Final

Plant Community Parameter Estimates and Documentation for the Across Trophic Level System Simulation (ATLSS)

Data Report Prepared for the

ATLSS Project Team

The Institute for Environmental Modeling

University of Tennessee–Knoxville

Louis J. Gross, Director

Prepared By

Paul R. Wetzel

Department of Biological Sciences

East Tennessee State University

Johnson City, TN 37614–0703

October 18, 2001

Table of Contents

List of Tables 4

List of Figures 4

Document Objectives 5

Limitations and Assumptions Used to Estimate

Botanical Parameters and Develop Succession Models 5

Hydroperiod Parameter Determination 7

Succession Models 16

Introduction 16

Integrating the Succession Models 16

Pine/Scrub/Flatwood Succession 18

How to Read the Pine/Scrub/Flatwood Succession Diagram 18

Plant Classes Included in the Pine/Scrub/Flatwood Succession Table 19

References for the Pine/Scrub/Flatwood Succession Model 19

Cypress Forest Succession 21

How to Read the Cypress Forest Succession Diagram 21

Plant Classes Included in the Cypress Succession Model 22

References for the Cypress Forest Succession Model 22

Herbaceous Plant Communities Succession 24

How to Read the Herbaceous Plant Communities Succession Diagram 24

Plant Classes Included in Herbaceous Succession Model 26

References for the Herbaceous Succession Model 26

Coastal Community Succession 29

Introduction 29

How to Read the High Energy Coastal Communities–Eastern Coast Succession Diagram 32

Plant Classes Included in High Energy Coastal Communities–Eastern Coast Succession Model 32

Plant Classes Included in Low Energy Coastal Communities–Western and Southern Coasts Succession Model 33

References for the Coastal Communities 33

Table of Contents continued

Mangrove Forest Succession 34

Estimation of Deer Browse Parameters 36

Introduction 36

Deer Forage Estimation for Each Plant Community 36

Deer Forage Growth Rate Estimation 38

Water Depth Parameter Estimation 44

Limitations of the Water Depth Parameter Estimates 45

Future Additions or Changes to the ATLSS Model 50

Literature Cited 51

List of Tables

Table 1. Hydroperiod data and estimates for all 48 plant communities used in the ATLSS model and the aerial coverage of each plant community. CG=Compositional Group, EC=Ecological Complex. Full reference citations are given in the Literature Cited section. 7

Table 2. Successional relationships for the pine/scrub/flatwood plant communities. 20

Table 3. Successional relationships for the cypress plant communities south and north of the southern edge of Lake Okechobee. 23

Table 4. Successional relationships for the herbaceous and forested plant communities. 27

Table 5. Additional successional relationships that occur in the central Everglades not represented on Table 4 (after White 1994, p. 453). Note that only a reduction in hydrology would cause succession of the plant communities to occur in the opposite direction indicated. 28

Table 6. A. Successional relationships for the high energy coastal plant communities. Information on the effects of fire disturbance was available only for the communities found on old dune ridges. B. Plant communities found on low/moderate energy shorelines. Adequate information was not available to develop a succession model for these communities. 31

Table 7. Deer forage data and estimates for the plant communities used in ATLSS. Biomass, productivity data, and parameter estimates are also listed. 37

Table 8. Biomass and growth rate data and estimates for ATLSS plant communities. These values were used to estimate the growth rates of the deer browse component of the plant community. CG=Compositional Group, EC=Ecological Complex. Full reference citations are given in the Literature Cited section. 39

Table 9. Water depth data and estimates for ATLSS plant communities. These values were used to relate the productivity of a plant community with hydrology. CG=Compositional Group, EC=Ecological Complex. Full reference citations are given in the Literature Cited section. 46

List of Figures

Figure 1. Relationship of deer forage to hydrology in the ATLSS model. Modeling this relationship requires the estimation of six parameters for each plant community: the rate of growth of deer forage, the minimum and maximum water depths at which the plant community grows, the minimum and maximum optimal growth depths, and the rate that the forage growth declines (loss rate) after the maximum growth depth. 44

Figure 2. Schematic of how water depth parameters were estimated for high and moderate deer forage growth rates. 45

Document Objectives

This document describes the botanical parameters and the methods used to estimate those parameters needed to run the Across Trophic Level Systems Simulation (ATLSS). It also describes a series of simple successional models that incorporate hydrologic and fire disturbances into ATLSS. The ATLSS covers the Florida peninsula from Lake Okechobee southward. It uses plant communities defined by the Florida GAP (FGAP) analysis (version 6.6) as its basic ecosystem units. The objectives of this document are listed below.

