Fire-adapted pineland
vegetation of Florida:
A framework for
inventory, management and restoration.
Introduction
Fire-dependent pineland vegetation once
dominated the uplands of the Southeastern Coastal Plain from southern Virginia
south to the tip of Florida and west to eastern Texas. The natural communities that occupied this
landscape, as well as many of the plants and animals that inhabited them, are
now rapidly disappearing. Economic
development has removed the natural vegetation from much of this region,
particularly from the finer-textured soils that readily support
agriculture. In addition, fire
suppression has transformed most of the remaining natural vegetation to the
point that it no longer resembles the vegetation present when European settlers
first arrived. The consequence is that
we are on the verge of irrevocably losing much of the biotic diversity of the
Southeastern United States. Many of the
communities that were maintained by fire, and the species that occupied them,
will be permanently lost if action is not taken almost immediately. The situation is particularly critical in
Florida, which contains the greatest diversity of natural, fire-maintained
pinelands, as well as the greatest diversity of organisms dependent on these
systems.
Tropical rainforests are often claimed
to be the world's most species-rich plant communities. However, richness can be measured at many
scales. Whereas tropical rainforests are
the most species-rich plant communities at scales of 100 m2 and
larger, at scales between 0.001 and 10.0 m2, the most species-rich
communities include the pine-dominated, fire-maintained temperate grasslands and open woodlands of
the Southeast (Peet et al 1990, Peet & Allard 1995; Figure 1). Values of 25-30 species per 0.25 m2,
and 40 per m2 are not uncommon in these ecosystems. Moreover, at scales of 100 and 1000 m2,
the highest values currently known from the temperate zone are from
Southeastern longleaf pine woodlands (95 / 100 m2; 150 / 1000 m2). Although not all fire-maintained pinelands
of the Southeast support high levels of species packing, the exceptionally high
species-packing that characterizes much of this vegetation clearly marks these
communities as important in efforts to maintain biodiversity. In the absence of
fire, the high species-packing, as well as the bulk of the unusual biota, is
quickly lost as woody plants grow up and crowd out the species-rich herbaceous
layer and the animals that depend on it.
Preservation of the fire-dependent biota
of the Southeast will not be easy.
Detailed documentation and analysis of the range of variation is needed
so that major portions are not overlooked in conservation planning. Moreover, simply reintroducing fire into a
landscape transformed by fire suppression is not sufficient for restoring
fire-maintained systems; in many cases the original species are no longer even
present. A multi-step research program
will be needed to develop management and restoration methods, and not all the
steps can be anticipated in advance.
However, one of the first steps must be to create a site-based template;
if we are to manage, restore or recreate the original pineland vegetation, we
need to know the range of variation in the original vegetation and we need to
know how that vegetational variation mapped onto local site conditions. Clearly, the success of future efforts to
manage or restore conservation lands representing fire-maintained pinelands in
Florida and throughout the Southeast will hinge on the availability of
comprehensive documentation of the original composition and structure to serve
as the appropriate template.
Given the distinctiveness, biotic
diversity, and original vast extent of fire-maintained pinelands in Florida,
there is shockingly little detailed, consistent or comprehensive information
available on the compositional variation of this vegetation. Available
descriptions either use classes so broad as to be nearly meaningless (e.g.,
flatwoods, high pine, savanna) or concentrate on only a handful of plots from a
single locality. Much of the best data is buried in unpublished
dissertations (e.g., Edminston 1963)
and government reports (e.g., Clewell 1971, Wolfe et al. 1988), and these data
have been collected in such varied ways as to preclude meaningful comparisons
and compilations. When Christensen
(1998) reviewed the regional literature, he concluded that classification of
flatwood vegetation was intractable, given available information. Myers (1990), Abrahamson & Hartnett
(1990), Stout & Marion (1993), Ware et al. (1993) and Peet & Allard
(1993) all have recently attempted to review the literature on these vegetation
types for Florida with markedly little success in providing a comprehensive and
detailed overview. I was particularly
struck when I attempted to compile original data from across the range of
longleaf pine for a preliminary region-wide classification, that effectively no
quantitative published data are available for Florida (see Peet & Allard
1993). The research I am proposing is
designed to provide for the first time a quantitative, comprehensive
description of the composition and structure of this fire-maintained pineland
vegetation.
Vegetation description and
classification play a key role in many areas of conservation, land management
and scientific research. In effect, the
classification identifies the relevant natural communities. When a scientist places a specimen in a
museum, he or she almost always attaches a label that contains critical
information about the origin of the specimen.
