Jennifer Rehage (Florida International University) Michael Robblee (USGS)
The objectives of this study were to (a) Develop quantitative sampling programs and methods from studies of stream-channel fish communities through a literature search, (b) Test and develop methods for measuring catch per unit effort and density in oligohaline habitats, (c) Document the composition of native and introduced fishes and their habitat use, (d) Measure important correlative physical measurements such as relative water depths in channels and wetlands, salinity, temperature, and (e) Test methods of analyzing elemental ratios in bones to determine extent of fish movements from interface refuges into sloughs and peripheral wetlands using a combination of caged and wild fish.
11200 SW 8th Street
Eklund, A. M.
S. Davis and J. C. Ogden, editors
Loftus, W. F.; Jordan, F.; Chick, J. H.; Kandl, K. L.; McElroy, T. C.; Bass, Jr., O. L.
J. W. Porter and K. G. Porter, editors
The full article is available via journal subscription or single article purchase. The abstract may be viewed on the American Fisheries Society website.
Manning, R. B.
The full text is available free for download
Loftus, W. F.; Trexler, J. C.; Ulanowicz, R. E.
The full article is available via journal subscription or single article purchase. The abstract may be viewed on the SpringerLink website.
Kushlan, J. A.
The article may be viewed online at the website below.
Loftus, W. F.
Posted with permission from the Bulletin of Marine Science
Gutreuter, S.
The full article is available via journal subscription or single article purchase. The abstract may be viewed on the website below
Kennen, J. G.; Goldstein, R. M.; Hambrook, J. A.
All small species of fishes and macroinvertebrates are preserved for vouchers and bought back to the lab for processing. We make notes about minnow trap catch per unit effort (CPUE) in the same CATCH datasheets used for the electrofishing. Back at the FIU lab, processing involves identifying each specimen to the species level and obtaining its wet weight (in grams) and measurements (usually standard length for fishes, and carapace length for shrimp). We use a 0.01 g accuracy scale, and digital calipers (0.01 mm accuracy) for all measurements.
In the dry season of 2008, supplementary sampling was begun to target palaemonid shrimp species along the salinity gradient of the entire Shark River Estuary. This sampling is conducted with minnow traps at 8 sites that expand the entire estuary.
Sampling is conducted three times per year, in the wet season, dry season, and in the transition between those seasons.
All species caught are identified to the species level, including the Palaemonid shrimp, unless specimens are too small to have/show key identifying characters.
YSI readings do not accompany minnow trap data for first 4 sampling events of project (3 samples of year 1 and first sample of year 2), but were estimated from the same site location from the electrofishing data.
Any use of trade, product, or firm names is for descriptive purposes only and does not constitute endorsement by the U.S. Government
Research Methods a. Field/laboratory: At each creek, we systematically sample three, 100 m-long sections of creek bank by electrofishing (0-100 m, 200-300 m, and 400-500 m). Each electrofishing bout lasts five minutes (pedal time). We use a boat-mounted, generator-powered electrofisher (two-anode, onecathode system with a Smith-Root GPP 9.0 control box). Electrofishing is an effective method for sampling large fishes in freshwater habitats, and electrofishing catch per unit effort (CPUE) provides a reliable index of fish abundance. For all bouts, electrofishing power is standardized to 1500 watts according to ambient temperature and conductivity conditions. Because creek width is considerably greater than the electric field generated by the electrofisher, we randomly select either a left or right bank for each bout. Following USGS-NAWQA guidelines, all electrofishing is conducted using intermittent application of electrical current to prevent fish from fleeing deep beneath mangroves. All fish captured are placed in a holding tank, identified, measured (to the nearest 1-mm standard or total length), weighed (if necessary), and released after full recovery. We measure mass of those species for which we do not have reliable length-mass regression equations with which to estimate mass - this procedure results in less handling of fishes and reduces stress on the specimens. Only non-indigenous species are saved and preserved in 10% buffered formalin to be returned to the laboratory for processing. Although electrofishing typically targets large fish species, we routinely capture specimens as small as 5-cm standard length (SL). We sample the upper 100-m reaches of each creek with 3-mm mesh metal minnow traps. The minnow trap is an entrapment gear that typically targets small fishes and invertebrates, as well as juveniles of larger species. Minnow traps have two conical-shaped funnel openings (2.54 cm diameter) that inhibit animals from escaping once they enter the trap. Minnow traps are most efficient in capturing bottom-dwelling fishes and invertebrates, although it has been our experience that top-dwelling fish such as eastern mosquitofish (Gambusia holbrooki) are caught commonly in traps set at the surface of the water column. Thus, we use paired traps, one set on the bottom and the other suspended at the top of the water column, attached to mangrove prop roots. We deploy five pairs of traps per creek randomly along the uppermost 100-m stretch in each creek. The traps are deployed unbaited overnight along the fringing mangroves for an approximately 24-h catch effort. Fishes and decapods captured from minnow traps are preserved in 10 % buffered formalin and returned to the laboratory for processing. According to standardized animal care guidelines, we euthanize all collected fish with an overdose of MS-222 prior to fixation. Specimens are retained in formalin for approximately one week to allow proper fixation, and then are preserved in 70 % ethanol before processing. During each sampling event, we use a YSI 85© unit to record physico-chemical parameters (water temperature, specific conductance/salinity, and dissolved oxygen) at the beginning of each electrofishing bout. We measured water clarity and bottom type with a measuring stick, and turbidity with an electronic turbidity meter. These measurements allow us to relate patterns of species abundance and community structure to abiotic conditions in the mangrove creeks. Beginning with the February 2006 collection, we also recorded those parameters when retrieving minnow traps by using the handheld YSI 85©, and we did the same during the palaemonid shrimp sampling.