1. Describe the limitations and assumptions used to estimate the botanical parameters and develop the succession models.

2. Determine hydroperiod ranges for all of the FGAP v. 6.6 plant communities in south Florida.

3. Estimate the amount of deer browse available in the plant communities where deer are expected to live.

4. Estimate the maximum and minimum water depths that vegetation grow in each plant community where deer forage.

5. Develop a simple set of succession models that incorporate hydrologic and fire disturbances. The succession models should include the direction and rate of succession for both disturbances.

6. Carefully document all parameter estimation and succession models with references from the scientific literature and expert professional opinion.

Limitations and Assumptions Used to Estimate Botanical Parameters and Develop Succession Models

It is important that the users of the data contained in this document understand how the information for the model parameters and succession sequences was gathered and synthesized. This is necessary to prevent them from making conclusions with the ATLS Simulation that go beyond the reliability of the input data. Gleaning data from a wide variety of sources in the scientific literature and the lack of data on certain plant communities limits the strength of the data as input to the ATLS Simulation. Use of the FGAP plant classification system also created certain limitations and assumptions. These limitations and assumptions are described below and should be read and carefully considered by all users of the data contained in this report.

1. Differences in Plant Community Classification

Plant community classification can vary significantly among different systems. For example between the different versions, v. 2.1, v. 3.0, and v. 6.6, of FGAP, between FGAP v. 6.6 and plant communities described in the literature, and between FGAP v. 6.6 and Harlow (1959), a reference used extensively in estimating deer browse. Many of the FGAP class descriptions are very vague and give few representative plant species. Discussions of plant communities in the literature are usually quite the opposite: detailed community descriptions often with species lists. Much time was spent matching similar plant communities between classification systems.

It is my impression that the plant associations created in the FGAP analysis mapping effort were based on the aerial signatures of plant communities that could be readily identified from aerial photographs. This results in the establishment of some plant associations that have wide hydroperiods or plant associations that do not correspond with the plant communities reported in the literature by botanists and ecologists working on the ground. Because some of the FGAP plant classifications do not match well with the plant communities described in the literature, the succession and hydroperiod data going into the model is either very broad or not very specific.

Incidentally, an alternative mapping system would be to establish a vegetation classification system, determine the air photo signature of each plant community in the classification system, and then proceed with the vegetation mapping. This procedure was followed by Madden et al. (1999) and I think that their vegetation classification system will be somewhat easier to adapt to the needs of the ATLSS model.

2. The Difference between Hydroperiod and Hydrologic Regime

Hydroperiod (the average number of days per year that the water level is at or above the soil surface) was estimated for the plant communities used in the ATLSS model. However, it is very important to note that wetland ecosystems have characteristic hydrologic regimes. A hydrologic regime has two components: the hydroperiod and a hydro pattern, that is, the seasonal occurrence of inundation and draw downs. When water is present in a wetland community is as important to the plants and animals as the length of inundation. Hydrologic regimes fluctuate seasonally, annually, and inter-annually. This report does not include information about the hydro patterns of specific plant communities.

3. Net Primary Production Parameter Estimates

To estimate the growth rates (kg/ha/month) of the portion of a plant community that could be used as deer browse, it was necessary to collect biomass estimates of each representative plant community. There are many methods of determining net primary production. However, the reported biomass values were not adjusted or calibrated with each other in any way.

4. Water Depth Parameter Estimates

In order to connect the biomass growth of deer forage to hydrology, several water depths were estimated for each plant community group. These water depths are tied to the minimum, optimal, and maximum growth rates of the plant communities in the ATLSS model. The water depths assigned to a particular plant community were obtained from the literature but are not related to the growth rates used in the model. Growth rate and water depth data were not collected at the same time, during the same season of the year, or in the same location. The development of the water depth parameters and their relationship to growth rates is purely a construct of the model.

