A critique of the historical-fire-regime concept in conservation.

Prescribed fire is widely accepted as a conservation tool because fire is essential to the maintenance of native biodiversity in many terrestrial communities. Approaches to this land-management technique vary greatly among continents, and sharing knowledge internationally can inform application of prescribed fire worldwide. In North America, decisions about how and when to apply prescribed fire are typically based on the historical-fire-regime concept (HFRC), which holds that replicating the pattern of fires ignited by lightning or preindustrial humans best promotes native species in fire-prone regions. The HFRC rests on 3 assumptions: it is possible to infer historical fire regimes accurately; fire-suppressed communities are ecologically degraded; and reinstating historical fire regimes is the best course of action despite the global shift toward novel abiotic and biotic conditions. We examined the underpinnings of these assumptions by conducting a literature review on the use of historical fire regimes to inform the application of prescribed fire. We found that the practice of inferring historical fire regimes for entire regions or ecosystems often entails substantial uncertainty and can yield equivocal results; ecological outcomes of fire suppression are complex and may not equate to degradation, depending on the ecosystem and context; and habitat fragmentation, invasive species, and other modern factors can interact with fire to produce novel and in some cases negative ecological outcomes. It is therefore unlikely that all 3 assumptions will be fully upheld for any landscape in which prescribed fire is being applied. Although the HFRC is a valuable starting point, it should not be viewed as the sole basis for developing prescribed fire programs. Rather, fire prescriptions should also account for other specific, measurable ecological parameters on a case-by-case basis. To best achieve conservation goals, researchers should seek to understand contemporary fire-biota interactions across trophic levels, functional groups, spatial and temporal scales, and management contexts.

Tracking postfire successional trajectories in a plant community adapted to high-severity fire.

In order to develop management strategies that maintain native biodiversity in plant communities adapted to high-severity fire, an understanding of natural postfire succession in the target ecosystem is essential. Detailed information on fire effects is lacking for the sand pine (Pinus clausa [Chapm. ex Engelm.] Vasey ex Sarg.) scrub of the southeastern United States, limiting our ability to decide how and when to apply prescribed fire in this ecosystem. We studied the effects of fire-severity heterogeneity on sand pine scrub following a 4700-ha wildfire in Florida's Juniper Prairie Wilderness Area (USA). We identified four levels of fire severity (unburned, low, moderate, and high) and three pre-burn stand conditions (sapling, mature, and senescent). Study plots were established in each severity-stand-class combination, and were sampled at one and two years following fire. Nonmetric multidimensional scaling (NMS) ordination was applied in order to identify differences in community composition and successional trajectories in each of the stand-class-fire-severity combinations. NMS analyses indicated a shift in dominance between the dominant understory oak species, from Quercus myrtifolia Willd. to Quercus geminata Small, as sand pine basal area increases. Our ordination and regression results showed that Q. myrtifolia was the most aggressive colonizer of postfire open space, which is an important structural and habitat component of a sand pine scrub. Successional trajectories were heavily influenced by Quercus myrtifolia Willd. and were more uniform in the mature class than in the senescent class, probably due to more consistent overstory basal area. In both mature and sapling stands, herbaceous species cover was highest in moderate-severity plots. Woody-debris load varied significantly with stand age, fire severity level, and time. Sand pine seedling recruitment was highest in mature stands burned at high severity, while sapling and senescent stands exhibited much lower sand pine seedling recruitment rates at all levels of fire severity. The observed differences in seedling recruitment are expected to influence the progressive development of vertical structure and composition in the sand pine forest, leading to community differences that will persist and influence the effects of subsequent disturbances.