Schobernd et al.: A comparison of numbers of fish larvae extruded from plankton nets of different mesh size 
241 
Hanisko et al., 2007), and vermilion snapper (Rhom- 
boplites aurorubens ; Hanisko et al. 2 ). Ichthyoplankton 
data from the SEAMAP surveys have also been used to 
describe larval transport, decadal changes in fish habi¬ 
tat, and annual variations in egg densities in the Gulf 
of Mexico (Johnson et al., 2009; Marancik et al., 2012; 
Hernandez et al. 3 ; Lyczkowski-Shultz et al., 2013). Ad¬ 
ditionally, SEAMAP samples have been used to assess 
the potential impacts of 1) entrainment of larvae in 
offshore liquefied natural gas facilities (Gallaway et 
al., 2007) and 2) larval mortality from the 2010 Deep¬ 
water Horizon oil spill (Muhling et al., 2012) to Gulf of 
Mexico fisheries. 
Application of such abundance-at-size data is predi¬ 
cated on the assumption that larvae caught and re¬ 
tained in plankton nets consistently and accurately 
represent the assemblage being sampled (Tranter, 
1968; Smith and Richardson, 1977). Inherent catch- 
ability issues with plankton nets, however, result in 
underrepresentation of larger, more developed larvae 
that are able to detect and avoid the net (Morse, 1989; 
Somerton and Kobayashi, 1989) and in underrepre¬ 
sentation of the smallest larvae in catches because 
they are extruded through net meshes (Colton et al., 
1980; Lo, 1983; Houde and Lovdal, 1984; Johnson and 
Morse, 1994). The effect of these sources of bias on 
larval abundance data have been widely investigated 
in studies outside the Gulf of Mexico but have never 
been addressed specifically for data generated from 
SEAMAP ichthyoplankton surveys. Only 2 previous 
studies with gear other than SEAMAP plankton nets 
investigated the extrusion of larval fish from plankton 
nets in northern Gulf of Mexico waters (Comyns, 1997; 
Hernandez et al., 2011). 
Ideally both of these biases should be evaluated 
before abundance data are interpreted and used in 
resource monitoring and environmental and fisheries 
assessments (Smith and Richardson, 1977). In recent 
stock assessments, the effect of avoidance was mitigat¬ 
ed by including in analyses only the largest size class 
of larvae that are consistently captured in the net—a 
decision based on examination of size frequency distri¬ 
butions (Hanisko et al., 2007; Hanisko et al. 2 ). Despite 
the importance of the earliest life stages in estimating 
absolute abundance or mortality, the underrepresen¬ 
tation of these values due to extrusion of the small¬ 
est larvae in SEAMAP samples has not been consid¬ 
2 Hanisko, D. S., A. Pollack, and G. Zapfe. 2015. Vermilion 
snapper ( Rhomboplites aurorubens) larval indices of relative 
abundance from SEAMAP Fall Plankton Surveys, 1986 to 
2012. Southeast Data, Assessment and Review SEDAR45- 
WP-05, 34 p. [Available from: website.] 
3 Hernandez, F. J., W. M. Graham, and K. Bayha. 2013. Spa¬ 
tial distribution and abundance of red snapper (Lutjanus 
campechanus), vermilion snapper (Rhomboplites aurorubens ), 
and red drum ( Sciaenops ocellatus ) eggs across the northern 
Gulf of Mexico based on SEAMAP continuous underway fish 
egg sampler (CUFES) surveys, 52 p. Final Report NOAA/ 
MARFIN Award Number NA09NMF4330153. [Available 
from Grants Branch, Southeast Reg. Off., Natl. Mar. Fish. 
Serv., NOAA, 263 13 th Ave. S„ St. Petersburg, FL 337101.] 
Table 1 
Cruise number, start and end dates, and number of 
tows conducted with bongo nets with different mesh 
sizes during Southeast Area Monitoring and Assess¬ 
ment Program surveys in the northern Gulf of Mexico 
between October 2005 and August 2007. 
No. of 
Cruise 
Start date 
End date 
tows 
04266 
22-Oct-2005 
01-Nov-2005 
45 
63062 
23-March-2006 
29-May-2006 
15 
63064 
31-Aug-2006 
27-Sep-2006 
7 
63075 
29-Aug-2007 
28-Sep-2007 
14 
ered in the development of SEAMAP indices of larval 
abundance. 
The use of standard gear and towing methods dur¬ 
ing plankton surveys has ensured consistent sampling 
within and among SEAMAP cruises and surveys (Mc- 
Clatchie et al., 2014; GSMFC 4 ). However, the specific 
characteristics of the SEAMAP bongo net in relation 
to loss of eggs and larvae though the meshes of the 
standard SEAMAP sampling gear have yet to be in¬ 
vestigated. Our objective was to evaluate the degree of 
extrusion of fish eggs and larvae through the standard 
SEAMAP bongo net, which has a 0.333-mm mesh, by 
comparing numbers of larvae from that net with lar¬ 
val numbers from a bongo net with a finer, 0.202-mm 
mesh. 
Materials and methods 
Field and laboratory methods 
Between October 2005 and August 2007, 81 bongo tows 
with side-by-side nets, each with a different mesh size, 
were performed during 5 SEAMAP surveys conducted 
in the Gulf of Mexico (Table 1, Fig. 1). Given the pri¬ 
mary objectives for these surveys, samples were taken 
as time permitted after standard sampling was com¬ 
pleted. The 6 i err. bongo net frame with a mouth open¬ 
ing of 0.29 m 2 that is used during standard SEAMAP 
sampling was used in the tows of our study. However, 
unlike the standard SEAMAP bongo net configuration, 
which consists of 2 nets with a mesh size of 0.333 mm, 
the configuration consisted of a net with 0.202-mm 
mesh on one side of the frame and a net with 0.333-mm 
mesh on the other side. Although no side-by-side effect 
was evaluated, it is thought to be minimal because of 
the short distance between the mouth openings of the 
2 nets on the bongo frame. Sampling was conducted 
4 GSMFC (Gulf States Marine Fisheries Commission). 2016. 
SEAMAP operations manual for trawl and plankton surveys, 
61 p. GSMFC, Ocean Springs, MS. [Available from web¬ 
site.] 
