554 
Abstract— In 1996, using the contin- 
uous underway fish egg sampler (CU- 
FES), we carried out a preliminary study 
with the daily egg production method 
(DEPM) for estimating fish biomass of 
Pacific sardine, Sardinops sagax. Full- 
water-column abundance of sardine eggs 
was correlated with the abundance of 
eggs taken at 3-m depth with a CUFES; 
however direct conversion of CUFES 
samples to the full-water-column abun- 
dance, required by the DEPM, would 
add considerable variance to an esti- 
mate of daily egg production. Our pre- 
liminary study also indicated that the 
average size of an egg patch for Pacific 
sardine was 22 km diameter and that 
all stages of sardine eggs were not 
equally vulnerable to a CUFES at 3-m 
depth. 
Using these findings as a guide, in 
1997 we carried out an adaptive alloca- 
tion DEPM survey in which the num- 
bers of sardine eggs collected with the 
CUFES were used to determine sub- 
sequent locations of vertical net tows. 
All the vertical net tows were taken 
in a high-density stratum where the 
CUFES collected at least two eggs per 
minute; egg density in this stratum was 
computed by using only vertical tows. 
The remaining survey area, where the 
CUFES collected fewer than two eggs 
per minute, constituted a low-density 
stratum where egg density was esti- 
mated by using the ratio of CUFES egg 
abundance in these two strata multi- 
plied by the egg density in the high-den- 
sity stratum. Conventional statistics 
were used because the sampling units 
were survey lines spaced at 22-km or 
greater intervals. An acceptable level 
of precision (CV=21%) for a daily pro- 
duction of 2.57 eggs 0.05 m 2 /d (51.4 
eggs m 2 /d), was achieved by using only 
141 vertical net tows. Therefore, the 
CUFES will enhance the DEPM by 
increasing precision of the estimates, 
or by reducing costs in relation to a 
survey composed of only vertical water- 
column tows, when the CUFES is used 
adaptively to establish sampling strata 
for vertical water column tows. 
Manuscript accepted 19 March 2001. 
Fish. Bull. 99:554-571 (2001). 
Use of a continuous egg sampler for 
ichthyoplankton surveys: application to 
the estimation of daily egg production of 
Pacific sardine ( Sardinops sagax ) off California 
Nancy C. H. Lo 
John R, Hunter 
Richard Charter 
Southwest Fisheries Science Center 
National Marine Fisheries Service, NOAA 
8604 La Jolla Shores 
La Jolla, California 92037 
E-mail address (for N C. H. Lo): Nancy.Lo@noaa.gov 
The continuous underway fish egg sam- 
pler (CUFES) (Checkley et ah, 1997; 
van der Lingen et ah, 1998; Checkley 
et ah, 1999; Watson et ah, 1999) is a 
new device that provides high-resolu- 
tion spatial maps of fish eggs by siev- 
ing the eggs from water pumped from 
a fixed depth while a survey ship is 
underway. These data may be used as 
an index of fish abundance, or to study 
spawning habitats, and if converted to 
the numbers of eggs in the full-water- 
column, they may be used to estimate 
fish biomass by using one of the egg 
production methods (Hunter and Lo, 
1997). 
The objective of our study was to 
evaluate the use of the CUFES in the 
daily egg production method (DEPM) 
of estimating the spawning biomass 
of pelagic fishes (Lasker et ah, 1985; 
Parker, 1985). In the DEPM, biomass is 
calculated from the number of staged 
eggs taken in plankton samples and 
the daily fecundity of the parents. In 
a standai'd DEPM estimate, eggs are 
sampled in vertical net tows, starting 
from a point below the maximum depth 
of the eggs (typically 70 m) and end- 
ing at the surface. These vertical sam- 
ples are taken within a grid of sta- 
tions located 4 nmi apart, a sampling 
interval known to produce uncorrelat- 
ed samples of anchovy eggs (Smith and 
Hewitt, 1985). Use of the CUFES in the 
DEPM has several potential advantag- 
es over the use of a standard fixed-grid 
survey. Continuous sampling may in- 
crease the precision of the biomass es- 
timate because it provides an increased 
spatial resolution of egg patches (Hunt- 
er and Lo, 1997). Continuous sampling 
may save ship time and thereby reduce 
the cost of sampling per transect mile, 
and the increased spatial resolution of 
egg patches provides new knowledge 
regarding the spawning behavior of the 
species. Several potential disadvantag- 
es of using a CUFES in the DEPM 
also exist; the gains in precision pro- 
vided by continuous sampling may be 
diminished because of the increases in 
variance due to converting numbers of 
eggs taken in a CUFES to a full-wa- 
ter-column abundance; a much more 
complicated formula for variance of 
the estimate may be needed because 
of correlated CUFES samples; and the 
numbers of staged eggs taken in a 
CUFES may be biased, either because 
a CUFES damages eggs, making the 
staging of them more subject to error 
or because all stages may not be equal- 
ly vulnerable to sampling at the 3-m 
depth. Clearly these issues need to be 
resolved if a CUFES is to be used in a 
DEPM survey. 
We used the following approach to 
evaluate the use of a CUFES in the 
estimation of daily egg production. We 
conducted a pilot CUFES survey in 
1996 with the objective of examining 
the spatial properties of Pacific sardine 
(Sardinops sagax ) eggs sampled by a 
CUFES. We determined how the num- 
bers of eggs taken in a CUFES were 
related to their abundance in the full- 
water-column, the spatial correlation of 
egg samples, and the condition of eggs 
after passing through a CUFES. Our 
