A. GENERAL INFORMATION:

Project Title: Use of Amphibian Communities as Indicators of Restoration Success
Project start date: 2003 Project end date: 2005
Project Funding: USGS's Greater Everglades Science Initiative (PBS), CESI
Principal Investigator: Kenneth G. Rice
Email address: ken_g_rice@usgs.gov
Phone: 954-577-6305 Fax: 954-577-6347
Mail address: USGS, University of Florida Field Station, 3205 College Ave., Fort Lauderdale, FL 33314

Other Investigator(s): Frank J. Mazzotti
Email address: fjma@gnv.ifas.ufl.edu
Phone: 954-577-6304 Fax: 954-475-4125
Mail address: University of Florida, 3205 College Ave., Fort Lauderdale, FL 33314

Other Investigator(s): H. Franklin Percival
Email address: percivalf@wec.ufl.edu
Phone: 352-392-1861 Fax: 352-846-0841
Mail address: USGS, Florida Cooperative Fish and Wildlife Research Unit, PO Box 110450, Gainesville, FL 32611

Project Summary: Amphibians are present in all habitats and under all hydrologic regimes in the Everglades. The species present and the occupancy rate of a given species differ greatly across those gradients. These differences are due to hydropattern, vegetation, and other environmental factors. The combination of species composition and proportion of each habitat occupied at a given time form unique communities defined by those environmental factors. Therefore, if these communities can be reliably defined and measured, Everglades restoration success can be evaluated, restoration targets can be established, and restoration alternatives can be compared. This project will develop methodologies for defining and measuring the membership and area occupancy and of amphibian communities. Further, we will investigate the relationship of occupancy, survival, movement probability, and density of amphibians with hydroperiod and other environmental factors. Finally, we will provide a method for measuring restoration success based on these communities.

The importance of amphibian communities to Everglades restoration has been recognized and listed as critical priority research needs (see USGS Ecological Modeling Workshop and the DOI Science Plan in Support of Greater Everglades Ecosystem Restoration).

Project Objectives and Strategy: We will use established sampling methodologies such as mark-recapture to investigate survival, movement, and density, develop new methods for sampling across hydroperiod gradients (drift fence arrays), and use newly developed statistical techniques to estimate the proportion of area occupied by and to define amphibian communities. Our objectives include:

• Define amphibian communities appropriate for evaluating restoration success.
• Develop methods for measuring the area occupancy of amphibian communities across habitats and environmental gradients.
• Investigate the relationship of occupancy, survival, movement probability, and density with hydroperiod and other environmental factors.
• Develop restoration targets for the amphibian community of the Everglades.
• Develop a restoration tool for amphibian communities that measures restoration success and compares restoration alternatives.
• Develop an index of biological integrity for amphibians that provides a framework for scientifically defensible decisions by restoration managers.

Potential Impacts and Major Products:

• Tools and scientific data necessary for evaluation of restoration success and comparison of restoration alternatives.
• Methods and data necessary for RECOVER's adaptive assessment process and monitoring program.
• Development of a cost-effective monitoring program for amphibians.
• Development of performance measures for amphibian communities.
• Peer-reviewed publications and published methodology for evaluation of restoration success.

Products for FY03:

Waddle, J.H., K.G. Rice, and H.F. Percival. 2003. Using personal digital assistants for collection of wildlife field data. Bulletin of the Wildlife Society. In Press. MS# 02-235E. 5 pp.

Maskell, A.J., J.H. Waddle, and K.G. Rice. 2003. Osteopilus septentrionalis: diet. Herpetological Review. In Press. MS #AN02-40. 3 pp.

We also gave several presentations at National Meetings and GEER in FY03. We are currently working on a further journal manuscript and have presentations accepted at national and international meetings.

Collaborators: University of Florida, NPS-Big Cypress National Preserve

Clients: NPS, USFWS

B. WORK PLAN

Title of Task 1: Use of Amphibian Communities as Indicators of Restoration Success

Task Leaders: Kenneth G. Rice, USGS; Frank J. Mazzotti, University of Florida; Franklin Percival, USGS
Phone: 954-577-6305
FAX: 954-577-6347
Task Status: Active
Task priority: High
Time Frame for Task 1: 2003-2005
Task Personnel: Hardin Waddle, UF

Task Summary and Objectives: see above

Work to be undertaken during the proposal year and a description of the methods and procedures:

During FY04, we will concentrate our work on:

• Developing methods for defining amphibian communities.
• Investigating the relationship of occupancy, survival, movement probability, and density with hydropattern.
• Developing methods for measuring the occupancy rate of communities across habitats and hydroperiod gradients.

