Quantitative studies of coral reefs are challenged by the three-dimensional hard structure of reefs and the high spatial variability and temporal dynamics of their metabolism. We used the non-invasive eddy correlation technique to examine respiration and photosynthesis rates, through O2 fluxes, from reef crests and reef slopes in the Florida Keys, USA. We assessed how the photosynthesis and respiration of different reef habitats is controlled by light and hydrodynamics. Numerous fluxes (over a 0.25 h period) were as high as 4500 mmol O2 m−2 d−1, which can only be explained by efficient light utilization by the phototrophic community and the complex canopy structure of the reef, having a many-fold larger surface area than its horizontal projection. Over diel cycles, the reef crest was net autotrophic, whereas on the reef slope oxygen production and respiration were balanced. The autotrophic nature of the shallow reef crests implies that the export of organics is an important source of primary production for the larger area. Net oxygen production on the reef crest was proportional to the light intensity, up to 1750 µmol photons m−2 s−1 and decreased thereafter as respiration was stimulated by high current velocities coincident with peak light levels. Nighttime respiration rates were also stimulated by the current velocity, through enhanced ventilation of the porous framework of the reef. Respiration rates were the highest directly after sunset, and then decreased during the night suggesting that highly labile photosynthates produced during the day fueled early-night respiration. The reef framework was also important to the acquisition of nutrients as the ambient nitrogen stock in the water had sufficient capacity to support these high production rates across the entire reef width. These direct measurements of complex reefs systems yielded high metabolic rates and dynamics that can only be determined through in situ, high temporal resolution measurements.

Based on noninvasive eddy correlation measurements at a marine and a freshwater site, this study documents the control that current flow and light have on sediment-water oxygen fluxes in permeable sediments. The marine sediment was exposed to tidal-driven current and light, and the oxygen flux varied from night to day between 229 and 78 mmol m 22 d 21 . A fitting model, assuming a linear increase in oxygen respiration with current flow, and a photosynthesis-irradiance curve for light-controlled production reproduced measured fluxes well (R 2 5 0.992) and revealed a 4-fold increase in oxygen uptake when current velocity increased from , 0 to 20 cm s 21 . Application of the model to a week-long measured record of current velocity and light showed that net ecosystem metabolism varied substantially among days, between 227 and 31 mmol m 22 d 21 , due to variations in light and current flow. This variation is likely typical of many shallow-water systems and highlights the need for long-term flux integrations to determine system metabolism accurately. At the freshwater river site, the sediment-water oxygen flux ranged from 2360 to 137 mmol m 22 d 21 . A direct comparison during nighttime with concurrent benthic chamber incubations revealed a 4.1 times larger eddy flux than that obtained with chambers. The current velocity during this comparison was 31 cm s 21 , and the large discrepancy was likely caused by poor imitation by the chambers of the natural pore-water flushing at this high current velocity. These results emphasize the need for more noninvasive oxygen flux measurements in permeable sediments to accurately assess their role in local and global carbon budgets.

Accurate light measurements are important in the analysis of photosynthetic systems. Many commercial instruments are available to determine light; however, the comparison of light estimates between studies is difficult due to the differences in sensor types and their calibrations. The measurement of underwater irradiance is also complicated by the scattering and attenuation of light due to interactions with particulates, molecules, and the bottom. Here, three sensor types are compared to evaluate the calibration of light intensity loggers to estimate photosynthetically active radiation (PAR). We present a simple calibration of light intensity loggers that agree within 3.8% to factory-calibrated scalar PAR sensors under a wide range of environmental conditions. Under the same range of conditions, two identical factory-calibrated PAR sensors showed a similar difference of 3.7%. The light intensity loggers were calibrated to a high-quality PAR sensor using an exponential fit (r 2 = 0.983) that accounts for differences in sensor types with respect to the angle of incoming light, scattering, and attenuation. The light loggers are small, robust, and simple to operate and install, and thus well-suited for typical subsurface research. They are also useful for small-scale measurements, when broad spatial coverage is needed, or in research requiring multiple sensors. Many studies have used these simple light intensity sensors to estimate PAR, yet their limitations and advantages in mimicking PAR have not been well defined previously. We present these small and user-friendly loggers as an excellent alternative to more sophisticated scalar PAR sensors.

The metabolism of seagrass ecosystems was examined at 4 sites in south Florida, USA, using the eddy covariance technique under in situ conditions. Three sites were located across a phosphorus-driven productivity gradient to examine the combined effects of dynamic variables (irradiance, flow velocity) and state variables (sediment phosphorus and organic content, seagrass biomass) on ecosystem metabolism and trophic status. Gross primary production and respiration rates varied significantly across Florida Bay in the summer of 2012 with the lowest rates (64 and −53 mmol O 2 m −2 d −1 , respectively) in low-phosphorus sediments in the northeast and the highest (287 and −212 mmol O 2 m −2 d −1 , respectively) in the southwest where sediment phosphorus, organic matter, and seagrass biomass are higher. Seagrass eco systems offshore of the Florida Keys had similar large daily production and respiration rates (397 and −217 mmol O 2 m −2 d −1 , respectively) and were influenced by flow through the permeable offshore sediments. Across all sites, net ecosystem metabolism rates indicated that the seagrass ecosystems were autotrophic in the summertime. Substantial day-today variability in metabolic rates was found due to variations in irradiance and flow velocity. At all sites the relationship between photosynthesis and irradiance was linear and did not show any sign of saturation over the entire irradiance range (up to 1400 µmol photons m −2 s −1). This was likely due to the efficient use of light by the large photosynthetic surface area of the seagrass canopy, an effect which can only be examined by in situ measurements that integrate across all autotrophs in the seagrass ecosystem.