5. Succession Models

The simple succession models designed for inclusion in the ATLSS model were developed with a number of assumptions. First, it was assumed that hydroperiod is the primary determinant of individual plant communities. The hydroperiods used in the succession models are reported in Table 1.

Second, it was assumed that plant communities succeeded for only two reasons: hydroperiod disturbance and fire disturbance. Therefore, each disturbance must have a time counter in the model that corresponds to the years since the last shift in average hydroperiod or years since last fire disturbance. Annual seasonal dry downs were not considered to be hydrologic disturbances. These annual changes in hydroperiod are a characteristic of the hydroperiod of the plant community. There was no data on the intensity of these disturbances and how the varying intensity of the disturbance affected the plant communities

Clearly, the plant communities in south Florida experience many other disturbances and many of them have been reported or studied in the scientific literature. Some of the disturbances described in the literature include:

· Anthropogenic disturbances on the landscape ranging from agriculture to urbanization

· Hurricanes

· Freezes

· Nutrient level

· Salinity gradient near the coasts and estuaries (also relates to sea level rise)

· Seed/propagule sources and dispersal (particularly with exotics).

Hydroperiod Parameter Determination

The ATLSS hydroperiod of a plant community is the average number of days per year that the water level is at or above the soil surface. Hydroperiod values ranged from 0 to 365 days and were determined from references in the scientific literature (Table 1). All hydroperiod references found for particular plant communities are listed and the hydroperiod range assigned to a plant community either encompassed the ranges reported in the literature or were averaged from the data available. The data collection period and the nature of the hydroperiod data reported were also factored into the final hydroperiod assignment.

Hydroperiod estimates were not found in the literature for three plant community types: Mixed Evergreen–Cold Deciduous Hardwood Forest [19], Coastal Strand [33], and Sea Oats Dune Grassland [40] (Table 1). The hydroperiods for these plant communities that are reported in Table 1 were estimated by Paul Wetzel using best judgment. Fortunately, these three plant communities represent only 0.021% of the project area (Table 1).

Other plant communities, such as Sand Pine Forest [14], Sand Cordgrass Grassland [48], or Casuarina Compositional Complex [12], have only one reference or indirect references to their hydroperiod. The indirect references are explained in Table 1 or in the notes on Table 1 whenever necessary. Hydroperiod references abound for well studied plant communities, such as Sawgrass Marsh and Cypress Forest (Table 1). All references found for any plant community are reported in Table 1 so that the reader may make their own judgments of the average hydroperiod for a particular plant community.

Table 1 (next page). Hydroperiod data and parameter estimates for all 48 plant communities used in the ATLSS model and the aerial coverage of each plant community. CG=Compositional Group, EC=Ecological Complex. Full reference citations are given in the Literature Cited section.