During FY05, we will:

• Establish restoration targets for amphibian communities in appropriate habitats.
• Develop models and methods to measure restoration success across these communities and compare restoration alternatives.
• Develop an overall index of amphibian community integrity.

Duellman and Schwartz (1958) produced the first scientific survey of the amphibians of South Florida. This work serves as an excellent reference for the historical distribution of many species before the extensive habitat loss in South Florida during the second half of the 20^th century. Meshaka et al. (2000) produced a species list of the herpetofauna for ENP, but little information about the habitat associations and population status of the species was contained in that report. Dalrymple (1988) provided a good description of the herpetofauna of the Long Pine Key area in ENP, but no attempt has been made to sample amphibians throughout the Everglades.

We will use 3 primary methods to accomplish the objectives of the project:

• Proportion area occupied (PAO) by a species.
• Vocalization survey
• Time-constrained searches
• Mark-recapture.
• Proportion area occupied by a community.

All study areas will fall within Big Cypress National Preserve, but will differ with each method outlined below.

Proportion area occupied by a species.-- One problem with many of the methods used to sample amphibians is the lack of any control of the myriad environmental factors that affect the behavior and activity of the animals. Abiotic factors like temperature, humidity and hydrology as well as biotic factors like the presence of predators or conspecifics can affect the observability of amphibians. The observability of species' population is a function of the population size, the behavior of the individuals, and the ability of the observer to locate the animals in the particular habitat. Many monitoring programs simply count animals and do not control for this observability or capture probability (p). Therefore, comparisons over time or space are not possible or are biased. If the monitoring program can assume the cost of marking individual animals, then p can be determined and population size or density determined (standard mark-recapture methods, see Williams, et al. 2002). However, this would be cost prohibitive in a monitoring program for all amphibian species throughout the Everglades. MacKenzie, et al. (2002) has developed a novel approach to this problem. Rather than mark the individual, we "mark" the species. Therefore, presence/absence data from several plots within a habitat (or along a hydroperiod gradient in our study) provides an estimate of p and allows estimation of the proportion of a stratum occupied by a given species at a given time.

Sampling units will be chosen randomly within each stratum. Strata will be defined by the hydroperiod observed from existing hydrologic data and habitat type as defined by existing GIS vegetation layers. Our standardized sampling unit will be a circular plot of 20m radius. Plots will be sampled after dark to increase the probability of observing nocturnal amphibians. At each plot we will begin by listening for anuran vocalizations for 10 minutes. The abundance of each species will be categorized as: no frogs calling, one frog calling, 2-5 calling, 6-10 calling, >10 calling, or large chorus. The intensity of the vocalizations will be categorized as: no frogs calling, occasional, frequent, or continuous. After the vocalization survey, we will perform a 30-minute visual encounter survey (VES) in each plot. During this time, all individual amphibians observed will be identified to species and captured if possible. We will record the species, categorize the age (egg, larvae, juvenile, sub-adult, or adult), measure and record the snout-to-vent length and record the sex if it can be determined. We also will record the substrate and perch height of the animal. In addition to VES, in plots that are completely flooded, we will use dipnets and funnel traps to attempt to capture aquatic amphibians. We also will record several ancillary variables at each plot (air temperature, relative humidity, presence of water, water temperature, wind speed, cloud cover).

Individual species capture histories (matrix of presence/absence of each species at a sampling period and plot) and corresponding covariates (habitat, hydroperiod, temperature, humidity) will be assembled. We will then estimate the proportion of each stratum occupied by a species and the capture probability (using MLE and the logistic regression for covariates; MacKenzie et al. 2002). The best model will minimize AIC and adequately estimate the parameters in the model (the candidate model list will be developed a priori based on ecological knowledge and will not include all possible combinations). We can then use these estimates to construct appropriate communities for each stratum (see proportion of area occupied by a community below).

Mark-recapture. - We will use standard mark-recapture methods to estimate survival, capture, and movement probabilities for amphibians in Big Cypress National Preserve. We will use the Cormack-Jolly-Seber model and/or the Robust design to estimate the above parameters with maximum likelihood methods (Williams, et al. 2002). Further, we will use covariates such as hydroperiod to investigate the effects of hydrology on the population parameters. Our main objective is to define the controlling environmental factor on the population parameters. This factor will be used in the definition of communities (see proportion of area occupied by communities below).

