213 
Advantages of using crest nets 
to sample presettlement larvae 
of reef fishes in the Caribbean Sea 
Cormac J. Nolan (contact author) 
Email address: cormac. nolan@ucd.ie 
Marine Biodiversity, Ecology and Evolution 
School of Biology and Environmental Science 
University College Dublin 
Belfield, Dublin 4, Ireland 
Bret S. Danilowicz 
Allen E. Paulson College of Science and Technology 
Georgia Southern University 
Statesboro, Georgia 30460-8044 
Identifying the spatial and tempo- 
ral patterns of larval fish supply and 
settlement is a key step in under- 
standing the connectivity of meta-pop- 
ulations (Sale et ah, 2005). Because 
of the potentially dispersive nature of 
the pelagic larval phase of most reef 
fishes, tracking cohorts from hatching 
to settlement is extremely difficult 
(but see Jones et ah, 1999). However, 
for many studies it is sufficient to 
sample larvae immediately before set- 
tlement. Many coral reef fish species 
use mangrove and seagrass beds as 
nursery habitats (Nagelkerken et al., 
2001; Mumby et al., 2004) and larvae 
of these species must pass over the 
reef crest in order to arrive at their 
preferred settlement habitats. The 
ability to sample this new cohort of 
larval fishes provides opportunities 
for researchers to explore the intri- 
cacies of the transition from larva 
to juvenile (Searcy and Sponaugle, 
2001). Quantifying the potential set- 
tlers also provides valuable informa- 
tion about the spatial and temporal 
supply of presettlement larvae (Victor, 
1986). Therefore a number of larval 
sampling methods were developed, 
one of which is the use of crest nets 
(Dufour and Galzin, 1993). 
Crest nets are rigid-frame ta- 
pering nets that are fixed to the 
substrate in shallow water imme- 
diately behind the crest of the reef 
(see Doherty and Mcllwain, 1996 
for full description). The top of the 
crest net is above the surface of the 
water and currents and wave action 
force larvae into the mouth of the 
net. Because of the turbulence of the 
water coming over the reef crest and 
the fact that the whole water column 
is filtered, net avoidance by larval 
fishes is estimated to be minimal. 
Channel nets (Shenker et al., 1993) 
and light traps (Doherty, 1987), on 
the other hand, remain the domi- 
nant methods for sampling settle- 
ment-stage larval fishes on western 
Atlantic reefs. Surface channel nets 
are floating nets that are free to 
swivel with the prevailing current. 
Where crest nets are positioned in 
the shallow back reef, channel nets 
are positioned in deeper channels 
between mangroves, further away 
from the reef. Crest nets have been 
widely used in the Pacific Ocean to 
quantify the larval abundance of 
coral reef fishes immediately before 
settlement (Leis et al., 1998; Du- 
four et al., 2002; Leis et al., 2003; 
Mcllwain, 2003; Lecchini et al., 
2004). Despite the apparent success 
of sampling reef fishes in the Pa- 
cific Ocean with crest nets, there 
are currently no reports of crest 
nets being employed for sampling 
reef fishes in the Caribbean Sea. 
The first objective of this study was 
to simultaneously deploy crest and 
channel nets to compare the abun- 
dance and species richness of larval 
fishes sampled. It was hypothesized 
that crest nets would capture more 
larvae by sampling the whole water 
column on the reef crest as opposed 
to channel nets that sample only 
surface waters. 
Larval reef fish possess impres- 
sive swimming capabilities (Leis and 
Carson-Ewart, 1997) and have the 
ability to detect reefs at a distance 
(Myrberg and Fuiman, 2002) and can 
therefore influence their own disper- 
sal. However, many other abiotic fac- 
tors can still influence their growth, 
survival, transport, and eventual ar- 
rival at a suitable settlement habitat. 
The abundance of larvae present is 
related to lunar period in some ar- 
eas (Robertson et al., 1988, Thorrold 
et al., 1994; Sponaugle and Cowen, 
1996), but this abundance is not 
fully correlated with peaks in abun- 
dance in other areas (Kingsford and 
Finn, 1997). Larval growth rates and 
swimming ability vary with water 
temperature in some species (Green 
and Fisher, 2004) and winds can al- 
ter the strength and direction of sup- 
plying currents. The second objective 
of this study was to explore correla- 
tions between certain abiotic factors 
(lunar phase, water temperature, and 
prevailing wind) and the number of 
species and individuals collected by 
each net type. 
Materials and methods 
Study site 
Fieldwork was conducted at Turneffe 
Atoll, Belize (17°16'5"N, 87°48'57"W, 
Fig. 1A). Turneffe Atoll is part of the 
Meso-American Barrier Reef System 
(MBRS) that runs along southern 
Mexico through the waters of Belize, 
Guatemala, and Honduras. The 
MBRS is the world’s second largest 
coral reef system after the Great Bar- 
Manuscript submitted 1 June 2007. 
Manuscript accepted 10 January 2008. 
Fish. Bull. 106:213-221 (2008). 
The views and opinions expressed or implied 
in this article are those of the author and 
do not necessarily reflect the position of the 
National Marine Fisheries Service, NOAA. 