High Pine and Scrub	14	Sand Pine Forest	Pinus clausa, sand pine Dry sand ridges interior & coast	0.06	0	Implied 0 d–Myers, 1990, pp. 159, 176 “Excessively well drained”–Abrahamson and Hartnett 1990, p. 111, Fig. 5.2
[Community Aerial Coverage = 0.31%]	26	Sandhill EC	Longleaf pine, Pinus palustris, xeriphytic oaks, Q. incana, Q. geminata, Q. laevis, and a wiregrass/sporobolus understory on sand	0.03	0	Implied 0 d–Myers, 1990, pp. 159, 176 “Excessively well drained”–Abrahamson and Hartnett 1990, p. 111, Fig. 5.2
	35	Xeric Scrubland	Shrublands on inland sand and coastal dune ridges; Q. chapmanii, Q. geminata, Q. inopina, Q. myrtifolia, Ceratiola ericoides (FL Rosemary), and Lyonia ferruginiea. Scattered P. clausa, P. palustris, and P. elliottii.	0.22	0–15	Implied 0 d–Myers, 1990, pp. 159, 176 “Moderately well drained”–Abrahamson and Hartnett 1990, p. 111, Fig. 5.2
Mesic Temperate Hammock	4	Xeric–Mesic Live Oak EC	Xeric to mesic hydrologic conditions. Q. virginiana, Q. geminata	0.36	0–15	“Moderately well drained”–Abrahamson and Hartnett 1990, p. 111, Fig. 5.2
[Community Aerial Coverage = 0.96%]	5	Mesic–Hydric Live Oak/Sabal Palm EC	Mesic to hydric hydrologic conditions; hydric hammocks. Q. virginiana, S. palmetto	0.04	0–60	0d for 1yr measurmnt–Vince et al. 1989, p. 13, Fig. 10 60–120d–Vince et al. 1989, p. 14 0–30d–Drew and Schomer 1984, p. 99, Fig. 62
	6	Bay/Gum/ Cypress EC	Gordonia lasianthus, Magnolia virginiana, Persea palustris (bays), Nyssa spp. (gum), Taxodium spp.	0.55	60–160	60–160d–Schomer and Drew 1982, p. 110
	19	Mixed Evergreen–Cold Deciduous Hardwood Forest	Southern mesic hardwood forest or upland hardwood forest. East: Q. hemispherica, Q. virginiana and Carya glabra. West: Fagus grandifolia and Magnolia grandiflora	0.01	0–15	No data available. Estimated. 1–15d–M. Dennis, personal communication
Tropical Hardwood Hammock	2	Tropical Hardwood Hammock	Coastal and interior hardwood hammocks.	0.49	10–45	0–60d–Schomer and Drew 1982, p. 110 10–45d–Drew and Schomer 1984, p. 104
[Community Aerial Coverage = 0.84%]	20	Buttonwood Woodland	Buttonwood (Conocarpus erectus) woodland. Found inland and adjacent to the mangrove zone over marl soils or on exposed bedrock.	0.35	44–120	Mean 244d for river fringe–Kolipinski and Higer 1969, pp. 16a, 27 See Note 1.
Pine Flatwood and	13	South Florida Slash Pine Forest	S. FL pine forest, Pinus elliottii var. densa. Found on sand in the north and limestone in the south of FL.	0.95	0–60	30–60d–Abrahamson and Hartnett 1990, p. 109 0–60d–Schomer and Drew 1982, p. 110
Rockland	16	Mesic–Hydric Pine Forest CG	Multiple pine forest types. Dominated by Pinus elliottii var. elliottii	3.22	30–60	30–60d–Abrahamson and Hartnett 1990, p. 109 0–60d–Schomer and Drew 1982, p. 110
[Community Aerial Coverage = 6.64%]	25	South FL Slash Pine Woodland	Open, generally low stature south FL slash pine (P. elliottii var. densa) on sand, marl or rock. Understory usually graminoid w/ occassional Taxodium distichum	1.32	30–60	30–60d–Abrahamson and Hartnett 1990, p. 109 0–60d–Schomer and Drew 1982, p. 110 70–160d–Sun et al. 1995, wet edge of plant association
	30	Gallberry/Saw Palmetto CG	Shrub and graminoid communities in association with wet flatwoods. Gallberry (Ilex glabra and I. coriacea), fetterbush (Lyonia lucida), sweet pepperbush (Clethra alnifolia) and titi (Cyrilla racemosa)	0.98	30–60	30–60d–Abrahamson and Hartnett 1990, p. 109 0–60d–Schomer and Drew 1982, p. 110 ≤90d–Krauss 1987 70–160d–Sun et al. 1995, wet edge of plant association 30–90d–M. Dennis, personal communication
