Kupchik and Shaw: Effects of recruitment through a coastal boundary layer on growth of larval Brevoortia patronus 
209 
only monthly. In addition, 2 sampling efforts occurred 
in September 2007. The sampling design was developed 
to focus on atmospheric cold front passages, which are 
intermittent wind-dominated meteorological events 
common from the late fall through early spring along 
the Louisiana coast. Individual sampling dates were 
chosen to maximize sampling during astronomical 
tidal ranges to better evaluate the potential impacts 
of these atmospheric events. Ichthyoplankton were col¬ 
lected randomly and passively from both the surface 
and near-bottom. Collections at the surface were 6 min 
in duration, and net samples collected near-bottom 
were 10 minutes. These differences in sampling dura¬ 
tion were chosen to attempt to have similar volumes 
of water filtered through the net. To prevent contami¬ 
nation of the net during deployment for near-bottom 
sampling, the net mouth was closed until the ring net 
was at depth, and was closed again after sampling for 
retrieval. Nets were rinsed and washed down with 
fresh water to avoid sample contamination. All ichthyo¬ 
plankton sampling was conducted under pre-approved 
planning and authorization by the Institutional Animal 
Care and Use Committee. 
Ichthyoplankton samples were initially preserved in 
10% buffered (sodium phosphate, dibasic NaH 2 PO 4 -H 20 
and monobasic Na 2 HP 04 ) formalin—a short-exposure, 
long-term fixative—for approximately 3.5 h. Samples 
were then rinsed and transferred to a 70% ethanol so¬ 
lution for long-term storage and for later examination 
of the larval fish otoliths. 
Estuarine hydrographic parameters were measured 
at dockside during each plankton sampling event by 
using a portable YSI Model 85 instrument (YSI Inc., 
Yellow Springs, CO) to collect data on temperature, 
conductivity (salinity), and dissolved oxygen at each 
depth during net deployment. Data concerning pre¬ 
dicted diurnal tides, measured tide height, and the 
resulting alteration in the expected tidal prism were 
collected from a nearby tide gauge station (station 
ID: 8762075; NOAA Tides and Currents, website) at 
the Port of Fourchon, Belle Pass, Louisiana (29°6.8'N, 
90°11.9'W). 
Laboratory methods 
A Motodo Plankton Splitter (Aquatic Research Instru¬ 
ments, Hope, ID) was used to split samples with vol¬ 
umes greater than 200 mL in half and to split samples 
with a volume greater than 400 mL into quarters. All 
ichthyoplankton were removed from the samples and 
placed into 10-mL scintillation vials by using a dis¬ 
secting stereoscope. To ensure that all ichthyoplankton 
were removed, a random subset of samples from both 
surface and near-bottom collections were checked by a 
second party after initial processing. 
Ichthyoplankton were identified to the lowest taxo¬ 
nomic classification possible; however, size and physical 
condition were potentially limiting factors for definitive 
confirmation of identification. Alizarin blue and aliza¬ 
rin red were used to confirm meristic counts for indi¬ 
vidual ichthyoplankton that were difficult to identify. 
Gulf menhaden larvae were separated and stored in 
70% ethanol for otolith analysis. Identifications of lar¬ 
val fish were based on identification guides by Richards 
(2005) and Fahay (2007). 
Gulf menhaden were subsampled from each surface 
and near-bottom sample for otolith analysis on the basis 
of the normal distribution of SL of all Gulf menhaden 
larvae collected. Measurements of SL to the nearest 0.1 
mm were taken with a Leica MZ6 stereoscope (Leica 
Microsystems, Buffalo Grove, IL) calibrated against a 
microscope stage micrometer. Gulf menhaden larvae 
were subsampled from every sampling effort that con¬ 
tained the target species. In samples where 3 or fewer 
Gulf menhaden larvae were collected, all larvae were 
selected for otolith removal. In samples that contained 
more than 3 Gulf menhaden larvae, 3 larvae were se¬ 
lected so that a larva with the longest SL, shortest SL, 
and a SL from the normal distribution was chosen for 
otolith removal. Removal, preparation, analysis and 
otolith interpretation were undertaken according to the 
methods described in Kupchik and Shaw (2016). 
Age determination and spawning dates 
Age of larval gulf menhaden, recorded in days after 
spawning (das), was determined from the counts of 
growth increments (otolith radii) by using a semi-au¬ 
tomated image analysis method (Kupchik and Shaw, 
2016). Daily increment deposition has been confirmed 
to have an increment-count to age-regression slope of 
1 for larval Gulf menhaden growth (Geffen, 1992). As 
with the method used by Raynie (1991), we applied a 
5-day lag for the first increment formation after spawn¬ 
ing for Gulf menhaden larvae on the basis of laboratory 
research (Warlen, 1988). This lag resulted in a calcula¬ 
tion of total age das where 5 days were added to the 
number of increments from read otoliths. For modeling 
growth, we applied a 3-day lag for first increment for¬ 
mation after hatching (Warlen, 1988). This resulted in 
a calculation in total age in days after hatching (dah) 
where 3 days are added to the number of increments 
determined from otolith reading. Ages were estimated 
for larvae not selected for dissection by using frequen¬ 
cy of age-at-length keys and the FSA package, vers. 
0.7.4, for R software, vers. 3.1.1 (R Core Team, 2014). 
Spawning dates were calculated for all Gulf men¬ 
haden larvae; direct calculation was made for those 
larvae where otolith radii were analyzed, and also for 
those where the age was estimated with the method 
described by Isermann and Knight (2005), namely with 
a semi-random method in the FSA package. The spawn¬ 
ing date was determined as the difference between the 
date of capture and the age in days after spawning. 
Growth rates 
Distributions of lengths and calculated ages based on 
increment counts were tested for normality by using a 
Shapiro-Wilk’s test. Instantaneous larval growth (per 
