Sampling Nekton in different habitats

Chapter 9 -Nekton in Estuaries
Sampling Nekton in different habitats

• 
  Fish and invertebrates that swim use and migrate between different habitats
  

• 
  We would like to have a quantitative estimate of density or biomass for nektonic species so that we can understand and model estuarine ecosystems, but it is still largely guesswork because of:
  
• 
  Gear selectivity (net have different sized meshes, trawls scare nekton)
  
• 
  Net avoidance by nekton (they move!)
  
• 
  Habitat interacts with gear (seines and trawls get hung on oysters)
  
• 
  Lack of quantitative estimates with some gears (density in environment unrelated to density in gear)
  
• 
  Lack of absolute density estimation (relative estimates are achievable)
  
• 
  Fish behavior affects gear effectiveness (pelagic vs. benthic species)
  
• 
  catch efficiency = recovery efficiency * capture efficiency
  

• 
  catch efficiency is estimated by releasing a known number of nekton in a large enclosed region (a pond or pool or section of the estuary), then sampling with the gear within the enclosed region repeatedly. The average catch per m² is divided by the known stocked density per m² to give a catch efficiency
  

• 
  recovery efficiency is measured by releasing marked nekton inside the gear and estimating recovered upon examining the catch
  

• 
  capture efficiency is [catch efficiency/recovery efficiency]
  

• 
  example: catch efficiency = 0.76, recovery efficiency = 0.90 and capture efficiency = 0.84 for a 1.0 m² throw trap
  

Rozas and Minello (1997). Estimating densities of small fishes and decapod crustaceans in shallow estuarine habitats: a review of sampling design with a focus on gear selection. Estuaries 20(1): 199-213.

Towed nets:

Dip nets - Not quantitative; not effective due to vegetation

Otter trawl - easy to use, large area sampled; catch efficiency is low and variable (17 - 53 % for shrimp; 6 % for spot; 26 % for croaker; 7 % for anchovies); difficult to standardize; ineffective in vegetation

Beam trawl - easy to use, large area sampled; catch efficiency is low and variable; ineffective in vegetation

Epibenthic sled - easy to use, large area sampled; catch efficiency is low and variable; difficult to standardize; ineffective in vegetation

Surface trawl - easy to use, large area sampled; catch efficiency is low and variable; difficult to standardize; ineffective in vegetation

Seine - easy to use, large area sampled; catch efficiency is low and variable (53 % for striped killifish; 52 % for Atlantic menhaden; 40 % for white mullet; 33 % for mummichog; 23 % for spot) but no corrections can be made because these numbers are not stable; sampling area is difficult to define; difficult to use in vegetation, rocky bottoms

Passive samplers:

Channel net - large sampling area; recovery low for some species; gear avoidance by some species; catch efficiency unknown

Fyke net - large sampling area; recovery low for some species; gear avoidance by some species; catch efficiency unknown

Flume net - Recovery is high and measurable; large sample area; restricted to intertidal, near marsh edge; structure may attract nekton; catch efficiency unknown

Light trap - inexpensive to construct; catch efficiency difficult to estimate; sampling area undefined; species selective;

Breder traps - inexpensive to construct; catch efficiency difficult to estimate; sampling area undefined; species selective, may not sample all fishes

Heart traps - inexpensive to construct; catch efficiency difficult to estimate; sampling area undefined; may not sample all fishes, species selective

may not sample all fishes

Pit traps - Inexpensive to construct; catch efficiency difficult to estimate; sampling area undefined; may not sample all fishes, species selective; predators and sediment are problems

Simulated Aquatic Microhabitats (SAMs) were used by Kneib (1997) to estimate abundance of salt-marsh nekton in Georgia (petri-dish in marsh)

• 
  Primary species collected: mummichog Fundulus heteroclitus
  
• 
  Correlated actual density with SAM-measured density
  
• 
  Of fish introduced, 34 % - 72 % were collected by the SAMs
  

• 
  a standing stock of 15.8 million young nekton were estimated to be present in the marsh and could not be estimated any other way
  

Encircling gear:

Block nets - Recovery efficiency measurable, but variable, depending on method of removal; large area sampled; may not sample all fishes, low capture efficiency; tide dependent

Throw traps (Wegner ring) - Catch efficiency high (70 - 76 %); recovery efficiency high (90 % or higher for most species), but poor for heavy vegetation; destructive sampling; does not capture large fishes well, small sampling area

Drop net - Catch efficiency high ( 97 % for shrimp 98 % for goby; 96 % for penaeid shrimp; 94 - 100 % for fish) recovery efficiency can be measured, usually high (82 - 98 %); small sampling area; under-samples large fishes

Barrier seine - catches large fishes; catch efficiency has not been measured; recovery efficiency measured (44 %), but variable, depending on method of removal; large area sampled; may not sample all fishes; structures may attract nekton

Purse seine (haul seine, swipe net) - Recovery efficiency measurable, but variable, depending on method of removal; large area sampled; may not sample all fishes, low capture efficiency

Flume weir - Recovery efficiency is high; large sample area; added structure attracts nekton.

Pop net - Capture efficiency high (94 - 100 % for fishes); recovery efficiency is high; inexpensive; added structure attracts nekton, difficult to deploy in oysters, thick emergent vegetation

Bottomless lift net - Recovery efficiency is high; inexpensive; intertidal habitat only; small sampling area
Entangling gear:

Gill nets - useful for collect large adult nekton for stomach analysis or otolith studies; poor recovery efficiency and capture efficiency; species selective (catches spiny fishes); sampling area difficult to define

Trammel nets - useful for collect large adult nekton for stomach analysis or otolith studies; poor recovery efficiency and capture efficiency; species selective (catches spiny fishes)
Other (experimental) approaches:

Acoustics - water is transparent to sound; easy to measure

• 
  passive acoustics useful in identifying certain species of fishes that produce sounds (Sciaenidae); cannot estimate absolute abundance at this time (but we are trying to do this here at ICMR/ECU Biology and Physics);
  
• 
  active acoustics useful in estimating density and biomass, and fish size; but not species identity; species need to have swim bladder; some fish (shad) may avoid high frequency sounds (echosounders use ultrasonic frequencies, but fish may hear them because that's what dolphins use, too). Active and passive together may be useful.
  

Video/visual census - great if the water is clear, but it is almost never clear enough in an estuary to use visual census techniques.

• 
  Line transects
  
• 
  point circle counts have been used on coral reefs.
  
• 
  SCUBA required.
  
• 
  Visual identification difficult for novices