Sanchez-Rubio and Perry: Meteorological and hydrological regimes and their influence on recruitment of Brevoortia patronus 
395 
Rubio et al., 2011a) was reclassified as an AMO cold 
year. 
The long-term phases of NAO were identified accord- 
ing to the proportion of total number of seasonal values 
(winter: January-March; spring: April-June; summer: 
July-September; and fall: October-December) that 
were negative and positive within a long-term NAO 
phase. A long-term period with a higher proportion 
of negative seasonal values was classified as a NAO 
negative phase, and a long-term period with a higher 
proportion of positive values was identified as a NAO 
positive phase. If the NAO negative and positive phas- 
es also had negative and positive averaged values, the 
above classification of phases of NAO was confirmed. 
The NAO positive (1959, 1995, 1996) and NAO nega- 
tive (1971, 1972) years classified by Sanchez-Rubio et 
al. (2011a) were reclassified as NAO negative and NAO 
positive years, respectively. 
The prior classification of ENSO years by Sanchez- 
Rubio et al. (2011a) was also revised with the new 
available monthly values of ENSO. To classify a year’s 
event, 3 or more monthly values from May through 
February were averaged. As a result, the previously 
classified ENSO cold years in 1971 and 1995 were re- 
classified as ENSO neutral years. 
Meteorological and hydrological data for coastal divisions 
Data sets of monthly precipitation, PDSI, and air tem- 
perature were obtained by climate division, the spa- 
tial scale by which the NOAA National Climate Data 
Center divides data from stations within a state. Data 
sets were acquired for 17 climate divisions along coast- 
al areas from Texas to Florida: Texas (divisions TX6- 
10), Louisiana (divisions LA5-9), Mississippi (division 
MS10), Alabama (division AL8), and Florida (divisions 
FL1-5). Monthly precipitation (in inches to the hun- 
dredths place) and PDSI values were averaged for the 
water year (defined as September of the current year 
to August of the subsequent year) from 1899 through 
2011. Annual precipitation values were converted into 
millimeters. Monthly (November-March) air tempera- 
ture values were averaged across the period 1899-2011. 
To compare hydrological conditions (precipitation 
and PDSI) among climate divisions, single linkage 
(nearest neighbor) agglomerative clustering based on 
Pearson correlation coefficients was carried out with 
SPSS Statistics 5 software, vers. 20.0 (IBM Corp., Ar- 
monk, NY). Three clusters of precipitation and PDSI 
climate divisions from the eastern (FL2-5), central 
(LA5-9, MS10, AL8, FL1), and western (TX6-10) re- 
gions were found along the Gulf Coast. A Mann-Whit- 
ney test was used to test differences among correlation 
coefficients in groups derived from the cluster analysis. 
When the null hypothesis of no difference was rejected, 
annual values of precipitation, PDSI, and air tempera- 
5 Mention of trade names or commercial companies is for iden- 
tification purposes only and does not imply endorsement by 
the National Marine Fisheries Service, NOAA. 
ture from grouped climate divisions were averaged to 
obtain data sets for each variable by region. 
For this study, annual environmental data sets were 
restricted to the western and central regions (from 
Texas to Alabama) where data were collected for men- 
haden. Annual regional anomalies were calculated by 
subtracting the average value by year from the annual 
value of wind momentum (north-south and east-west 
directions) and air temperature. Annual values of air 
temperature (November-March) and sea level (Novem- 
ber-May) were taken for the months that corresponded 
to the recruitment of Gulf menhaden to inshore nurs- 
ery habitats and to their early juvenile development. 
Annual values of precipitation, PDSI, and river flow 
were calculated for the period September-August, the 
months of early inshore development. Annual regional 
anomalies were calculated by subtracting the average 
value by year from the annual value of river flows. 
Meteorological and hydrological data: offshore waters 
Hourly SST, wind speed, and wind direction were ob- 
tained from offshore monitoring stations (platform 
42001, monitored since 1975, and platform 42002, 
monitored since 1973) maintained by the NOAA Na- 
tional Data Buoy Center (available at website). Hourly 
values of SST were transformed to daily and monthly 
values by using averaged values for each of the buoys. 
To increase the number of available years, we averaged 
highly correlated (coefficient of correlation [r]=0.946, 
P<0.001) monthly values of SST from buoys at both 
stations. Because of the potential influence of SST on 
the development of Gulf menhaden larvae, annual av- 
erage values of SST were calculated for the spawning 
season of Gulf menhaden (October of the previous year 
to March of the following year). 
For each of the stations, hourly values of wind direc- 
tion were used to categorize winds as easterlies, west- 
erlies, northerlies, or southerlies. Wind data from each 
direction were treated separately. Hourly wind speed 
data were averaged and transformed to daily and 
monthly values. Hours of recorded winds from each di- 
rection were added to obtain monthly values. Because 
the sampling hours were different among months, the 
monthly hours for each direction of the wind were di- 
vided by the total monthly hours when wind direction 
was sampled. Monthly values of wind speed and direc- 
tion were correlated (r>0.726, P<0.001), and data from 
both stations were averaged. These data were used to 
calculate wind stress (T), measured as newtons per 
square meter: 
T = p x CD x Dio 2 , (1) 
where p (the density of air) = 1.225 kg/m 3 ; 
Dio (wind speed in m/sec) = wind speed at 10 m 
above the water surface; and 
CD = the drag coefficient for which Smith (1980) 
proposed a formula to calculate CD: 
1000 CD = 0.44 + (0.063 x Di 0 ). (2) 
