| Summary: | We investigated coral reef carbonate chemistry dynamics and metabolic rates using an automated system that measured total alkalinity (TA, 30 min intervals), pH on the total scale (pHT, 10 min intervals) and the partial pressure of carbon dioxide (pCO2, 1 min intervals) over 2 weeks at Heron Island (Great Barrier Reef, Australia). The calculation of pHT (using the pCO2 and TA pair) and pCO2 (using the pH and TA pair) had similar values to the measured pHT and pCO2 values. In contrast, calculated TA from the pCO2-pH pair showed a large discrepancy with measured TA (average difference between measured and calculated TA = 52 μmol kg−1). High frequency sampling allowed for detailed analysis of the observations and an assessment of optimum sampling intervals required to characterise the net ecosystem calcification (NEC) and production (NEP) using a slack water approach. Depending on the sampling interval (30 min–2 h time steps) used for calculations, the estimated daily NEC and NEP could differ by 12% and 30%, respectively. Abrupt changes in both NEC and NEP were observed at dawn and dusk, with positive NEC during these periods despite negative NEP. Integrating NEC and NEP over a full diel cycle using 1 or 2 h integration time steps resulted in small differences of 2–7% for NEC and 1–3% for NEP. A diel hysteresis pattern rather than a simple linear relationship was observed between the aragonite saturation state (Ωar) and NEC. The observed hysteresis supports recent studies suggesting that short-term observations of seawater Ωar may not be a good predictor of long-term changes in NEC due to ocean acidification. The slope of the DIC to TA relationship was slightly higher (0.33) in 2014 than in an earlier study in 2012 (0.30). The automated, high frequency sampling approach employed here can deliver high precision data and can be used at other coral reef research stations to reveal long-term changes in NEC and NEP potentially driven by ocean acidification, eutrophication or other local changes.
|
|---|