Dataset: Urchin Respiration Rates

Environmental context and study sites:
Oceanic conditions of the Galápagos archipelago are highly spatially and temporally variable due to a complex ocean current regime and the El Niño-Southern Oscillation cycle (ENSO) (Houvenaghel 1984; Ruttenberg 2001; Wellington et al. 2001). The convergence of a number of ocean currents (Panama current, Perú current and Cromwell or Equatorial Undercurrent) results in variation (14-29°C) of the sea surface temperature among islands and between seasons (Wellington et al. 2001). Both temperature and upwelling intensity vary across the Archipelago: high upwelling and nutrient-rich zones are usually located in the colder western section of the Archipelago, and low-upwelling zones in the warmer, northern sites. Because of this environmental variance and oceanographic conditions, the Galápagos is divided into five distinct bioregions, where the assemblages of fish and macroinvertebrate species vary (Harris 1969; Wellington 1984; Jennings et al. 1994; Edgar et al. 2004). There is also a strong seasonality (resulting from the migration of the Intertropical Convergence Zone (Houvenaghel 1978; Wellington et al. 2001)) with a warm and rainy season from December to May, and a cooler, dry season from June to November. The maximum average sea surface temperature typically occurs in February/March and the minimum in September/October (Houvenaghel 1978; Schaeffer et al. 2008).

We performed the urchin physiology experiments in August 2018 at four different sites), accessed via the RV Queen Mabel. We recorded the temperature at each site by deploying one temperature logger (HOBO Water Temperature Pro v2 Data Logger- U22 001, Onset corporation, USA) during a previous research cruise in March 2018. Temperature was recorded at each site every 30 min at 7-12 m depth from March to August 2018. Punta Espinosa, located in the northeastern point of Fernandina Island in the western bioregion of the Archipelago, is within a major upwelling zone (Houvenaghel 1978; Schaeffer et al. 2008). La Botella and Punta Cormorant are located in the western and central-northern sides of Floreana, respectively, a southern island in the central-southeastern bioregion. Bartolomé is located in the south-eastern side of Santiago Island, in the central bioregion (Edgar et al. 2004). Punta Espinosa and La Botella are located in high-upwelling zones (Houvenagel 1978; Witman et al. 2010); while Bartolomé and Punta Cormorant are located in low upwelling zones (Houvenagel 1978).

Study species:
Sea urchins are important herbivores in many nearshore benthic marine habitats, often limiting algal biomass and thereby affecting community structure and function (Chapman and Johnson 1990; Andrew 1993; Steneck et al. 2002; Siddon and Witman 2003; Graham 2004; Irving and Witman 2009; Somero 2010). The pencil sea urchin (E. galapagensis) is the most abundant echinoid species of the shallow waters of the Galápagos archipelago (Lessios et al. 1999; Brandt and Guarderas 2002; Lawrence and Sonnenholzner 2004; Alvarado and Solís-Marín 2013). This species is one of the most significant mesograzers and bioeroders in the system (Brandt and Guarderas 2002; Irving and Witman 2009; Brandt et al. 2012; Feingold and Glynn 2014; Manzello et al. 2014; Glynn et al. 2017). Its densities across the Galápagos Archipelago average 3.2 ind·m-2 (Brandt and Guarderas 2002), however some sites have densities up to 28 ind·m-2 (Lawrence and Sonnenholzner 2004; Alvarado and Solís-Marín 2013). At these high densities, E. galapagensis can convert macroalgal assemblages to urchin barrens or pavements of encrusting coralline algae (Ruttenberg 2001; Edgar et al. 2010) and reduce hermatypic coral cover (Glynn 1988). Notably, after the 1982-1983 El Niño event that devastated coral reefs around the Galápagos archipelago (Glynn 1984, 1990; Feingold and Glynn 2014; Glynn et al. 2017), densities of E. galapagensis increased six-fold and led to some of the highest reported bioerosion rates in the world (Glynn 1988). Therefore, any changes in E. galapagensis behavior, grazing rates, physiology or abundance could have a significant impact on the ecosystem functioning of Galápagos rocky and coral reefs (Steneck et al. 2002; Siddon and Witman 2003; Graham 2004).