Figure 1: Schematic of sampling grid showing trap locations and location of adjacent upland habitats. [larger image]

We will erect 2" PVC pipe refugia on wooden stakes in a grid pattern 10m apart in every direction. The grid will be rectangular shape 50m X 150m, 96 total traps. The long dimension of the grid will extend from the core of the focal prairie to the edge of an adjacent upland habitat. See Figure 1 for an illustration. Each trap will be assigned a number that identifies its location within the grid, and each capture record will indicate the trap number.

A sample will consist of checking each pipe for treefrogs. All captured frogs will be identified to species, age class, and sex (if possible), measured (snout-urostyle length) and weighed to the nearest 0.5g. Each animal will be individually marked by toe removal. We will use the minimum amount of toes possible to uniquely mark the individuals, and all appropriate techniques will be employed to avoid infection at the removal site. Frogs will be released immediately after processing at the capture location.

We will sample each site four times during the closed period of the robust design. We will choose the order in which sites are sampled randomly at the beginning of each sampling period, and we will continue to sample sites continuously in that order over consecutive days until each site has been sampled four times. The number of sites that can be checked in a day will vary based on number of new and previously marked individuals captured and the location of each site, but effort will be made to complete all sampling in less than two weeks to avoid violating the closure assumption. Sampling will resume one month after completion of the previous sampling period.

The first sampling occasion will be timed to fall approximately one month before the onset of the wet season and the subsequent inundation of the prairie habitat (e.g. May). Sampling will continue at one month intervals throughout the summer and then may be decreased to once every two months during the dry season, depending on the amount of time required to continue sampling. Sampling will continue for two years in this manner to document the annual pattern of abundance, survival, and movement of treefrogs in BICY.

Proportion area occupied by a community. - Given that species occupancy rates differ across hydroperiod gradients and that hydrology is the controlling factor of this difference (see above), we can begin to construct "communities." In the figure below (letters represent species, the size of the circle represents PAO, numbers represent hydroperiod), we can see that in short hydroperiod sites, species A and D dominate. However, as we move to longer hydroperiod sites, other species emerge as the dominate species in the community. This pattern of species composition and PAO forms the set of "communities" along the hydroperiod gradient.

[larger image]

We have seen this pattern begins to emerge in preliminary data from the Everglades (values are an estimate of the proportion of a stratum occupied by that species):

Hydroperiod	Cricket Frog	Southern Toad	Squirrel Treefrog	Pigfrog	Leopard Frog
Short	0.0000	0.5277	0.7058	0.0000	0.3101
	0.0000	0.5155	0.6495	0.3123	1.0000
	0.1525	1.0000	0.1865	1.0000	0.8564
	0.3391	0.0000	1.0000	0.8708	0.8646
Long	0.7080	0.4333	0.1718	0.7068	0.3558

At present, the method for defining and then predicting community composition and PAO is not complete. This study will develop this methodology for the Everglades. We will use data from preliminary studies (primarily recent and ongoing inventories of amphibians in ENP, BICY, and BISC) to choose the members of a given community and then model, within the PAO framework, those communities. As in PAO for a species outlined above, we will construct capture histories for each community and estimate PAO and capture probability. This community model can then be used to develop an amphibian community index (see below).

Index of Biological Integrity. -- Indices of biological integrity (IBI) were originally developed to assess conditions of riverine systems (Karr 1991, 1993) and also have been developed successfully for use in terrestrial environments (O'Connell et al. 1998). The basic premise of IBI's is that a range of conditions of ecological integrity can be defined based on the structure and composition of a selected biological community (e.g. amphibians, fish, birds, macroinvertebrates). The concept of biological integrity provides an ecologically-based framework in which species-assemblage data can be ranked in a manner that is more informative than traditional measures such as richness and diversity (Karr and Dudley 1981, Brooks et al. 1998). Therefore, the final step in this project will be to develop an amphibian community index (ACI) for evaluating the success of restoration and management of Greater Everglades Ecosystems. The ACI will be modeled after previously developed IBI's (Cronquist and Brooks 1991, Karr 1991,1993, Books et al. 1998, O'Connell et al. 1998). Essentially, we will use the PAO of communities estimated above to index or define the integrity of a given stratum. As restoration proceeds, we can use changes in the index to make informed management decisions and to measure success. Further, we can use the pattern of these communities based on hydopattern to develop restoration targets and to compare alternatives. By providing a reliable and repeatable measure of ecological quality an ACI will help managers reach scientifically defensible decisions (Brooks et al. 1998).

