Intraspecific variation in the organismal stoichiometry of the Least Killifish tracks spatial variation in periphyton composition

Ecological stoichiometry, the mass-balance relationships in elemental composition among species in an ecosystem, is fundamental to understanding a wide variety of ecological phenomena, from foraging patterns to nutrient cycling (Sterner and Elser 2002). Variation in organismal stoichiometry, be it interspecific or intraspecific, can drive substantial variation in community and ecosystem dynamics (El Sabaawi et al. 2015, Des Roches et al. 2017, Leal et al. 2017a). Here we report a survey of the organismal stoichiometry and trophic niche of H. formosa in eight populations, four from each of two contrasting habitat types. First, we describe variation in water chemistry among a larger set of 17 locations, from which we selected our eight populations for study. We also document variation among these eight locations in the elemental composition of periphyton in the littoral zone, where H. formosa forage. Second, we describe the trophic niche of H. formosa in those eight populations*,* establishing that the fish occupy a trophic position consistent with their being a primary consumer. Third, we examine the organismal stoichiometry in those populations. We show that there is less population variation than has been reported in other species, but the variation that we detected tracks variation in the carbon and nitrogen profiles of the periphyton on which these fish feed.

Methods
We sampled water at 17 locations with populations of H. formosa. These locations embraced a variety of habitats, including springs, wooded swamps, open canopy lakes, and freshwater marshes. We selected eight of these locations for stoichiometric and stable isotope analyses: four freshwater springs (McBride’s Slough, Newport Bridge, Shepherd’s Spring, Wacissa River) and four lakes (Harper’s Eyelet, Little Lake Jackson, Moore Lake, Trout Pond). We chose these eight locations for two reasons. First, they spanned a wide range of water chemistry parameters. Second, the populations of H. formosa in these locations displays a wide range of population densities and life histories (Schrader and Travis 2012, Macrae and Travis 2014). 

We collected fish via dip netting in the shallow littoral zone, typically at depths less than 0.30 m. We sacrificed fish via rapid chilling in an ice slurry, following approved IACUC protocols. We obtained periphyton samples by scraping stems and leaves of aquatic plants, larger rocks, and fallen limbs wherever we saw H. formosa individuals foraging. We placed all samples on ice for return to the laboratory.

Water Chemistry Methods

We collected water samples by lowering a Nalgene bottle on a long pole into clear water, approximately 1.5 meters from the shoreline and away from aquatic vegetation. We filtered the water on site by pouring through a 10 μl filter and immediately placed the filtered water on ice for transport the same day to Akuritlabs, Inc., in Tallahassee (National Environmental Laboratory Certification E81350) for analyses. We obtained data on pH, concentrations of chlorophyll a, total nitrogen, organic nitrogen, nitrite, nitrate, ionized and unionized ammonia, total phosphorus, and orthophosphate. For analyses, we used data only on pH, chlorophyll a, nitrate, and total nitrogen because only for these variables were concentrations above the detection limit at all locations.

Elemental composition and isotope methods

We prepared fish tissue and periphyton samples for elemental and isotope analyses following Aresco et al. (2015). In brief, upon return to the laboratory, we removed and discarded all internal organs, including reproductive tissue, gut, and liver/viscera, where lipids are stored for future use (McManus and Travis 1998), and freeze-dried the carcasses. We freeze-dried all periphyton samples in the same manner.  We ground all freeze-dried material to a fine powder using a WiglBug Model 3110B, rinsing the capsule and grinding bearing with ethanol between samples. For analyses of carbon and nitrogen, we prepared ground material in small tin capsules with, on average, 1 mg fish tissue and 2 mg periphyton. These samples were analyzed using the elemental analyzer and continuous flow isotope ratio mass spectrometer at the University of California, Davis, Stable Isotope Facility.

We analyzed total phosphorus concentration in fish tissue as concentration of 31P with the Thermo 2 inductively coupled plasma mass spectrometry system (ICP-MS) in the Geochemistry Lab at the National High Magnetic Field Laboratory in Tallahassee, Florida. The ICP-MS method is especially accurate for small amounts of sample material or samples with low concentrations of total phosphorus (Cooper et al. 2005; Wilschefski and Baxter 2019). We prepared samples by dissolving them in nitric acid, following protocols described in Wilschefski and Baxter (2019), using 1 ppm phosphorus and 2 ppb indium as internal standards.

Calculation of baseline δ15N values and H. formosa trophic position

We estimated the trophic position of individual H. formosa by correcting for the variation in δ15N values of the identifiable primary producers (trophic level 1; Vander Zanden et al., 1999). For baseline δ15N values, we used those estimated from the periphyton samples at each site. We used the equation

Trophic Position = [(δ^15^N-δ^15^Nbaseline)/3.4]+1,

where 3.4% is the mean enrichment of δ15N between trophic levels as determined in previous studies of fish that have the same range of longevity and tissue turnover rates as fish species in the sites we sampled (Vander Zanden et al., 1999; Aresco et al., 2015).