This dataset is the average daily temperature from 2-3 years of data for each location. For example, January, Calendar Day 1, Julian Day 1 is the average temperature for all recordings from Jan. 1 averaged over years 2005, 2006, and 2007.

To determine the latitudinal pattern of sex determination, embryos from M. menidia were collected from 31 locations along the species’ range from Amelia Island, FL (30° 31.13’ N, 081° 27.78’ W) to the Magdalen Islands, Quebec, Canada (47° 23.60’ N, 061° 50.76’ W) (Table 1, See Hice et al. (2012) for detailed methods). This resulted in collection sites that averaged 60-80 km from each other for complete coverage of M. menidia’s range (Figure 1). Briefly, spawning adult M. menidia were collected with a 30m seine net, stripped spawned following the methods employed by Lagomarsino and Conover (1993) and embryos were transported to Flax Pond Laboratory in Old Field, NY or the Bluepoint Laboratory in West Sayville, NY. Larvae of mixed parentage were pooled and randomly allocated into 15 °C or 28 °C in triplicate containers prior to sex determination (Conover and Fleisher 1986). Density per container was periodically adjusted and fish were euthanized when they reached a minimum of 21-23 mm TL. Time to reach this size ranged from one to six months, depending on growth rate and temperature treatment.

Sex was determined macroscopically with a dissecting microscope and total lengths were recorded. Females were identified by opaque or lightly pigmented fragile ovaries, as compared to thin, translucent and flexible testes in males (Conover and Fleisher 1986). In some replicates (27%), a few individuals did not reach a size at which sex could be definitively assessed, but this accounted for a small proportion (1.2%) of all fish. An average of 228 + 36 fish were sexed from each location.

Sea surface temperature (SST) data was collected from the NOAA National Buoy Data Center archives for estuarine or nearshore sites within the U.S.A. and from the Oceanographic Database maintained by the Dept. of Fisheries and Oceans, Canada. Data from two to three years between 2004 and 2007 were used to calculate the average SST (to reduce noise of an anomalous year), as well as mean values for the absolute range (maximum summer temperature - minimum winter temperature), summer temperature (July through September), and 30 day winter minimum temperature (coldest 30 day period). Season length was calculated as the mean number of days that SST within the estuary was > 12 ^oC, as somatic growth is approximately zero below this temperature (Conover and Heins 1987a). Data for SST was available for 19 of the 31 sites.

Level of TSD was calculated as the difference in sex ratios of juvenile M. menidia reared at feminizing (15 °C) and masculinizing (28°C) temperatures (defined as, [F/(F+M) at 15 °C]-[F/(F+M) at 28 °C], Conover and Heins 1987a). Sex ratios were calculated by weighting the sex ratio from each replicate by the number of individuals. Measurements of TSD were repeated during an additional year for two populations that experienced high mortality during the common garden rearing (Norfolk, VA and Patchogue, NY). The repeated calculations of TSD were used in the data presented here. Piecewise linear regression was applied to the latitudinal distribution of TSD to determine breakpoints in the form of sex determination. The validity of five models (1 to 5 segments) was assessed using the methods in Hice et al. (2012). Confidence limits (95th percentile) were calculated using Sigmaplot (version 10.1, Systat Software, San Jose, CA, USA).

Related Reference:

Lyndie A. Hice,Tara A. Duffy, Stephan B. Munch and David O. Conover (2014) Spatial scale and divergent patterns of variation in adapted traits in the ocean. Ecology Letters 15: 568-575 doi: 10.1111/j.1461-0248.2012.01769.x

Related Datasets:

Menidia sex ratios - 2 temps
Temperature-dependent sex ratio of Menidia and SST data

mean_daily_sst_flat_site.csv