Abstract
Invasive annual grasses, such as cheatgrass (Bromus tectorum), pose a significant threat to western rangelands by outcompeting native plants and increasing the frequency and intensity of wildfires. Frequent fires have reduced or eliminated native shrubs as annual grasses have invaded rangelands. This research aimed to quantify three aspects of the management of cheatgrass-invaded rangelands: (1) control of annual grasses, native plant community response, and modeled fire behavior after an aerial application of the herbicide indaziflam, (2) flammability of cheatgrass and two native perennial grasses, and (3) classification of shrub canopy volume with structure-from-motion photogrammetry techniques.Invasive annual grasses pose a significant fire risk and can quickly expand, even in low elevation mountain big sagebrush (Artemisia tridentata ssp. vaseyana) plant communities that have naturally high resistance to invasion and resilience to disturbance. Reducing the impact of annual grasses requires an effective tactic for annual grass control. The chosen tactic must minimize any negative effects and be maintained long enough to reduce plants emerging from the soil seedbank. One potential solution is the use of indaziflam, a new pre-emergent herbicide that has demonstrated promising results in reducing annual grasses for at least three years in natural areas throughout the western United States. The first chapter of this dissertation evaluated the efficacy of indaziflam in reducing annual grass foliar cover and its impact on native plant foliar cover and fuel continuity in mountain big sagebrush plant communities near Hailey, Idaho. A total of 19 ha were treated with indaziflam in September 2020, and 32, 900 m2 permanent assessment plots were established within treated and untreated areas. Foliar cover and fuel continuity were measured one and two years post-treatment along 3, 30-m long transects per plot. To better understand the potential impact of control measures on fire behavior, customized fuel beds based on field measurements were developed to model fire behavior across the treated and untreated areas under three different environmental scenarios. The results showed that indaziflam treatment significantly reduced annual grass foliar cover across all plant community types, while not affecting fuel continuity or native perennial plant foliar cover, even in areas with dense shrub cover. Mean untreated annual grass foliar cover was 11% and 38% one- and two-years post-treatment, while treated cover was 4% and 10%, respectively. Total and herbaceous fuel loading did not differ by indaziflam application, and shrubs contributed the most to total fuels. Indaziflam treated areas had slightly lower modeled rates of spread, flame lengths, and reaction intensity than untreated areas, but these differences were only present in areas with low shrub cover. These findings suggest that indaziflam application is an effective tactic for controlling annual grasses with minimal impacts to resident native plants in high resistance and resilience sagebrush communities for at least two years.
Cheatgrass increases fire risk and alters plant communities in the sagebrush steppe grasslands of the Great Basin (USA) and adjacent sagebrush steppe areas, yet no studies have contrasted its flammability to native perennial grasses. The second dissertation chapter focused on the flammability of cheatgrass compared to two native perennial grasses, Columbia needlegrass (Achnatherum nelsonii) and bluebunch wheatgrass (Pseudoroegneria spicata), across a range of fuel moistures. Cheatgrass increased overall mass consumption, flaming duration, and flame length when combined with the two perennial grass species. The extent of flammability increase was influenced by the proportion of cheatgrass in the mixture, with flaming duration and thermal dose being particularly sensitive to cheatgrass fuel moisture. Maximum temperature and flame length during perennial grass combustion remained similar, regardless of the presence of cheatgrass. Furthermore, the flammability of Columbia needlegrass was higher when burned with cheatgrass than anticipated based on the flammability of each species alone. Cheatgrass may pre-heat Columbia needlegrass, leading to increased mass consumption, flaming duration, and thermal dose. This study provides experimental evidence supporting previous qualitative observations of cheatgrass altering fire behavior and increasing fire risk. Cheatgrass increased flammability of these two native perennial bunchgrasses throughout a range of fuel moistures and should be considered in fire management decisions.
Shrub canopy volume is an important ecological indicator in rangeland ecosystems and is closely related to biomass, fuel loading, wildlife habitat, site productivity, and ecosystem structure. Traditional field techniques for estimating shrub canopy volume are tedious and time-consuming and pose challenges that alternate methods may alleviate. The third dissertation chapter explored the suitability of using drone-collected data to estimate shrub canopy volume for seven dominant shrub species within mountain big sagebrush plant communities in southern Idaho, USA. First, height and canopy widths of 103 shrubs of eight species was measured in the field. Then, canopy volume for each individual shrub was re-estimated using two techniques (allometric or volumetric) from a 3D representation of the study area which was created using structure-from-motion photogrammetry. The volumetric method, which involved converting point clouds to raster canopy height models, outperformed the allometric method and was more reproducible and robust to user-related variability. Drone-estimated volume closely matched field-estimated volume (R2 > 0.9) for three larger shrub species: A. tridentata subsp. tridentata, A. tridentata subsp. vaseyana, and Purshia tridentata. These findings demonstrate that drone-collected images can be used to assess shrub canopy volume for at least five upland sagebrush steppe shrub species and support the integration of drone data-collection into rangeland vegetation and fuels monitoring. Additionally, the study demonstrates the potential for automating canopy volume estimates using point cloud-based automatic shrub detection algorithms.
The research presented here provides compelling evidence of the threat posed by invasive annual grasses in sagebrush plant communities with naturally high resistance and resilience. The presence of cheatgrass increases the flammability of two native perennial grasses across various fuel moisture levels. The interaction across fuel types (grasses and shrubs) emphasizes the importance of low cost, accurate assessment of shrub canopy volume through use of unmanned aerial systems. Rangeland condition, fuel abundance and fuel distribution are important parameters measured with this drone-based technique. Mitigating the risk of annual grasses is imperative, and indaziflam is an effective tool for controlling them for at least two years, with minimal impact on native species.