To conduct biological or ecological field research without providing
precise information on the ecological community or communities where the work
was done is much like omitting the label from a museum specimen; critical
information on ecological context or habitat will be missing. Inventories of natural areas can be no
better than the classification used to define the inventory units. Managers of conservation lands require
accurate and detailed descriptions of the vegetation attributes they need to
preserve or recreate, but without a well-formulated vegetation classification
and description, a quality template for management or restoration is not
possible. Widespread realization of these
needs has led to broad collaboration among federal agencies, conservation
organizations and professional ecologists to develop standards for vegetation
classification (see Federal Geographic Data Committee 1997, Weakley 1997). The work I propose will provide the
ecological context and necessary foundation for future research and management
on Florida's fire-dependent pineland ecosystems.
The focus of the proposed work is
explicitly fire-dependent, pine-dominated vegetation and is intended to include
primarily open woodland communities with herbaceous or low-shrub dominated
understories that burn regularly with low-intensity ground fires. In the somewhat ambiguous traditional
terminology, these communities would include the areas of flatwoods, high pine,
piney woods, savannas, some prairies, and some scrubby flatwoods. By far the most conspicuous plants are the
open-grown pines (longleaf, slash, and pond), though in some cases they are
extremely sparse. Scrub areas, such as
those dominated by sand pine, will not be included as those areas are not so
much fire-maintained as they are periodically fire disrupted. The subtropical rocklands and flatwoods of
the southern tip of the Florida (south of the terminus of the Lake Wales Ridge;
see Synder et al. 1990) dominated by South Florida slash pine (Pinus elliottii var. densa) are beyond the scope of this
study and will be the focus of a separate study using similar methodology
conducted by Dr. William Platt and others.
(Inclusion of these southern areas in the present project would be
desirable, but is not possible within the existing timeframe and budget
constraints.)
Research
context and long-term objectives
There is an urgent need to document the
composition and structure of the fire-maintained pineland vegetation of the Southeast
before these communities are lost to development and fire-suppression. If this is not done soon, the opportunity
will be lost forever. This urgency has
motivated me to undertake a series of studies to document the diversity of
vegetation composition and structure throughout this little-known biome. Thus far, I have completed the survey
(roughly 425 samples and 600 species) and analysis for the Carolinas and much
of Georgia. Three manuscripts are
nearing completion and should soon be submitted for publication. The editors of both Castanea and American Midland
Naturalist have actively encouraged submission of these papers to their
journals.
I view journal publication of regional
papers as merely stepping stones toward a comprehensive and synthetic
monographic book on the ecology of the fire-maintained pinelands of the
Southeast. A prospectus and timetable
for this book have been completed in collaboration with Dr. William Platt, and
we are currently in the process of soliciting potential contributing authors
and contacting potential publishers.
The prospectus is included as Appendix 1 (note that in the prospectus we
there refer to fire-maintained pinelands generically as "savannas", a
usage of the term much broader than that traditionally employed by biologists
in the Southeast.) The present proposal
is designed to provide funding for the comprehensive inclusion of Florida
ecosystems in our synthetic treatment.
Without this funding, much of the diversity of Florida pinelands would,
of necessity, be omitted or receive only inappropriately cursory treatment.
Objectives
& benefits
Despite the broad focus of my overall
research program, the focus of the proposed research will be exclusively on the
identification and description of the natural diversity of the fire-maintained
pinelands of Florida. This can be
viewed in terms of the series of specific objectives and benefits listed below.
1.
I will identify the best remaining examples of fire-maintained pineland
vegetation across the range of sites conditions and geographic regions of
Florida (excepting only the southern tip).
2.
I will collect detailed quantitative samples representative of the range
of fire-maintained pineland vegetation in the state using a standard sampling
protocol so as to assure maximal
compatibility with other datasets.
3.
I will analyze the resultant data to develop a comprehensive
classification and associated summary information for each type with respect to
species composition, species diversity, woody plant population structure, and
environmental and geographic situation.
4.
I will develop or expand community nomenclature in a fashion consistent
with the developing National Vegetation Classification (Federal Geographic Data
Committee 1997) so as to assure maximal utility to and application by
government agencies and conservation organizations. (Our work in the Carolinas and Georgia have identified nearly 70
community types, roughly half of which were new to and have subsequently been
added to the National Classification; see Weakley 1997. For illustrative purposes, the community
types recognized in that work are listed in Appendix 2 using the nomenclature
of the National Classification.)
5.
