale analyses