Fire interacts with ecosystem carbon in North American fire-grazing rangelands through processes that span chemical forms, soil depths, and temporal horizons. We synthesize empirical evidence on fire-driven carbon redistribution across the continental region where frequent fire and bison grazing historically co-occurred. Across these rangelands, temperate grasses allocate roughly 75–80% of biomass belowground, outside the combustion zone. Herbaceous aboveground biomass typically recovers within one to three growing seasons. Incomplete combustion produces pyrogenic carbon (PyC) amounting to ∼12% of the carbon emitted by fire. The stable component of that PyC persists in soil on centennial to millennial timescales. Frequent burning can reduce bulk soil organic carbon relative to fire exclusion, especially in dry systems, while shifting carbon inputs toward pyrogenic material. Across the Great Plains, shorter modeled fire intervals are associated with a larger mineral-associated fraction and a smaller particulate fraction, whereas the pyrogenic fraction changes little. Where fire is excluded, woody plant encroachment shifts a larger share of ecosystem carbon into aboveground biomass exposed to fire, drought, and dieback, while altering hydrologic function, biodiversity, and resilience to disturbance. The fire-grazing interaction distributes combustion and recovery across a shifting patch mosaic in which litter retention, grazing pressure, and belowground carbon inputs vary among patches. Carbon accounting frameworks that measure only stock size can mistake the carbon vulnerability created by fire suppression for carbon gain, inadvertently penalizing disturbance that sustains persistent soil carbon in these systems.
Mediterranean rangelands provide a substantial part of the annual diet of small ruminants in the extensive livestock systems of southern Europe. Woody vegetation, plant diversity and strong seasonal and structural heterogeneity complicate the estimation of the feeding value of these pastures. Conventional approaches based on botanical composition or herbaceous-layer indicators provide only a partial assessment of feed resource diversity, quality and nutritional role. We propose an alternative approach where we build a classification, based on near-infrared spectroscopy applied to 304 samples representative of the resources selected by sheep and goats in Mediterranean rangelands. Bite types were identified through behavioural observations and sampled across three sites in southern France during spring and autumn over two years. Spectra from dried and ground samples were pretreated and analysed using principal component analysis followed by hierarchical clustering. The resulting clusters separated resources according to plant organs and displayed consistent chemical and structural characteristics. Together, they formed a gradient ranging from protein-rich and highly digestible young organs, to fibre-rich and structurally dense resources. The classification was evaluated through cluster stability analyses, spatio-temporal cross-validations and projection of an independent dataset, complemented by qualitative assessment from farmers and technical advisors. Results revealed stable cluster cores and biologically-meaningful transition zones, particularly for reproductive organs such as grass ears and legume pods. Projection of the independent dataset showed strong overlap with the reference spectral space, supporting transferability beyond the initial sampling conditions. This approach provides an operational framework for integrating pastoral resource diversity into rangeland assessment and feeding management.
Successful rangeland revegetation on western rangelands can be challenging with natural and human-related influences. Defoliation from a wide range of vertebrate and invertebrate grazers can reduce seedling growth and increase plant mortality, thereby impeding restoration efforts. Understanding how herbivory can impede plant establishment and growth may help predict restoration outcomes and create effective solutions for rangeland restoration. Bottlebrush squirreltail (Elymus elymoides (Raf.) Swezey), a native perennial grass species often used in seed mixes, stabilizes soils, provides quality forage, and competes with annual plants. The purpose of this study was to identify the impact of herbivory on bottlebrush squirreltail seedling damage and mortality. This study was conducted on the Utah Test and Training Range (UTTR) in western Utah, USA, tracking herbivory, seedling emergence, and timing and cause of seedling death. We placed cameras in 28 randomly placed plots arranged in a randomized split-plot design with fenced and unfenced plots and seeded with two rows of bottlebrush squirreltail. We tracked individual seedlings and recorded their condition (alive, dead, grazed or damaged), comparing establishment and survival between fenced and unfenced plots. Seed predators reduced initial seedling establishment in unfenced plots by four times ( P = 0.0 0 02). Seedlings were seven times more likely to survive in fenced vs. unfenced plots. Of total seedling mortality, 73.6% of seedling death was caused by herbivory from black-tailed jackrabbit (Lepus californicus), invertebrate herbivores, and Botta's pocket gopher (Thomomys bottae). Strategies to mitigate the effect of herbivores should be considered to increase seeded plant establishment during restoration efforts. These results highlight the broader need for restoration projects in arid and semi-arid ecosystems to account for herbivore communities when planning seeding projects, as early-stage herbivory can strongly limit plant establishment across many species and ecosystems. (c) 2026 The Society for Range Management. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Rangeland ecosystems store substantial terrestrial carbon and contribute to climate mitigation, yet their spatial heterogeneity, dynamic disturbance regimes, and multiple carbon pools make accurate quantification challenging. This introduction outlines and compares methodological approaches for measuring aboveground biomass, soil organic carbon, and fire-derived pyrogenic carbon across rangelands, with emphasis on how sampling design, sensor platforms, modeling frameworks, and accounting tools scale across ecological gradients. We outline field-based protocols, remote sensing technologies, process-based simulators, and machine learning upscaling strategies, highlighting trade-offs in resolution, cost, uncertainty, and spatial-temporal coverage. Particular attention is given to how these methods complement one another across spatial and temporal scales, including the conversion factors and sampling assumptions used to translate biomass estimates into carbon stocks. We then compare practical considerations across spatial and temporal scales to guide method selection, and connect them to rangeland carbon accounting, monitoring, and management by showing how method selection influences assessment reliability and outcomes. By integrating ecological insight, technological innovation, and standardized measurement, this overview provides a practical foundation for designing robust monitoring programs and reducing uncertainty in rangeland carbon assessments.
The introduction and proliferation of invasive annual grasses have increased wildfire size and frequency, contributing to the degradation of sagebrush steppe rangelands in the Great Basin of the western United States. The availability of timely and accurate burn severity data following a wildfire is critical for making well-informed postfire management decisions to address this degradation. However, most commonly used burn severity metrics were developed in the context of high biomass, high cover forested systems, which limits their utility in sagebrush steppe. The goal of this paper is to provide an overview of the current state of burn severity assessment in sagebrush steppe communities and to highlight areas for improvement. First, we compared common burn severity metrics, finding differences among them. We then assessed their utility in shrublands and found that most metrics fail to adequately capture key ecological conditions that drive postfire management decisions, such as perennial grass mortality or fine-scale heterogeneity in burn severity. Consequently, we advocate for the development of a standardized procedure that characterizes the state of the postburn plant community in the context of potential successional trajectories under different environmental and management scenarios. Finally, we provide a suite of ecological burn severity indicators that serve as a starting point in capturing the response of rangelands to fire. Most postfire restoration decisions and underlying burn severity estimates in rangelands are made during the weeks following wildfire. However, the effects of wildfire are not always readily apparent immediately after fire, necessitating an adaptive sampling procedure that allows for reassessment of site conditions and adjustments to management strategies. A standardized procedure designed specifically for rangelands is important because it will enable rapid assessment of the indicators relevant to postfire decision-making.
