Marancik et al.: Spatial and temporal distribution of grouper larvae in the Gulf of Mexico and Straits of Florida 
13 
West Florida shelf— Fall 
Year 
27.0 27.5 28.0 28.5 29.0 29.5 
Surface temperature (°C) 
Surface salinity 
Mississippi-Alabama- 
Florida shelf— Fall 
Texas-Louisiana shelf— Fall 
Year Surface temperature (°C) 
Figure 5 
Partial effects [s(covariate); y-axis] of the generalized additive model (GAM) for 
each covariate (x-axis) used in the best subregion by season model of percent 
frequency of occurrence (%FO) of grouper larvae (all taxa combined) collected 
during Southeast Area Monitoring and Assessment Program fall sampling (Sep- 
tember-October, 1986-2005) in the three subregions analyzed. Modeled partial 
effects (solid lines) and 95% confidence intervals (dotted lines) are shown for 
each included covariate. The dark dashed line marks no effect. Percent frequency 
of occurrence was higher for partial effects greater than 0, and lower for values 
less than 0. Vertical dashes along the x-axis denote actual sample values for the 
covariate. ****P<0.001, ***P<0.01, **P< 0.05. 
decades (7 of the 10 years between 1996 and 2005) 
and are often collected in multiple months within a 
year. Graysby larvae have also become a significant 
percentage of the total catch of grouper larvae col- 
lected in the recent decade: 3-33% (mean = 18.7%) 
before 1995; 5—100% (mean = 40.3%) after 1995. A 
comparison of survey coverage for the two periods 
(1986-95, 1996-2005) revealed comparable sampling 
effort during spring and fall surveys. Compared with 
percent coverage during the period since 1995, the 
percent coverage during fall surveys before 1995 
was similar or slightly higher, whereas in spring 
the percentages were similar or slightly lower before 
1995. Therefore, differences in sampling effort did not 
