USGS - science for a changing world

Southeast Ecological Science Center

 
An evaluation of three aquatic sampling techniques for amphibians: implications for inventory and monitoring project design

Jennifer S. Staiger
USGS - Florida Integrated Science Center
7920 N.W. 71st St., Gainesville, FL 32653

Presented at the 2005 Joint Meeting of Ichthyologists and Herpetologists, Tampa, FL, 6-11 July 2005.

SE ARMI Logo - click to go to the homepage.


Introduction and Background

Crayfish trap (CT) - click to enlarge image Crayfish trap (CT)

In response to global declines and threats to amphibian populations, the US government implemented the national Amphibian Research and Monitoring Initiative (ARMI).  The ARMI goal is to inventory and monitor amphibian populations on federal lands and to evaluate potential causes of declines.  An important component of this effort is to develop efficient sampling techniques for the diversity of amphibian types.

Automated frog call data logger (FL) - click to enlarge image Automated frog call data logger (FL)

Inventory efforts generally require sampling methods that will detect the range (common to rare) of species present, work well in a variety of habitat types, and provide data that can be used to calculate detection probabilities.  These probabilities can then be used to develop population monitoring protocols for the monitoring phase.  Methods used during this phase must reliably detect the suite of species of interest and provide data that can be used to estimate the proportion of area occupied by these species.  Of course, time and cost considerations also influence which methods may be employed in either phase.

The southeastern ARMI region encompasses a wide range of habitats and areas of high amphibian diversity across six states.  Consequently, a variety of sampling methods have been utilized during amphibian inventories in the region. 

Objective

As Southeast ARMI moves from the inventory phase to the monitoring phase at several sites, the sampling techniques used must be evaluated for their efficacy in terms of detection and effort, and to identify biases associated with their use.

Methods

Dip-net survey (DN) - click to enlarge image Dip-net survey (DN)

The data presented here were collected during inventories at four USFWS national wildlife refuges: St. Marks, FL, Okefenokee, FL and GA, Harris Neck, GA, and Savannah, GA and SC.

Time constrained dip-net surveys (DN), aquatic funnel traps (modified crayfish traps, CT), and automated frog call data loggers (FL) were regularly used to sample amphibians at these sites.  Generally, dip-net surveys consisted of 2 or more people netting for an amount of time based on the size of the wetland and the number of netters. Crayfish traps were typically deployed for two to three nights and checked daily.  Automated call loggers were typically set to record one minute every hour from approximately 18:00 to 08:00 over two to three nights. However, the use of these techniques varied somewhat between sites and surveys.


aquatic_sampling_4Okefenokee graph showing species accumulation - click to enlargeSt. Marks graph showing species accumulation - click to enlargeHarris Neck graph showing species accumulation - click to enlargeSavannah graph showing species accumulation - click to enlarge

Figures 1-4. Amphibian species accumulation curves comparing three aquatic sampling techniques at four national wildlife refuges.  Because automated call loggers (FL) cannot detect salamander species, a pair of curves for crayfish traps (CT) and dip-net surveys (DN) are presented, one excluding and one including salamander species.


Table 1. Detection probabilities for each species by method and site. 0 indicates species was not detected using that method but has been detected at the site using other methods.  Hyphen (-) indicates species does not occur at the site (based on Southeast ARMI inventories).

Table 1.  Detection probabilities for each species by method and site.  0 indicates species was not detected using that method but has been detected at the site using other methods.  Hyphen (-) indicates species does not occur at the site (based on Southeast ARMI inventories). - click to enlarge


clear pixel

Figure 5.  Comparison of number of amphibian species detected across 4 refuges using crayfish traps (CT), dip-net surveys (DN), and automated frog call data loggers (FL).  Numbers in areas of overlap indicate shared species, numbers in () are total species, followed by those exclusive for each method. - click to enlarge Figure 5.  Comparison of number of amphibian species detected across 4 refuges using crayfish traps (CT), dip-net surveys (DN), and automated frog call data loggers (FL).  Numbers in areas of overlap indicate shared species, numbers in () are total species, followed by those exclusive for each method.

clear pixel clear pixel

Table 2. Advantages and disadvantages of three aquatic sampling techniques.Table 2.  Advantages and disadvantages of three aquatic sampling techniques. - click to enlarg

     


Conclusions

  • Each method appears to work best for a particular group of species or life stage (Table 1). Large aquatic salamanders such as Siren lacertina were detected almost exclusively using CT, whereas the larval stage of small salamanders such as Eurycea quadridigitata and Pseudobranchus striatus were detected by DN only. Obviously, FL cannot detect caudates, but it consistently detected more anuran species than the other methods (Fig. 5, Table 2).
     
  • At each site, none of these methods detected all the amphibian species found during all inventory activities.  The reasons vary; some of the absent species are terrestrial (lacking an aquatic stage, e.g., Plethodon grobmani), some are explosive breeders (having very discrete breeding periods, e.g., Scaphiopus holbrookii), while others are likely extremely rare (e.g., Rana capito at Okefenokee).
     
  • Based on the detection probabilities and species accumulation curves, DN may be the best method for general inventories, whereas FL may be well suited for monitoring projects (Table 1, Figs. 1-4). However, CT will be needed if large aquatic salamander species are of interest.
     
  • Because detection probabilities will be used to determine the number of sites and visits needed to monitor amphibian populations using an occupancy model (e.g., Proportion of Area Occupied), it is important to maximize detection of a range of species.

National ARMI Website - click to go to the national site.

National ARMI Website

For more information:
jennifer_staiger@usgs.gov

Accessibility FOIA Privacy Policies and Notices

Take Pride in America logo USA.gov logo U.S. Department of the Interior | U.S. Geological Survey
URL: http://fl.biology.usgs.gov/posters/Herpetology/Aquatic_Sampling_Techniques/aquatic_sampling_techniques.html
Page Contact Information: SESC Webmaster
Page Last Modified: Friday, 07-Jan-2011 10:52:43 EST