In the 2008 dry season, we started a new sampling effort focused on documenting the distribution and abundance of palaemonid shrimp along the entire Shark River estuary. The dry-season sample was conducted during the spring, and the wet-season sample will be conducted in the fall. We sample shrimp species at eight sites from the marsh-mangrove ecotone at the downstream end of Shark River Slough to the marine environment in Ponce de Leon Bay.
All sites were sampled with minnow traps, set deep in the water column (avg. dry season depth = 0.79 m) along mangrove prop roots at the edges of the open channels of creeks, bays, and rivers. The sampling sites are co-located with our MAP sites and the FIU LTER sites, and are evenly distributed to capture the gradual shift in salinities from freshwater upstream to brackish and marine conditions downstream. At each site, 10 unbaited Gee® minnow traps (3-mm metal mesh, 2.5-cm opening) are deployed overnight and unbaited along 100-m stretch of shoreline. Four nights of consecutive sampling are conducted in both seasons (10 traps x 8 sites x 4 nights x 2 seasons = 640 traps of effort). Preliminary sampling showed that traps average approximately 5-10 shrimp per trap.
Specimens collected in the field are placed in 10% buffered formalin and brought back to the lab for processing. After about one week, specimens are transferred to 70% ethanol. All specimens are identified to the species, measured and counted using a Leica dissecting microscope-mounted camera to image and measure specimens.
These items are used in every sampling trip: 1. Boat-mounted, generator-powered electrofisher (two anode, one-cathode system with a Smith-Root GPP 9.0 control box) 2. Unbaited Gee® minnow traps (3-mm metal mesh, 2.5-cm opening) 3. YSI 85 multimeter unit 4. Garmin MAP 75 GPS 5. Water turbidity meter
All species caught are identified to the species level, including the Palaemonid shrimp, unless specimens are too small to have/show key identifying characters. We use a Leica 7.5MZ dissecting microscope for identification of our smaller specimens, including the palaemonid shrimp species.
QA/QC Methods a. Field/laboratory: Because all large fishes, except for the non-native species, are returned to the water after being identified, weighed, and measured, it is critical that we take steps to accurately record all data and safeguard the data sheets in the field. The CATCH datasheets we use for recording catches are waterproof and are secured in a metal clipboard to prevent loss in the field. On return from the field, we immediately make photocopies of the CATCH datasheets and of the SAMPLING datasheets that describe the sampling effort (site, time and date of sample, physiochemical covariates, and field notes).
We identify large fishes in the field, with one PI verifying the identification of the second person. PIs and technicians are intimately familiar with the fishes that occur in the Everglades, but if there is a question, we preserve the fish and bring it back to the laboratory for verification. We send any unusual fish to experts on that particular taxon to confirm our identifications. We also photograph all species taken in the field at their first collection to use as identification vouchers, and preserve unusual or rare specimens to be added to our reference collection. We preserve all small species of fishes and macroinvertebrates for vouchers and bring them back to the lab for processing. We make notes about minnow trap CPUE in the same CATCH datasheets used for the electrofishing. Back at the FIU lab, processing involves identifying each specimen to the species level and obtaining its wet weight (in grams) and measurements (usually standard length for fishes, and carapace length for shrimp). We use a 0.01 g accuracy scale, and digital calipers (0.01 mm accurary) for all measurements. Lab PROCESSING datasheets are filled out detailing specimen measurements. These datasheets are kept in a binder in the lab, and copies of these PROCESSING datasheets are kept at the PI’s office and home. In the binder, the original PROCESSING datasheets are kept along with copies of the CATCH and SAMPLING datasheets from the field. A binder is used per sampling year: 2004-05, 2005-06, 2006-07, and 2007-08. Original CATCH and SAMPLING datasheets are kept at the PI (Rehage)’s office at FIU. Any questionable specimen is set aside in a vial, and its identity is verified with experts in the field, and sent to them if needed. We use a number of field guides to aid in identification, as well as our own reference collection of specimens built from past sampling.