I will publish the results of this research in peer-reviewed journal
articles. These publications would
ultimately provide the basis for the synthetic overview monograph described in
Appendix 1.
Note that the product is a
classification and description based on the best remaining sites we can
identify, and is not an inventory of all the sites in Florida. Classification is a necessary step prior to
inventory; once the difficult step of classification has been completed,
inventory can be conducted using more extensive and less intensive sampling
than we propose for this project.
Proposed
research (Methods)
Identification
of sample locations
The most difficult component of this
project will likely be identification of the best remaining examples of fire-maintained
pinelands across the range of available site conditions. I am not so presumptuous as to think that
our group alone can locate and sample all variation in the fire-maintained
pineland vegetation of Florida within what is effectively a two-year
study. The only way to approach
complete coverage is through extensive consultation with regional experts. Representatives of state and private
conservation organizations, employees of federal agencies like the Forest
Service and the military, private and corporate land managers, consulting
foresters, and local academics all will be contacted and can be expected to
contribute in substantive ways. For
many years I have been building a network of contacts for such purposes. We will not be shy about contacting others;
we know that this project cannot be done well if done in isolation.
One important mechanism to assure
relatively complete coverage is to stratify sites by critical environmental
factors. For example, we expect to
first stratify the state by physiographic units, probably using those defined
by Cooke (1939), which, although not the most recent (see Randazoo and Jones
1997), have a much finer resolution than most other schemes. We will attempt to cover the range of
relevant soil series in each region.
Large blocks of land such as managed by the U.S. Forest Service, state
agencies, the military, and the forest products industry will receive
special. Aerial photos will be
consulted to identify large blocks of undeveloped and fire-maintained
vegetation. We anticipate that we will
continue to enjoy success at obtaining permission to work on large private land
holdings, industrial lands and military lands.
In summary, I am confident that we can
make substantial progress toward identification of the best remaining sites,
and in the process provide the essential skeleton for an evolving
classification and description, plus flesh out the more important pieces. With this critical core in place, professional
and amateur botanists and ecologists should be able to identify what has been
missed and contribute substantially to the refinement of the work over the
subsequent years.
In addition to identifying candidate sites, recognition of what
constitutes the best remaining vegetation is especially challenging. We will use a strategy similar to that which
we have used successfully in the Carolinas and Georgia. An initial literature review (starting with
Peet and Allard 1993, Abrahamson & Hartnett 1990, Myers 1990) will identify the range of variation that
might be expected. Historical
literature and especially photographs can be invaluable (e.g., Harper 1914,
Schultz 1907, Kurz 1942). We will then consult widely as described above. Our broad experience with fire-maintained
pinelands provides an invaluable experience base. We will examine patterns in species diversity, the presence of
weedy species, vegetation texture, presence of indicator species like
wiregrass, presence of old trees, and other attributes. Integrity of the ground
layer will be emphasized over structure of the tree canopy. The bulk of the biodiversity resides in the
groundlayer, this portion of the vegetation is most sensitive to site
conditions, and it the most difficult to restore following disturbance. Moreover, most pine canopies have been
modified by cutting, planting and other activities.
A few examples may serve to illustrate
some of the criteria that are useful.
When we sampled the flatwoods on the east side of the Okefenokee
Wildlife Refuge, we found that they looked altogether natural owing the recent
burning, but that small-scale diversity was substantially lower than found on
similar soils elsewhere. In the end, we
deduced that these systems were substantially degraded from years of fire
suppression, and that recent control burns have not yet fully restored the
system. Experimental plots at Tall
Timbers Fire Ecology Lab have been burned annually (and at other intervals) for
decades, but contain no trace of wiregrass and are clearly an artifact of land
use prior to the current management practices.
The oldgrowth longleaf stands of the Patterson Natural Area at Eglin Air
Force Base have clearly been in place for a a very long time, but they lack any
sign of wiregrass, despite the species occurring immediately to the north on
Blackwater and Conecuh forests. Here we
must join with other puzzled researchers and admit that we simply cannot tell
whether wiregrass was once part of that ecosystem. On the other hand, we were informed by ecologists at the Savannah
River Ecology laboratory that wiregrass was not an important component of the
natural vegetation of the Savannah River Site.
In this case, our4 extensive investigation provided evidence that
wiregrass had at one time been a local dominant, although today it is rare.
Vegetation
sampling
We will use the Carolina Vegetation
Survey (CVS) vegetation sampling protocol, a methodology designed to
simultaneously facilitate long-term studies of vegetation, assessment of biodiversity,
and inventory of contemporary vegetation composition and structure (Peet et al.
