Fire Ecology. Fire is the key ecological disturbance in many Florida ecosystems, including the scrub, sandhill, and flatwoods vegetation that dominates Archbold Biological Station. Fire events, fire suppression, and variations in fire patterns and intensities affect plant communities and populations in many interesting and often complex ways. In this biennial report, we outline some of our research directions in fire ecology.

Community Responses to Fire. Florida scrub community responses to individual fires involve both resprouting of many of the dominant shrubs and seedling recruitment pulses in other species. Most previous work has emphasized the resprouting vs. seeding dichotomy and has not detailed the mechanisms of compositional shifts with fire. In a detailed study of scrub response to a single fire at Lake Wales Ridge State Forest, we documented postburn reductions in subcanopy, shrub, litter, and lichen cover that paralleled increases in the frequency, abundance, and diversity of herbs. Although some scrub herbs resprout, most postburn herbaceous species increases were due to seedling recruitment. Individual species that did resprout had varying rates from 15-98%. This suggests that a dichotomy of resprouters vs. seeders oversimplifies species and community level responses to fire.

Other types of upland vegetation burn more frequently than scrub and also have different vegetation/fire interactions. Seasonal ponds and flatwoods dominated by the cutthroat grass are vulnerable to fire exclusion and drainage, which can cause invasion by bayhead species, as documented in recently published articles by former research assistant Rebecca Yahr and former intern George Landman. Bayhead invasion of seasonal ponds was context dependent, tending to occur most often when ponds and bayheads were in close proximity. Pine flatwoods dominated by wiregrass shift gradually toward palmetto dominance with fire suppression, as detailed in a recent article by former intern Satya Maliakal. Former intern Kurt Reinhart, resampling areas first sampled by Eric Menges, showed that fire-suppressed southern ridge sandhill vegetation can be restored by frequent burns that reduce litter cover, litter depth, and shrub cover. All these frequently burned plant communities have substantial grass coverage that is reduced by fire suppression and, in turn, may increase fire frequency in local landscapes.

We plan to continue our work on fire effects in sandhill vegetation with an experimental burn at Carter Creek (USFWS National Wildlife Refuge). Here, we will integrate community ecology with demographic work on several scrub endemic plants and an experimental introduction of the extremely rare shrub Ziziphus celata into various postburn microsites. Introduced seedlings and seeds will be genotyped with RAPD techniques, in an ongoing project conducted by Carl Weekley in collaboration with Thomas Kubisiak, and Tammera Race.

Population Biology and Population Viability Analyses with Fire. Fire clearly affects many aspects of the demography of scrub plants, so that population viability assessments must explicitly incorporate fire regimes. Pedro Quintana-Ascencio and Menges completed the first population viability analysis for a Florida scrub plant for Hypericum cumulicola. The results suggest that fires every 50 years or less are necessary to avoid local extinctions. We are nearly finished with a population viability analysis of Eryngium cuneifolium, another Florida scrub endemic. Frequent fires with return intervals of 15 years or less are necessary for E. cuneifolium persistence (see Fig.1, page 7). Since rosemary scrub may burn less often, local extinctions and metapopulation dynamics may be the norm for this species. Since other rosemary scrub specialists (e.g. H. cumulicola) thrive with less frequent fires, we believe that variation in fire regimes will allow co-existence and hedge against local extinctions. Further population viability analyses with fire are planned for other Florida scrub species.

The mechanisms behind fire effects are probably many and vary by species. For example, for Eriogonum longifolium var. gnaphalifolium, (scrub buckwheat), former intern Kelly McConnell and Menges have found that top clipping causes a flowering response typically found postfire, while litter removal enhances seedling recruitment (see also; Student Research, page 37). Shrub removal and ash additions had little affect on this species. Warea carteri population increases postfire are consistent with increased germination in litter-free microsites (studied in collaboration with Michael Kelrick).

Fire Intensity. Fire is essential to many ecosystems worldwide, but fire effects may vary greatly with small-scale variations in fire intensity. Unfortunately, various methods to estimate fire intensity are not always congruent. We (Alaä Wally, Weekley, and others, in collaboration with Joan Walker) are comparing fire intensity measured by dataloggers (which provide precise, detailed information on fire temperature and duration, but can only be deployed in a limited area) and more numerous but less precise pyrometers (metal tags with paints that melt at given temperatures) and calorimeters (aluminum cans with water that evaporates proportionately to heat output). In one burn, a combination of pyrometer and calorimeter data explained a large proportion of variation in thermocouple temperature data.

