A fire burns and we work the flames. But next time I hope to find a way to cross the road, to understand and defuse his anger. Cleckley, William, O. Ecosystem Restoration Workshop.
Accessed 3 March Giltzenstein, J. Historic Fire Regimes of Longleaf Pine. The Encyclopedia of Southern Fire Science. Accessed 22 October Hoyle, Zoe. Revitalizing Wiregrass at Fort Gordon. Accessed 4 March Kush, John, J. Restoring Fire to the Longleaf Pine Forest. Fire Science Brief Mackowiak, Cheryl. Burning Questions Addressed in Wiregrass Production. Accessed 2 March Noss, Reed, F. North Carolina Agric.
Lemon, P. Tall Timbers Fire Ecol. Lewis, W. Machlis, L. San Francisco: Freeman Metz, L. Mendoza, A. Tropical Ecology in press. Muller, C. Torrey Bot. Mutch, R. Nelson, W. Odum, E. Parrott, R. Unpublished M. Ralston, C. In: Prescribed Burning Symposium Proc. Roberts, P. Steward, K. Vlamis, J. Range Manage. Volg, R. Wahlenberg, W. Lathrop Pack For. Wells, B. Wells, C. In: Prescribed Burning Symp. Download references. You can also search for this author in PubMed Google Scholar.
Reprints and Permissions. Fire and soil-plant nutrient relations in a pine-wiregrass savanna on the coastal plain of North Carolina. Oecologia 31, 27—44 All relevant data are within the manuscript and its Supporting Information files. Restoring fire regimes is a major goal of biodiversity conservation efforts in fire-prone ecosystems from which fire has been excluded. In the southeastern U. In these savannas, frequent fires that support biodiversity are driven by vegetation-fire feedbacks.
Understory grasses are key components of these feedbacks, fueling the spread of fires that keep tree density low and maintain a high-light environment. When fire is reintroduced to long-unburned sites, however, remnant populations of bunchgrasses might experience high mortality from fuel accumulation during periods of fire exclusion.
Our objective was to quantify fire effects on wiregrass Aristida beyrichiana , a key component of vegetation-fire feedbacks, following 16 years without fire in a dry pine savanna typically considered to burn every 1—3 years. We examined how wiregrass size and fuel duff depth and presence of pinecones affected post-fire survival, inflorescence and seed production, and seed germination. Wiregrass exhibited high survival regardless of size or fuels.
The ability of bunchgrasses to persist and reproduce following fire exclusion could jumpstart efforts to reinstate frequent-fire regimes and facilitate biodiversity restoration where remnant bunchgrass populations remain. Fire regimes are important drivers of biodiversity patterns in fire-prone ecosystems [ 1 ].
In many regions of the world, variations in climate and human activity have caused fire regime shifts ranging from fire suppression to increased fire frequency and altered seasonality [ 2 — 4 ].
These changes have affected biodiversity in many ways, including shifts in plant functional groups [ 5 ] and animal species distributions [ 6 ]. Promoting biodiversity by restoring and maintaining fire regimes in flammable ecosystems, particularly where fires have been suppressed, is therefore a major goal of biodiversity conservation efforts [ 7 ].
Reintroduction of fire regimes in fire-prone pine savannas of the southeastern U. These ecosystems sustain an exceptional species richness and endemism [more than 50 species per m 2 ; 8 ], making it crucial to restore the frequent ground fires every 1—3 years that maintain them [ 9 — 11 ]. Pine savannas are currently experiencing a long period of fire exclusion [ 12 ], which has led to visible changes in ecosystem structure and reductions in biodiversity [ 13 ].
A fundamental step towards their restoration is promoting vegetation-fire feedbacks [ 10 , 14 ], whereby flammable vegetation, such as grasses, fuel the spread of fires that keep midstory and overstory tree cover low.
In turn, the high-light environment is conducive for the growth of understory vegetation that fuels future fires, resulting in a frequent fire regime every 1—3 years. If fire is excluded, tree recruitment into the overstory increases, shading out the flammable understory and disrupting vegetation-fire feedbacks [ 14 , 15 ].
Restoration efforts often target grasses in the understory to support fire spread and reinstate vegetation-fire feedbacks. Aristida beyrichiana wiregrass is a perennial, endemic C4 bunchgrass that commonly dominates pine savanna understories, contributing to vegetation-fire feedbacks and biodiversity dynamics [ 16 — 18 ].
In savannas where fire has been excluded however, the size and number of wiregrass individuals are greatly reduced [ 19 , 20 ], which could be attributed to competition, lower light levels from overstory shading, or heavy litter [ 21 — 23 ].
Moreover, because sexual reproduction in wiregrass is primarily stimulated by fire, lack of fire inhibits wiregrass seed production [ 16 ]. Restoring populations of this dominant species and key component of pine savanna function is therefore a conservation and restoration challenge [ 24 ].
Although restoring frequent fire to pine savannas should increase biodiversity in the long-term, the initial state of the ecosystem could alter the restoration trajectory [ 25 ]. During long periods of fire exclusion, woody species typically increase in cover and shade out the understory, resulting in herbaceous species declines [ 23 ].
The fuel buildup e. For example, long fire-free intervals allow fuels to accumulate, which can increase tree mortality through increased depth and duration of heating at the bases of trees [ 26 ].
Similarly, pine litter and duff accumulation around wiregrass plants could increase the amount and duration of heating and cause unusually high wiregrass mortality [ 27 , 28 ]. Previous studies suggest that wiregrass responses to fire after long, unburned periods should differ from responses to frequent return intervals.