1998). Variations on this protocol have
been used to obtain in excess of 2800 vegetation samples in the Carolinas and
Georgia, including 427 that represent the range of fire-maintained pinelands in
those states.
The CVS methodology is designed to be
flexible so as to accommodate as many sampling situations as possible, to
effectively sample species composition over a broad array of spatial scales, and
to produce results compatible with the majority of commonly used sampling
schemes. The methodology is built
around 10x10 meter plots as the basic sampling module. In an intensively sampled 10x10m module, two
or more sets of nested quadrats are established, each radiating from a
permanently marked corner and increasing in area with each quadrat by a factor
of ten. All woody stems above breast
height in the module are recorded by diameter class. Typically 10 modules are
combined to form a standard 0.1 ha (1000 m2) plot, though local
heterogeneity in vegetation sometimes requires plots as small as a single
module. In a standard 0.1 ha plot,
there will typically be four central modules sampled intensively, and an
additional 6 (to make an array of 2 x 5) sampled for woody stems and additional
herbaceous species not encountered in the four intensive models. The method has considerable built-in
flexibility to allow for such contingencies as sampling of species with either
very high or very low densities of stems at breast height. Analysis can be conducted using samples
scales ranging from 0.01 m2 to 1000 m2. A full elaboration of the methodology is
beyond the scope of this proposal, but a detailed description is available upon
request or by anonymous ftp
(panicum.bio.unc.edu/NC_Veg_Survey/sampling/castanea.rtf); our suite of data
management and analysis software written in SAS is also available from this
site).
Soil samples will be collected from the upper 10 cm of the A horizon as
well as the B horizon of each intensively sampled module. Soils will be dried,
sieved, bagged, composited across the 4 modules, and sent Brookside Labs in New
Knoxville, Ohio for chemical and
textural analysis. The compositing of samples across modules is required to
minimize costs. Our previous work in similar systems in the Carolinas and
Georgia has shown that, unlike soils derived from igneous rock, soils of
coastal plain pinelands are spatially rather consistent in soil chemistry and
texture. This work has also shown strong
correlations between soil attributes and vegetation composition, making the
soil analysis a critical part of the proposed research.
Sampling efforts will be stratified to
cover a broad range of geographic and edaphic conditions. The actual data collection is time
consuming, in no small part because of the exceptional species richness of
these systems. Past experience suggests
that we can anticipate collection of between 100 and 140 samples per season for
a total of somewhere around 250 detailed samples for the two field
seasons. Our final dataset will be
available to state agencies and conservation professionals.
Data
management and analysis
We have developed a suite of SAS
computer programs for error checking CVS data and for data summarization. Once the data are entered and error checked,
species nomenclature will be revised to conform to Kartesz 1994 to provide
consistency across all our Southeastern pineland datasets.
Numerical classification programs will
be used to develop an initial classification of vegetation samples for each
major geographic region. Past
experience suggests that agglomerative classification using a Bray-Curtis
similarity metric and either the Wards or Lance-Williams algorithm produces the
most ecologically interpretable results, an observation confirmed by Belbin and
McDonald (1993). Two-way indicator
species analysis (Hill 1979) may be used for exploratory analysis, but we agree
with Belbin and McDonald that this method can produce misleading results.
We will use ordination techniques to
relate compositional patterns to environmental and site variation, mostly
though a combination of multidimensional scaling (Minchin 1987, 1990),
detrended correspondence analysis (ter Braak 1987, Peet et al. 1988) and
canonical correspondence analysis (ter Braak 1987). Once the primary vegetation gradients have been recognized, the
vegetation pattern and distribution of community types will be represented in
terms of these gradients (as in Peet & Loucks 1977, Austin et al. 1990,
Allen et al. 1991). To facilitate the
untangling of complex, interacting environmental factors, we will use a
strategy of progressive fragmentation (Peet 1980, Peet and Allard 1995) where the
dataset is broken into progressively smaller portions as critical environmental
factors are identified. The nonlinear
interaction of environmental gradients makes this approach better than one of
simply partialling out the variance attributable to particular factors (as in
Carleton 1984, ter Braak 1987, Wiser et al. 1996). Multiple discriminent analysis will be used to develop site-based
templates for community management and restoration (see Van Lear & Jones
1987, Allen & Wilson 1991).