Literature Cited:

Boughton, R. G., J. Staiger, and R. Franz. 2000. Use of PVC pipe refugia as a sampling technique for hylid treefrogs. American Midland Naturalist 144: 168-177.

Brooks, R.P., O'Connell, T.J., Wardrop, D.H., and Jackson, L.E.: 1998, 'Towards a Regional Index of Biological Integrity: The Example for Forested Riparian Systems,' Environmental Monitoring and Assessment, 51, 131-143.

Croonquist, M.J. and Brooks, R.P.: 1991, 'Use of avian and mammalian guilds as indicators of cumulative impacts in riparian-wetland areas,' Environmental Management 15, 701-714.

Dalrymple, G. H. 1988. The herpetofauna of Long Pine Key, Everglades National Park, in relation to vegetation and hydrology. Pp 72-86 In: Szaro, R. C., K. E. Stevenson, and D. R. Patton, eds. The management of amphibians, reptiles and small mammals in North America. U.S. Dept. of Agriculture, U.S. Forest Service Symposium, Gen. Tech. Rept. RM-166, Flagstaff, AZ.

Donnelly, M. A., C. Guyer, J. E. Juterbock, and R. A. Alford. 1994. Techniques for marking amphibians. In Heyer, W. R., M. A. Donnelly, R. W. McDiarmid, L. C. Hayek, and M. S. Foster, editors. Measuring and monitoring biological diversity: Standard methods for amphibians. Smithsonian Institution. Washington, D.C.

Duellman, W.E. and A. Schwartz. 1958. Amphibians and reptiles of southern Florida. Bull. Florida State Mus., no. 3.

Enge, K. M. 1997. A standardized protocol for drift-fence surveys. Florida Game and Fresh Water Fish Commission Technical Report No. 14. Tallahassee. 69 pp.

Karr, J.R. : 1991, 'Biological integrity: a long-neglected aspect of water resource management,' Ecological Applications 1, 66-84.

Karr, J.R. : 1993, 'Defining and assessing ecological integrity: beyond water quality,' Environmental Toxicology and Chemistry 12, 1521-1531.

Karr, J.R. and Dudley, D.R. : 1981, 'Ecological perspective on water quality goals,' Environmental Management 5, 55-68.

MacKenzie, D.I., J.D. Nichols, G.B. Lachman, S. Droege, J.A. Royle, and C.A. Langtimm. 2002. Estimating site occupancy rates when detection probabilities are less than one, Ecology. In Press.

Meshaka, W.E., W.F. Loftus, and T. Steiner. 2000. The Herpetofauna of Everglades National Park. Florida Scientist 63(2): 84-103.

O'Connell, T. J., Jackson, L.E., and Brooks, R.P. : 1998, 'A Bird Community Index of Biotic Integrity for the Mid-Atlantic Highlands,' Environmental Monitoring and Assessment, 51, 145-156.

Williams, B.K., J.D. Nichols, and M.J. Conroy. 2002. Analysis and management of animal populations. Academic Press, London. 817 pp.

Planned Outreach:

• Peer-reviewed publications and published methodology for evaluation of restoration success.
• Interpretation of results for management personnel and restoration groups through web-based reports and presentations at local meetings.
• Presentations at national conferences.
• Data and methodologies will be made available through the USGS SOFIA website.

C. BRIEF DESCRIPTION ON HOW PROJECT TASKS SUPPORT THE DOI AND USGS EVERGLADES RESTORATION SCIENCE PLANS:

• This project addresses several science objectives in the USGS Science Plan in Support of Everglades Restoration. Primarily, this work is concentrated under Goal 2B "Restore, Preserve, and Protect Natural Habitats and Species - Ecological Indicators." The tasks directly address 4 of the 5 science objectives:
• We examine the effects of hydropattern and develop information required for restoration targets (2B-SG1).
• We began a monitoring program to establish baselines and examine the current state of Everglades amphibian communities (2B-SG3).
• We have developed a monitoring program for amphibian communities throughout the Everglades for use in evaluation of restoration success (2B-SG4).
• We are developing models as tools for predicting the effects of restoration alternatives and determining restoration success (2B-SG5).
• The need for monitoring and modeling amphibian populations during CERP is specifically mentioned in the DOI Science Plan in Support of Everglades Restoration under several projects such as the Golden Gate Estates Hydrologic Restoration Project and the Ten Mile Creek Reservoir Assisted STA.
• In the DOI Science Plan, the need for monitoring and simulation projects for indicator species is listed in at least the F, K, L, O, R, and S hydrology projects.