In a stand that had been fire-suppressed for at least 35 years, [ 13 ] found that wiregrass declined after a reintroduction fire and still had not reached pre-burn levels of cover after eight years. None of these studies, however, examined the burning of accumulated fuels as a mechanism by which reintroduction fires affect wiregrass plants. Our objective was to examine the effects of burning duff and pinecone fuels on remnant wiregrass populations when fire is reintroduced to long-unburned sites.
Weather conditions and burning techniques were favorable to cautiously achieving a low-intensity fire, which is typical of reintroduction fire objectives. We hypothesized that pinecones and deeper duff would increase plant mortality and result in decreased reproduction and seed germination.
The site encompasses more than 3, hectares of lowland swamps and marshes and upland ecosystems that include xeric sandhill pine savannas, hammocks, mixed-pine hardwoods, and isolated wetlands. We conducted this experiment in a longleaf pine Pinus palustris sandhill savanna that was last burned in June 16 years prior; Fig 1A.
The unit is located on excessively drained, Candler fine sand, 0—5 percent slopes, with an overstory dominated by longleaf pine, and the midstory dominated by hardwood trees, mainly turkey oak Quercus laevis. The groundcover was composed of scattered wiregrass and pineywoods dropseed Sporobolus junceus and a continuous layer of duff decomposed material between the intact litter layer and the soil surface and fuels such as pine needles and hardwood leaves, interspersed with pinecones and small branches.
This location would likely have experienced regimes of frequent 1—3 yr natural fires, primarily in the late spring and early summer, coincident with increasing lightning frequencies and dry weather [ 12 ]. Wiregrass is a warm-season, caespitose, C4 bunchgrass that is a major component of pine savanna understory communities in the southeastern Coastal Plain [ 12 ].
With meristems several centimeters or more below the soil surface [ 30 ] and thin leaves and aerated architecture [ 16 ], the species is highly flammable and regrows rapidly after fire [ 16 , 17 ]. Reproduction is primarily vegetative, although sexual reproduction can be stimulated by fire and other disturbances [ 16 , 31 ]. The species rarely flowers without fire, however, but even after fire not every plant will necessarily produce flowering culms. We haphazardly identified and tagged wiregrass individuals across a range of sizes i.
We considered an individual to be a compact unit of ramets at least 10 cm distant from other groups of ramets. We calculated plant basal area as the area of an ellipse, using the longest axis of the base of the plant and the corresponding perpendicular axis. Each plant was tagged by pinning a uniquely numbered aluminum tag into the soil next to the individual. In demographic studies, the unit of replication is the individual on which vital rates survival, reproduction are measured, not the plot or burn unit level [ 32 ].
Replication at the unit level would have entailed burning another unit on a different day, in which case units would not have been true replicates owing to intra-day differences in weather conditions, human-ignition patterns, and fire behavior.
Fires are heterogeneous across various scales, and thus is it is likely each of our replicate plants experienced different fires even though they were burned on the same day, providing random variation in fire characteristics across treatments. Plants were categorized into three fuel treatments: high duff, low duff, and low duff with added pinecones. We measured duff depth as the average depth at four equidistant points around the base of each individual.
We did not add pinecones to plants in high duff because there were fewer individuals available in this category. Differences in litter depth between treatments were negligible. The unit was burned with a backing fire lit on 26 June Weather conditions and burn techniques were typical of those preferred when initially restoring prescribed fires to long-unburned pine savannas. In November , we surveyed plant survival and reproduction.
For each reproductive plant, we counted all culms and measured the length of the inflorescence on every flowering culm. To quantify fecundity, we calculated the relationship between inflorescence length and number of glumes on 20 flowering culms and used this relationship to calculate number of seeds per plant.
In November , we also collected seeds from plants by hand-stripping them from inflorescences. In March , we examined seed viability by separating filled seeds from empty and smut-infected seeds using the press test [ 33 ]. We counted seeds from each treatment, separating filled from empty and infected seeds until we obtained 96 filled seeds in each treatment. The boxes were placed on a lab bench at room temperature.
We periodically monitored and removed germinated seeds in each box and ended the experiment after two months. We also placed the empty and infected seeds in similarly prepared boxes to confirm our seed quality assessment. Because nearly all plants survived, it did not make sense to statistically analyze survival. To investigate whether fuel treatments affected seed production, we used the 77 plants which produced flowering culms.
We used negative binomial regression MASS package; [ 34 ] after testing the overdispersion of the number of culms data with respect to a Poisson distribution by comparing the residual deviance to the residual degrees of freedom [ 35 ].
A comparison of the full model for culm production to the null model was significant, indicating that the predictors improved the model. We analyzed seed germination data using logistic regression on the probability of germination lme4 package; [ 36 ].
We used the status of filled seeds i. There was not enough variation among boxes to include germination box as a random effect. All analyses were conducted in R [ 37 ]. Wiregrass plants exhibited low overall mortality regardless of size or fuel treatment. Of the plants burned across the three treatments, all but four individuals survived. Wiregrass fecundity varied by size.
The probability of reproduction significantly increased with basal area Table 1 , with plants greater than cm 2 more likely to be reproductive Fig 2. In reproductive plants, the number of culms per plant also increased significantly with basal area, regardless of fuel treatment Table 1. The germination of filled seeds did not differ between fuel treatments Table 1.
Individuals were subject to different fuel treatments during a reintroduction fire in a long-unburned pine savanna. Probability of reproduction increased with size.
0コメント