	36	St. Johns Wort Shrubland	Often found in isolated, small, acid wetlands. Hypericum fasciculatum may cover entire wetlands or the fringe of deeper water bodies.	0.17	30–150	Winchester et al. 1985. See Note 2.
Dry Prairie [Community Aerial Coverage = 1.08%]	29	Dry Prairie EC	Sparsely wooded savannas with mosaic of Serenoa repens and grasses Aristida spp., Sporobolus spp., and Andropogon spp.	0.91	30–60	30–60d–Abrahamson and Hartnett 1990, p. 109 ≤50d–Duever et al 1984a, p. 301 0–30d–M. Dennis, personal communication
Wet Prairie [Community Aerial Coverage = 4.75 %]	45	Muhly Grass Marsh	Marls soils and on dry coastal sands and shells. Muhlenbergia capillaris	2.51	60–120	<180d–Kushan, 1990, p. 337 60–210d–Schomer and Drew 1982, p. 110 100d–Porter 1967, p.938 111–155d–Duever et al. 1978, p. 537 70d mean–Gunderson and Loope 1982b ≥90d–Krauss 1987 but not more than 135d? 60–120d–Loope 1980
	52	Sparsely Wooded Wet Prairie CG	Graminoid or forb understory and sparse wooded overstory. Includes Taxodium distichum or Pinus spp.	0.01	60–120	60–210d–Schomer and Drew 1982, p. 110 70d–Duever et al 1984a, p. 301
	53	Dwarf Cypress Prairie	Graminoids (Muhlenbergia capillaris, Rhynchospora spp.) with sparse shrub overstory (Taxodium distichum)	1.62	120–150	120–210d–Schomer and Drew 1982, p. 110 150–210d–Ewel 1990, p. 298 120d–Brown et al. 1984, p. 308 120d– Duever et al 1984a, p. 301
	54	Temperate Wet Prairie	Located in northern and central FL. Graminoids, forbs, and hydrophyllic species.	0.61	166–290	166–290d–Goodrick 1974 146–237d–Hagenbuck, et al. 1974, p. 10b
Marsh [Community Aerial Coverage = 19.4 %]	42	Graminoid Emergent Marsh CG	Graminoid marshes.	4.75	120–270	224–278d–Duever et al. 1978, p. 537 180–270d–Kushan 1990, p. 337 84–365d–McPherson 1973, p. 18 146–237d–Hagenbuck, et al. 1974, p. 10b ≤250d– Duever et al 1984a, p. 302
	43	Sawgrass Marsh	Cladium jamaicense	13.27	130–330	180–270d–Kushan 1990, p. 337 150–300d–Schomer and Drew 1982, p. 117 365d–Steward and Ornes 1975 117–310d–David 1996, p. 22 168–303d–Lowe 1986, p. 220 175–365d–McPherson 1973, p. 18 Mean 285d–Kolipinski and Higer 1969, pp. 16a, 27 90–240d–Loope 1980 73–180d–Hagenbuck, et al. 1974, p. 10c 133–335d, mean=259–Ross et al. 2000, p. 107
	44	Spikerush Marsh	Eleocharis spp.	0.54	150–300	180–270d–Kushan 1990, p. 337 150–300d–Schomer and Drew 1982, p. 117 193–310d–David 1996, p. 22 Mean 327d–Kolipinski and Higer 1969, pp. 16a, 27 >270d–Loope 1980 73–180d–Hagenbuck, et al. 1974, p. 10c (Eleocharis was 15–35% frequency) 266–333d–Ross et al. 2000, p. 107
	46	Cattail Marsh CG	Typha domingensis and T. latifolia	0.53	180–280	180–270d–Kushan 1990, p. 337 ~180–300d–Schomer and Drew 1982, p. 110 208d–David 1996, p. 22
	55	Maidencane Marsh	Panicum hemitomon	0.31	180–300	180–270d–Kushan 1990, p. 337 180–300d–Schomer and Drew 1982, p. 117 222d–David 1996, p.22 270–350d–Lowe 1986, p. 218
Shrub Island [Community Aerial Coverage = 6.03 %]	28	Broad Leaved Evergreen/Mixed Evergreen Cold–Deciduous Shrubland CG	Fetterbush (Lyonia lucida) [North FL] and cocoplum (Chrysobalanus icaco) [South FL]. Freshwater red mangrove dwarf shrubland, Rhizophora mangle, C. icaco. Highest density on Gulf coast.	0.30	120–150	120–150d; min of 60d–Schomer and Drew 1982, p. 115
	32	Dwarf Mangrove EC	Shrub mangroves, regardless of species dominance.	1.23	150–300	116–360d, mean=300d–Ross et al. 2000, p. 107