The classification will be reviewed in
light of the gradient analysis results and refined if necessary. The resultant classification units will then
be compared with the existing National Vegetation Classification (Weakley et al
1997) being developed by the Nature Conservancy and the U.S Federal Geographic
Data Committee (1997). Nomenclature
from the National Classification will be used where appropriate, and new units
will be proposed for the National Classification where needed. Previous work in the Carolinas and Georgia
suggests somewhat over half of the community types recognized will be new to
the National Classification (see Appendix 2 and Weakley et al 1997). The composition and environmental setting of
the communities recognized will be summarized in terms of composition,
structure, diversity and environmental setting,
Project
schedule
The project
will be divided geographically and temporally into two phases. The first phase will focus on the pinelands
of the panhandle plus those of the northeast corner of the Florida. The second phase will focus on the bulk of
peninsular Florida The field work for
the first phase will be conducted during May-October 1999, and the field work
for the second phase will be conducted during May-October 2000. Each field work period will be preceded by a
reconnaissance period and each will be followed by a period of data analysis
and manuscript preparation. Each of the
two phases can be expected to produce three deliverable products: a list of
sites selected, a complete dataset, and a manuscript for publication. A comprehensive report will be provided at
the end which will contain the first approximation of a synthetic treatment of
fire-maintained pinelands of Florida and adjacent states.
The specifics
of the schedule can be summarized as follows:
The specifics
of the schedule can be summarized as follows:
(Revised
September 2000)
(i =
initiate work; w = work in progress; c = work completion; p =
product receivable)
Fiscal
Year 1999-00 2000-01
2001-02 2002-03 2003
Quarter 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2
Phase 1 –
Panhandle Florida
Study site
& selection i w w w . w c . . . . . . . .
Vegetation
sampling . i w w . . w c . . . . . . . .
Data
compilation and management . . . i w w . . w c . . . . . .
Data analysis . . . . . , . . w w c . . . . .
Subsection
report preparation . . . . . . . . i w w w c . .
Phase 2 –
Peninsular Florida
Study site
identification and selection . . . . i w w . w w c . . . .
Vegetation
sampling . . . . . i w w . . w c . . . .
Data
compilation and management . . . . . . . i w w . w c . . .
Data analysis . . . . . . . . . . i w w c . .
Subsection
report preparation . . . . . . . . . . . i w w c .
Final report
preparation . . . . . . . . . . . . i w w p
Literature
cited
Abrahamson,
W.G., and D.C. Hartnett. 1990. Pine
flatwoods and dry prairies. Pages
103-149 in Ecosystems of Florida.
R.L. Myers and J.J. Ewel, editors. The University of Central Florida Press,
Orlando.
Allen, R.B.,
R.K. Peet, and W.L. Baker. 1991.
Gradient analysis of latitudinal variation in Southern Rocky Mountain
forests. Journal of Biogeography
18:123-139.
Allen, R.B.
and J.B. Wilson. 1991. A method for determining
indigenous vegetation from simple environmental factors, and its use for
vegetation restoration. Biological Conservation 56:265-280.
Austin, M.P.,
A.O. Nicholls, and C.R. Margules. 1990.
Measurement of the realized niche; environmental niches of five Eucalyptus species. Ecological
Monographs 60:161-177.
Belbin,
L. and C. McDonald. 1993. Comparing
three classification strategies for uswe in ecology. Journal of Vegetation
Science 4:341-348.
Carleton, T.J.
1984. Residual ordination analysis: a
method for exploring vegetation-environment relatioinships. Ecology
65:469-477.
Christensen,
N.L. 1998. Vegetation of the
southeastern coastal plain. In M.G. Barbour and W.D.
Billings, eds. North American Terrestrial Vegetation, Second edition.. Cambridge University Press, Cambridge (in press).
Clewell, A.F.
1971. The vegetation of the Apalachicola National Forest: an ecological
perspective. Contract 38-2249,
Final Report. U.S. Forest Service, Atlanta, Georgia.
Cooke, C.W.
1939. Scenary of Florida, as interpreted by a geologist. Florida
Geological Survey Bulletin 17. 118 p.
Edminston,
J.E. 1963. The ecology of Florida pine flatwoods. PhD Thesis, Univ. Florida, Gainesville, FL.
Federal
Geographic Data Committee Secretariat.
1997. FGDC Vegetation classification
and information standards (http://www.nbs.gov/fgdc.veg/).
Harper, R.M.
1914. Geography and vegetation of North
Florida. Florida State Geological
Survey. Sixth Annual Report 163-437.
Hill, M.O.
1979. TWINSPAN - A FORTRAN Program for arranging multivariate data in an
ordered two-way table by classification of the individuals and attributes. Cornell University. Ithaca, NY.
