Submitted to: Rangeland Ecology and Management
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 10/11/2021
Publication Date: 1/24/2022
Citation: Polley, H.W., Kolodziejczyk, C.A., Jones, K.A., Derner, J.D., Augustine, D.J., Smith, D.R. 2022. UAV-enabled quantification of grazing-induced changes in uniformity of green cover on semi-arid and mesic grasslands. Rangeland Ecology and Management. 80:68-77. https://doi.org/10.1016/j.rama.2021.10.001.
Interpretive Summary: Ecological processes on grazing lands depend on the extent to which the surface area of these ecosystems is covered by living (green) vegetation together with spatial uniformity in green coverage. However, green plant coverage is difficult to measure frequently and at the small spatial scales required to evaluate spatial patterns in living plant cover. We describe a technique for rapid airborne (unmanned aerial vehicle; UAV) measurements of spatial coverage of living vegetation at small spatial scale. We then apply the technique to determine effects of locally relevant grazing treatments on spatial variation in coverage of living vegetation in pastures located in northeastern Colorado, USA (shortgrass prairie) and central Texas, USA (improved pasture). Variation in living plant cover was well-explained by differences in sunlight reflectance from the surface of pastures at both sites. Measurements of reflectance across entire pastures using a UAV revealed that green plant coverage was similar on average among shortgrass pastures with heavy, moderate, and light grazing treatments but reduced in improved pasture by a grazing treatment in which cattle were moved monthly among pastures (rotational grazing) rather allowed to graze year-round in the same pastures (continuous grazing). Heavy grazing in the shortgrass and rotational grazing in improved pasture increased spatial uniformity in coverage of living vegetation such that adjacent areas were more similar in green plant cover under heavy grazing (shortgrass) and rotational grazing (improved pasture) than other grazing treatments. Our results indicate that UAV-enabled remote sensing provides for rapid measurement of living plant cover at sufficiently small spatial resolution to characterize spatial variation in green plant cover on grasslands. The measurements we describe provide a science-based approach to evaluate impacts of grazing treatments and other management activities on grazing lands.
Technical Abstract: Coverage of living (green) vegetation influences rangeland processes and biodiversity but remains a challenge to quantify frequently and at small spatial grain. We describe a technique for rapid airborne (unmanned aerial vehicle; UAV) measurements of continuous spatial coverage of living vegetation at a resolution (spatial grain) of 8 cm ground sample distance. We then applied this technique at the pasture (paddock) scale in two contrasting grassland ecosystems (semi-arid shortgrass steppe in northeastern Colorado, USA and mesic grassland in central Texas, USA) to determine effects of locally relevant grazing treatments on spatial variability and dependence (pattern) in fractional coverage of green vegetation (fractional vegetation cover; FVC). Site-specific regression models developed using reflectance in visible and near-infrared (NIR) wavebands explained 90% and 89% of variance in FVC for semi-arid and mesic grassland, respectively. Mean FVC was similar among shortgrass steppe pastures differing in grazing treatment (light or heavy grazing by cattle, moderate cattle grazing with prairie dogs present). In contrast, FVC was lower with rotational compared to continuous, year-long grazing in the mesic grassland. Heavy grazing in shortgrass steppe and rotational grazing in mesic grassland increased the spatial uniformity in FVC by reducing spatial variability and increasing spatial dependence in FVC, the latter by increasing the similarity in FVC values among spatially separated patches. UAV-enabled remote sensing provides for rapid measurement of FVC at sufficiently small spatial grain to characterize pasture-level values of FVC and spatial variation in FVC. Results can be used to 1) detect within growing season temporal trends in grassland vegetation status, and 2) evaluate the effectiveness of management actions intended to alter spatial variation in FVC to achieve conservation or diversity objectives. Enhanced capacity to monitor FVC at small spatial grain promotes adaptive management of grasslands.