Fig 1.
Location of the Heron Island reef flat research site.
(a) Map of Australia with a black rectangle representing the Capricorn Bunker group of reefs at the southernmost end of the Great Barrier Reef with a circle for Heron Island. (b) Aerial photograph of Heron Island with the reef flat study site shown as a star, the Heron Island Research Station aquarium system as a black rectangle and the channel cut in the reef flat as a white arrow.
Table 1.
Environmental exposure metrics calculations.
Fig 2.
High resolution plots of temperature, light and water depth.
(a) Box plots of temperature, (b) light and (c) water depth data with hourly resolution. The edges of the box are the mid-range (25–75th percentile), the lines extend to the most anomalous data points not considered outliers, and outliers are plotted individually. (d) Diel composites shown for each month for temperature, (e) light and (f) water depth.
Table 2.
Monthly environmental data summary statistics including the average diel means, average diel minimum, average diel maximum and diel range for each parameter (± SD).
Table 3.
Seasonal environmental statistics, with absolute minima and maxima observed.
Table 4.
Krustal-Wallis test results for seasonal differences in environmental parameters.
Fig 3.
High resolution salinity, rainfall, windspeed and current velocity data.
(a) Environmental data included salinity, (b) rainfall, (c) wind speed and (d) current velocity. The salinity and current velocity data were measured as part of this study while the rainfall and wind speed data are from the Australian Bureau of Meteorology (http://www.bom.gov.au/).
Fig 4.
Carbonate chemistry parameters measured (pH, nDIC, nTA) and calculated (pCO2 and ΩARG) throughout the study.
(a) Hourly resolution data for pH in total pH units, (b) salinity normalized dissolved inorganic carbon (nDIC) in μmol kg-1, (c) salinity normalized total alkalinity (nTA) in μmol kg-1, (d) the partial pressure of CO2 (pCO2) in μatm and (e) the aragonite saturation state (ΩARG). The edges of the box are the mid-range (25–75th percentile), the lines extend to the most anomalous data points not considered outliers, and outliers are plotted individually. (f) Diel composites of the data are shown for each month for pH, (g) nDIC, (h) nTA, (i) pCO2 and (j) ΩAR.
Fig 5.
Comparison of high-resolution carbonate chemistry data during two weeks in June and December 2010.
(a) The data for June 7–22, austral fall, is shown for water depth in m, (b) pH in total pH units, (c) salinity normalized dissolved inorganic carbon (nDIC) in μmol kg-1, (d) salinity normalized total alkalinity (nTA) in μmol kg-1, (e) the partial pressure of CO2 (pCO2) in μatm and (f) the aragonite saturation state (ΩARG). (g) Data for Nov. 29—Dec 13, austral spring, for water depth, (h) pH, (i) nDIC, (j) nTA, (k) pCO2, and (i) ΩAR. During these periods, discrete samples of DIC and TA were taken at least two times per day and in some cases every 3 hours.
Fig 6.
Contribution analysis to determine the role of salinity, temperature, total alkalinity and DIC in influencing different aspects of the carbonate chemistry.
(a) Contributions of salinity, temperature, total alkalinity and DIC to changes in pH, (b) aragonite saturation state (ΩAR) and (c) pCO2. The complete sampling suite #s are times throughout the December 2010 study period when all of the parameters (S, T, pH, DIC, TA) were measured. The sign of percentage change (positive vs. negative) keeps track of the overall directionality of change in the pH, pCO2, or ΩAR value.
Table 5.
Mean and standard error of intensity (I), duration (D), and severity (S) of pH, temperature and combined events.
Fig 7.
Exposure metrics for pH, temperature and an additive, combined metric for pH and temperature.
(a) Frequency distribution of the mean intensity events, in total pH units, for each of the months in this study was plotted, while the duration of the events in hours was plotted in (b) and the mean severity in total ph units•hours in (c). The pH threshold (Th) was defined as 8.1 to represent the 2010 mean global open ocean pH. A similar procedure was used for the temperature data, but the mean monthly maximum temperature thresholds for each month (June-December) were determined based on temperature monitoring data available from 2008-present (Australian Institute for Marine Science, http://data.aims.gov.au/aimsrtds/station.xhtml?station=130). (d) The relative frequency distributions of monthly mean intensity of the high temperature events in °C, (e) the duration of these events in hours and (f) the mean monthly severity of these high temperature events in °C•hours. (g) The composite intensity including the mean monthly scaled pH and temperature intensity values, (h) combined duration including the mean monthly scaled pH and temperature duration values and (i) the combined severity.
Table 6.
Comparison between studies showing seasonal patterns in temperature, salinity, and carbon system parameters.