38 
Fishery Bulletin 107(1 ) 
Table 1 
Numbers of gray snapper (Lutjanus griseus) collected by each of the fishery-independent gears used in the eight major estuaries 
along the Texas coast. The increase in numbers of gray snapper from before to after the 1993 time period is presented as an 
x-fold increase. 
Estuary 
Bag seine 
Otter trawl 
Gill net 
Before 
1993 
After 
1993 
x-fold 
increase 
Before 
1993 
After 
1993 
x-fold 
increase 
Before 
1993 
After 
1993 
x-fold 
increase 
Sabine-Neches 
0 
6 
— 
0 
8 
— 
0 
4 
— 
Trinity-San Jacinto 
4 
20 
5.0 
0 
5 
— 
3 
52 
7.3 
Lavaca-Colorado 
17 
72 
4.2 
1 
6 
6.0 
14 
232 
16.6 
Guadalupe 
26 
71 
2.7 
1 
10 
10.0 
8 
163 
20.4 
Mission- Aransas 
7 
73 
10.4 
3 
22 
7.3 
8 
103 
12.9 
Nueces 
13 
35 
2.7 
0 
1 
— 
233 
926 
3.9 
Upper Laguna Madre 
2 
6 
3.0 
0 
2 
— 
6 
46 
7.7 
Lower Laguna Madre 
4 
34 
8.5 
22 
159 
7.2 
64 
461 
7.2 
Total 
73 
317 
4.3 
27 
213 
7.9 
336 
1987 
5.9 
collected each month in each bay, except Sabine Lake 
(Sabine Lake sampling began in 1986). From August 
1984 through January 1992, sampling frequency within 
each estuary increased to 12 samples per month, then 
to 16 per month, and finally to 20 samples each month. 
Since 1992, a total of 20 bag-seine samples have been 
collected from randomly selected locations within each 
bay in each month, resulting in 1920 samples per year. 
Otter trawls (6.1 m wide and with 3.8-cm stretched nylon 
multifilament mesh) were pulled for 10 minutes in water 
>1.0 m depth. Twenty trawl samples were taken from ran- 
domly selected sites in each bay each month (except for 
10 per month in Nueces and the upper Laguna Madre). A 
total of 1920 trawl samples were collected each year. Gill 
nets (monofilament, 183 m long, 1.2 m deep, and equipped 
with separate 45.7-m sections of 7.6-, 10. 2-, 12. 7-, and 
15.2-cm stretched mesh tied together in a sequence of 
ascending mesh size) were set overnight during each 
spring and fall season. The spring season begins with 
the second full week in April and extends for 10 weeks; 
the fall season begins with the second full week in Sep- 
tember and also extends for 10 weeks. Gill nets were set 
perpendicular to the shore with the smallest mesh facing 
shoreward. Nets were set within 1 hour before sunset 
and retrieved within 4 hours after the following sunrise. 
In each bay, a total of 90 gill nets were set at randomly 
selected sites (720 samples per year). 
Total lengths (TL) of gray snapper were measured 
for each gear type. Catch rates (relative abundance 
expressed as catch per unit of effort) from each gear 
were calculated for each bay, month, season, or year. 
Environmental measurements of surface salinity, water 
temperature (°C), and dissolved oxygen (mg/L and per- 
cent saturation) were measured both before the set and 
retrieval of gear for each gillnet collection and before 
each bag-seine collection. Bottom salinity, water tem- 
perature, and dissolved oxygen were measured before 
each trawl sample. 
Analytical techniques 
In order to directly compare the time series of abun- 
dance patterns of gray snapper across the disparate 
gears, relative abundance estimates for each gear were 
first Z-transformed (Snedecor and Cochran, 1980). The 
temperature records from each estuary were aggre- 
gated by month and year and expressed as monthly 
arithmetic averages to produce a time series compatible 
with the monthly abundance series. The temperature 
records were then analyzed with an empirical orthogonal 
function (EOF), a statistical tool used to decompose a 
spatially multivariate data set into its principal compo- 
nents. With this tool, the bulk of the variance can be 
described by a few orthogonal modes, so that the major 
properties of the data set can be more easily understood 
(Keiner and Yan, 1997). Principal modes of the spatial 
EOFs were seasonally detrended (multiplicative seasonal 
adjustment; SYSTAT, vers. 12, SYSTAT Software Inc., 
San Jose, CA) to reveal either periodicity or trends in the 
data. Detrended EOF slopes were assessed with linear 
regressions to test any departure from zero. 
Results 
From 1978 through 2006, a total of 2953 gray snap- 
pers were sampled from Texas bays and estuaries. The 
seasonal patterns revealed with the fishery-independent 
gears agreed well with published reports of spawning and 
movement patterns. Most juvenile and subadults taken 
in the bag seines and otter trawls were collected in the 
late summer to early fall, from August to October. Adult 
gray snapper were collected with gill nets in every month 
when this gear was deployed and were most prevalent 
during the fall from September to November. The vast 
majority of individuals were collected with gill nets from 
the middle Texas coast (from Lavaca-Colorado Estuary 