Kartesz, J.T.
1994. Synonymized checklist of the vascular flora of the United States,
Canada, and Greenland. Second
edition. Timber Press, Portland, OR
Kurz, H.
1942. Florida dune and scrub vegetation and geology. Florida Geological
Survey. Geological Bulletin 23. 154 p.
Minchin,
P.R. 1987. An evaluation of the relative robustness of techniques for ecological
ordination. Vegetatio 69:89-107.
Minchin,
P.R. 1990. DECODA Version
2.01. Anutech Pty. Lts., Canberra.
vegetation of north-central Florida. American Midland Naturalist 79:441-457.
Myers, R.L.
1990. Scrub and high pine. Pages 150-193 in R.L. Myers and J.J. Ewel
(editors). Ecosystems of Florida.
Univ. Central Florida Press, Orlando, FL.
Peet, R.K.
1980. Ordination as a tool for
analyzing complex data sets. Vegetatio 42:171-174.
Peet, R.K. and
D.J. Allard. 1993. Longleaf pine vegetation
of the Southern Atlantic and Eastern Gulf Coast regions: A preliminary
classification. Proc. Tall Timbers Fire Ecology Conf. 18: 45-81.
Peet, R.K.,
R.G. Knox, R.B. Allen, and J.S. Case. 1988.
Putting things in order: the advantages of detrended correspondence
analysis. American Naturalist 131:924-934.
Peet, R.K. and
O.L. Loucks. 1977. A gradient analysis
of Southern Wisconsin forests. Ecology 58:485-499.
Peet, R.K., E.
van der Maarel, E. Rosen, J., Willems, C. Norquist, and J. Walker. 1990. Mechanisms of coexistence in species-rich
grassland. Bulletin of the Ecological Society of America 71:283.
Peet, R.K.,
T.R. Wentworth and P.S. White. 1998. A
flexible, multipurpose method of recording vegetation composition and
structure. Castanea (in press). (available via ftp as panicum.bio.unc.edu/NC_Veg_Survey/Sampling/castanea.rtf
Randazzo, A.F.
and D.S. Jones (eds.). 1997. The geology of Florida. U. Presses of
Florida, Gainesville, FL. 327 pp.
Schwarz, G.F.
1907. The longleaf pine in virgin forest: a silvical study. J. Wiley
& sons, NY. 135 p.
Snyder, J.R.,.
A. Herndon and W.B. Robertson Jr. 1990.
South Florida Rockland. Pages
150-193 in R.L. Myers and J.J. Ewel (editors). Ecosystems of Florida. Univ. Central Florida Press, Orlando, FL.
Stout, I.J.
anad W.R. Marion. 1993. Pine flatwoods
and xeric pine forests of the Southern (lower) coastal plain. Pages 373-446 in W.H. Martin, S.G. Boyce and A.C. Echternacht, eds. Biodiversity
of the Southeastern United States: Lowland terrestrial communities. John Wiley and Sons, New York, NY.
ter Braak,
C.J.F. 1987. The analysis of
vegetation-environment relationships by canonical correspondence analysis. Vegetatio
69:69-77.
Van Lear, D.H.
and S.M. Jones. 1987. An example of
site classification in the Southeastern Coastal Plain based on vegetation and
land type. Southern Journal of Applied Forestry 11:23-28.
Ware, S., C.
Frost and P.D. Doerr. 1993. Southern
mixed hardwood forest: the former longleaf pine forest. Pages 447-493 in W.H. Martin, S.G. Boyce and A.C. Echternacht, eds. Biodiversity
of the Southeastern United States: Lowland terrestrial communities. John Wiley and Sons, New York, NY.
Weakley, A.S.,
K.D. Patternson, S. Landaal and Mark Gallyoun. 1997. International classification of ecological communities:
terrestrial vegetation of the Southeastern United States (Draft of November
1997). The Nature Conservancy,
Southeast Regional Office, Chapel Hill, NC
Wiser, S.K.,
R.K. Peet and P.S. White. 1996.
High-elevation rock outcrop vegetation of the Southern Appalachian
Mountains. J. Vegetation Science 7 (in press).
Wolfe, S.H.,
J.A. Reidenauer and D.B. Means. 1988. An ecological characterization of the
Florida panhandle. U.S. Department of Interior, Fish and Wildlife Service,
FWS Biological Report 88(12).
Wharton, C.H. 1978. The natural environments of Georgia. Georgia Dept. Nat. Res. Atlanta, GA.