Using SCUBA at rocky reefs of depths of 7-12 m, eight individuals of E. galapagensis were hand-collected from each of the four sites during the August 2018 cruise aboard the RV Queen Mabel. Selected sites displayed average urchin densities ranging from 2.5 to 5.0 ind·m-2 (Brandt and Guarderas 2002). After collections, urchins were allowed to stabilize in a bucket on the ship with seawater and an aerator at ambient temperature for 30 min. Sea surface temperature was recorded for each collection site using a calibrated digital thermometer (Traceable High Accuracy ±0.2°C Digital Thermometer S/N 170718701).

Thermal response measurements:
The thermal sensitivity of each urchin (n=8 per site) was measured in a closed system of ten 620 ml acrylic respiration chambers with magnetic stir bars. In this respirometry setup, there were eight replicate chambers that contained sea urchins and two chambers with only seawater as controls. Oxygen consumption and temperature were monitored in each individual chamber with a fiber-optic oxygen probe (Presens dipping probes [DP-PSt7-10-L2.5-ST10-YOP], Germany) and a temperature probe (Pt1000), respectively. Measurements were taken using a Presens Oxygen Meter System (OXY-10 SMA (G2) Regensburg, Germany) with temperature correction made for each probe independently. Oxygen concentration in the urchins and control chambers was measured every 1 s during trials, that lasted 6 to 10 minutes for a given temperature. Temperature was controlled [±0.2°C] using a thermostat system (Apex Aquacontroller, Neptune Systems), bucket heaters (King Work Bucket Heater 05-742G 1000W), and a chiller (AquaEuroUSA Max Chill-1/13 HP). At each site, the initial (and lowest) temperature was the local ambient. After each trial, the temperature was increased by 1-3°C, depending on the temperature. We decreased the range between treatment temperatures around the expected respiration peak (based on pilot data) because increased resolution improves curve fitting. We used the following temperatures (in °C) for urchins tested from each of the four sites: Punta Espinosa (19, 23, 26, 28, 30, 31, 32, 33, 34, 36, 38, 42), La Botella (20, 23, 26, 28, 30, 31, 32, 33, 34, 36, 38, 42), Punta Cormorant (22, 26, 28, 30, 31, 32, 33, 34, 36, 38, 42) and Bartolomé (23, 26, 28, 30, 31, 32, 33, 34, 36, 38, 41). It took 10 to 20 minutes to warm the water bath between treatment levels (temperature ramping rates did not differ between sites). Once stabilized at the new temperature treatment level, the water inside the chambers was replaced with new seawater to ensure that it matched the temperature of the water bath, and to reset O2 and CO2 levels. The water volume in each chamber was measured indirectly by measuring urchin volume (as the volume of water displaced from a graduated cylinder) and subtracting that from the known chamber volumes. After all measurements had been made, urchins were frozen on the ship and brought to the Marine Ecology Laboratory of the Galápagos Science Center (GSC) on San Cristóbal Island. Respiration rates were normalized to urchin Ash-Free Dry Weight, which was determined by first drying each sample in a drying oven for 24 hrs at 60°C and then burning it in a muffle furnace (Optic Ivymen System Laboratory Furnace 8.2/1100) for 4 hrs at 500°C.

TPC Characterization:
TPCs were used to characterize the relationship between urchin organic-biomass-normalized respiration rates and temperature for every individual. A TPC approach is a widely used model in climate change research to predict if organisms will be able to cope with increasing environment temperatures (Schulte et al. 2011; Vasseur et al. 2014) and to compare performance metrics across organisms, populations, species, localities and time (Sinclair et al. 2016; Silbiger et al. 2019). Acute TPCs were modelled with a modified Sharpe-Schoolfield equation for high temperature inactivation (Schoolfield et al. 1981; Padfield et al. 2017), using a non-linear least squares regression (Elzhov et al. 2013; Padfield et al. 2016) in the nls.multsart R package (Padfield and Matheson 2018).