Fire intensity affects some components of vegetation recovery. For example, fire intensity promotes bark beetles, which cause mortality of south Florida slash pine, as detailed in an article in press by Menges and Mark Deyrup. Jose Luis Hierro and Menges, however, found no effects of experimentally increased fire temperatures on the recovery of flatwoods vegetation at the Lake Placid Scrub Preserve.

Fire Management and Adjuncts to Fire. Many land managers would like to manage Florida scrub with substitutions or adjuncts to fire, such as mowing or logging. However, these treatments or pretreatments may not produce ecologically similar responses to fire. Our lab (in collaboration with Dawn Berry and the state agency managing several sites on the Lake Wales Ridge, the Florida Fish and Wildlife Conservation Comm.) is studying the effects of mowing (with a brown tree-cutter) and prescribed fire on rare plant populations and vegetation composition and structure in three central Florida scrub sites. In a 1999 fire at one site, both calorimeters and pyrometers recorded higher maximum temperatures in scrub plots than in mown plots. A similar study involving logging and fire (in cooperation with Florida Div. Forestry; see photo, page 6) has produced an unusually high abundances of several graminoid and herb species in the logged area, but our recent drought has limited the abundance of post-treatment seedlings at all sites. We will continue to study community responses during the coming years.

Gapology. We define gaps as openings among dominant shrubs (e.g. oaks, rosemary) that may have bare sand, lichens, herbs, or shrub stems less than 50 cm tall (see photo, this page). Gaps are important microhabitats for many Florida scrub plants. Not only are species such as E. cuneifolium and H. cumulicola found mainly in gaps, their demographic success is affected by distance from gap edge, the species that forms the edge, and the size of the gap. While it is clear that local distributions are affected by the distributions of gaps in the landscape, it is also possible that narrow endemics and soil specialists are particularly constrained by the larger-scale distribution of gaps. To address this possibility, Maliakal, (Ph.D. Candidate, Louisiana State Univ.), is comparing the importance of gap size on the demography of two Florida scrub endemics primarily restricted to gaps in rosemary scrub (Polygonella basiramia, Lechea cernua) with the demography of two widespread, less specialized congeners.

The persistence of gap specialists in the landscape may be a dynamic result of the distribution of occupied and unoccupied gaps. Using both gap and habitat patch scales, Owen Boyle (Ph.D. candidate, Univ. Wisconsin) is developing a spatially explicit metapopulation model for P. basiramia, including measuring subpopulation growth, extinction, and colonization rates, genetic structure, spatial structure, patterns of gene flow, dispersal, and habitat patch quality.

Metapopulation dynamics, gap dynamics, and fire may also affect spatial and temporal distributions of genetic variation. Following up on more broad-scale surveys of isozyme variation in H. cumulicola, we are collaborating with Rebecca Dolan (see page 9) to track fire-induced temporal and spatial shifts in genetic structure of H. cumulicola in an Archbold burn unit where all xeric gaps have been mapped and measured (see map [b], page 8).

The mechanisms by which gaps create suitable microsites for many scrub species are not well known. Soils in gaps contain more water than soils under shrubs (see Daniel Gagnon, page 9). Gaps also support biological soil crusts, which alter moisture and nutrient dynamics (see Christine Hawkes, page 37). Reduced root competition for water and nutrients may occur in larger gaps, although roots from neighboring shrubs commonly penetrate a meter into gaps, affecting spatial patterning of herbs within gaps. The extent to which gap effects are caused by belowground interactions is not known. Martina Petrů is examining the effects of experimental aboveground vs. belowground gaps on seedling recruitment, in rosemary scrub and scrubby flatwoods, as part of her Master’s research. Allelopathy from the roots of Florida rosemary inhibits germination of gap specialists, as discovered in research conducted by intern Molly Hunter and Menges, and differential germination success could also explain the effects of gap size and location within gaps on spatial patterning in some scrub plants (see also; Student Research, page 37).

Roadsides, abandoned firelanes, and other human-disturbed areas may perform much of the function of natural gaps, but their true role in demography of scrub plants is unknown. Research on H. cumulicola by Quintana-Ascencio and Weekley have found that demographic variation created higher risks of extinction and higher yearly population variation in roads than in Florida rosemary scrub.

Gapology is our attempt to integrate spatially-explicit data on gaps with the demography of many scrub plants. Beginning in 2000, we are locating and measuring thousands of gaps using GPS and ground measurements. In about 5 years, we will re-census these gaps, and derive estimates of gap demography such as "death" of small gaps, shrinkage in gap size, and coalescence and creation of gaps after fires. Many of our demographic projects are utilizing this common gapology framework.