We calibrate our water-quality meters in the lab before the sampling run, and recalibrate them in the field at the beginning of each day. We perform routine maintenance and calibration on all sample collection and sample processing equipment.
Data synthesis: Preliminary assessment of the data collected is performed at the end of each sampling event, upon completion of sample processing, data entry, and validation. A complete assessment and synthesis of the data collected (all years combined) is conducted yearly in conjunction with the preparation of annual reports.
Analysis Methods and design:
All data from SAMPLING, CATCH and PROCESSING datasheets are entered in spreadsheet format using Microsoft Excel. Separate files are created and managed for the two sampling methods (electrofishing and minnow trapping). Since analyses are multiyear, all years of data are combined into a single file, and as new data is gathered, it is added to a master datafile containing all previous years of data. Post data-entry, data are verified and validated by a different technician from the technician that entered the data. Then, the data files are cleaned up and formatted for statistical analyses. We use Systat and SAS for data analyses and Microsoft Excel and SigmaPlot 2000 for creating graphs and tables.
Statistical or comparative analysis: We use an analysis of deviance (ANODE) approach comparing models to examine the contribution of key hydrological and physiochemical parameters to the variation in the CPUE data. The full model contains three hydrological covariates: DSLD (days since last dry-down) for marshes upstream of creeks by using water depths from stations SH1 and P38, and water levels at those stations at the time of sampling aggregated over the 15 days prior to that sampling date. The model also includes four other physicochemical variables measured at the time samples were collected: salinity, water depths in creeks (cm), dissolved oxygen (mg/L), and water temperature (degrees C). For model selection, we use PROC GLMSELECT in SAS® with the stepwise selection option (with competitive drop), and compared AIC values. AIC’s represent the ‘distance’ between fitted models and an ideal model that perfectly represents the data, thus smaller AIC values indicate a better fit.
Analysis of Variance (ANOVAs) are then performed using the subset of variables identified via model selection. This model-selection approach is typically applied to four focal variables: (a) CPUE for large fishes caught in electrofishing samples (number 5 min.-1 pedal time); (b) CPUE of small fishes caught in minnow traps (sum of top and bottom trap per 24 h-1); (c) CPUE of only the marsh inhabitants caught via electrofishing; and (d) CPUE of only the marsh inhabitants caught via minnow traps. To satisfy assumptions of parametric tests, all CPUE measures are ln-transformed (observed value + 1) prior to analysis. Post-hoc, pairwise comparisons are performed using Tukey’s. We examine relationships among CPUE’s and between CPUE and key abiotic parameters using simple linear regression.
We analyze variation in palaemonid shrimp CPUE with an Analysis of Covariance (ANCOVA), which examines variation among sites, sampling months and days and included the following covariates: water temperature, dissolved oxygen, salinity, and trap depth (this is the depth at which the trap was set along mangrove prop roots). We also examine variation in shrimp CPUE by using the delta approach (following Serafy et al.), where we first look at the presence of palaemonid shrimp across sites (occurrence), their abundance once present (concentration), and then calculate an index of relative density called ‘delta-density’ (product of occurrence and concentration). All analyses are performed using Systat® 10.0 and SAS® 9.1.3.
Interpretation and analyses of the data is conducted yearly at the time of preparation of annual reports. Interpretation involves examining newly collected data within the context of all previous years of data, so that interpretation is comprehensive.
Any use of trade, product, or firm names is for descriptive purposes only and does not constitute endorsement by the U.S. Government
11200 SW 8th Street
Parameters collected for the minnow trap data include: DATE, SEASON, SITENAME, PAIR, LOC, S1_Count, S2_Count, S3_Count, S4_Count, S5_Count, S6_Count, S7_Count, S8_Count, S10_Count, S12_Count, S30_Count, S34_Count, S36_Count, S37_Count, S38_Count, S41_Count, S42_Count, S46_Count, S47_Count, S54_Count, S58_Count, S67_Count, S71_Count, S72_Count, S75_Count, S77_Count, S82_Count, S84_Count, S88_Count, S103_Count, S104_Count, S106_Count, S115_Count, CPUE, DEPTH_Creek, TEMP, DO, DO_perSat, and Specific _Conductance
Minnowtrap Data from Rookery Branch and the North, Watson, and Roberts Rivers Everglades National Park from November 2004 to April 2008, Jennifer Rehage, Florida Coastal Everglades Long Term Ecological Research, LT_TDCS_Rehage_002
Southeast Environmental Research Center University Park, ECS 253
Southeast Environmental Research Center University Park, ECS 253
U.S. Department of the Interior, U.S. Geological Survey
Comments and suggestions? Contact: Heather
Henkel - Webmaster
Generated by mp version 2.8.18 on Wed May 18 09:25:11 2011