	37	Saturated–Flooded Cold Deciduous Shrubland EC	Shrub wetlands. Salix spp., Cephalanthus occidentalis, Betula nigra, Alnus serrulata, and sometimes high proportions of Typha spp. or Cladium jamaicense	4.50	110–320	110–365d (tree islands)–Wetzel 2001 150–300d w/ willow–Schomer and Drew 1982, pp. 110,115 110–365d–McPherson 1973, p. 18 Mean 244d–Kolipinski and Higer 1969, pp. 16a, 27
Slough [Community Aerial Coverage = 0.96 %]	56	Forb Emergent Marsh	Pontederia cordata, Sagittaria lancifolia, and Thalia geniculata	0.96	230–360	240–360d–Schomer and Drew 1982, p. 117 ~222–350d–David 1996, p. 22 310–346d–Duever et al. 1978, p. 538
Pond [Community Aerial Coverage = 0.67 %]	57	Water Lily or Floating Leaved Vegetation	Eichhornia crassipes, Hydrocotyle spp., Nuphar luteum, Nymphaea odorata, and Nymphoides aquatica	0.67	330–360	330–360d–Gunderson and Loftus 1993, p. 205 220–350d–David 1996, p. 22 259–365d [“slough”]–McPherson 1973, p. 18 212–310d–Hagenbuck et al. 1974, p. 10a ~350d– Duever et al 1984a, p. 302
Bayhead [Community Aerial Coverage = 1.03 %]	3	Semi–Deciduous Tropical/ Subtropical Swamp Forest	Large strand swamps, low stature swamps or bayhead forest and tree islands. Taxodium distichum, Roystonea elata (royal palm), Quercus laurifolia, and Acer rubrum. In South FL, Annona glabra, Magnolia virginiana, and Persea palustris	1.03	60–180	120–150d–Schomer and Drew 1982, p. 115 60–180d–Gunderson and Loftus 1993, p. 213 Mean 244d–Kolipinski and Higer 1969, pp. 16a, 27 100–150d–Drew and Schomer 1984, p. 99, Fig. 62
Cypress & Mixed Swamp Forest/ Woodland	17	Swamp Forest CG	Deciduous and evergreen swamp forests of south and central FL. Some Taxodium spp. and the species of the Bay/Gum/Cypress EC (Class #6)	3.19	120–290	180–270d–Ewel 1990, p. 298 120–210d–Schomer and Drew 1982, p. 110 155–290d–Duever et al. 1978, p. 543
[Community Aerial Coverage = 6.78 %]	18	Cypress Forest CG	Cypress domes and river and lake fringes. May overlap with pines and cypress/gum ponds within pine flatwoods. Taxodium distichum, T. distichum	3.59	200–340	180–270d–Ewel 1990, p. 298 Nearly 365d–Sharitz and Mitsch 1993, p. 319 200–240d–Wharton et al. 1977 250–290d–Duever et al. 1984a, p 301 212–340d–Sun et al. 1995, p. 67, Msmts from pond. Range over 2yrs, one wet , one dry
Exotics	8	Cajeput Forest CG	Melaleuca quinquenervia	0.09	150–210	150–210d–Ewel 1990, p. 298 120–150d–Schomer and Drew 1982, p. 110
[Community Aerial Coverage = 0.34 %]	12	Casuarina Complex	Casuarina (equisetifolia, cunninghamia, glauca)	0.01	0–60	“Less than Muhly Prairie”–Schomer and Drew 1982, p. 14
	31	Brazilian Pepper Shrubland	Monotypic stands of Schinus terebinthifolius.	0.24	0–120	0–120d–Schomer and Drew 1982, p. 110
Mangrove [Community Aerial Coverage = 3.53 %]	9	Mixed Mangrove Forest Formation	Contains all three mangrove species (Laguncularia racemosa, Avicennia germinans, Rhizophora mangle) with varying levels of dominance. White and black species gernerally dominate.	2.61	60–240	208d–data from Jamacia for Laguncularia only Chapman 1976, p. 192 See Note 3.
	10	Black Mangrove Forest	Avicennia germinans Forest	0.16	120–240	255–355d–data from Jamacia Chapman 1976, p. 192 See Note 3.
	11	Red Mangrove Forest	Rhizophora mangle Forest	0.57	240–365	355–365d–data from Jamacia Chapman 1976, p. 192 See Note 3.
	21	Mixed Mangrove Woodland	Forest species the same as the mixed mangrove forest [9] but canopy coverage reduced to 26–60%. Reduced canopy from Hurricane Andrew.	0.12	60–240	208d–data from Jamacia for Laguncularia only Chapman 1976, p. 192 See Note 3.