Seagrass metabolism across a productivity gradient using the eddy covariance, Eulerian control volume, and biomass addition techniques

Abstract

The net ecosystem metabolism of the seagrass Thalassia testudinum was studied across a nutrient and productivity gradient in Florida Bay, Florida, using the Eulerian control volume, eddy covariance, and biomass addition techniques. In situ oxygen fluxes were determined by a triangular Eulerian control volume with sides 250 m long and by eddy covariance instrumentation at its center. The biomass addition technique evaluated the aboveground seagrass productivity through the net biomass added. The spatial and temporal resolutions, accuracies, and applicability of each method were compared. The eddy covariance technique better resolved the short-term flux rates and the productivity gradient across the bay, which was consistent with the long-term measurements from the biomass addition technique. The net primary production rates from the biomass addition technique, which were expected to show greater autotrophy due to the exclusion of sediment metabolism and belowground production, were 71, 53, and 30 mmol carbon m 22 d 21 at 3 sites across the bay. The net ecosystem metabolism was 35, 25, and 11 mmol oxygen m 22 d 21 from the eddy covariance technique and 10, 2103, and 14 mmol oxygen m 22 d 21 from the Eulerian control volume across the same sites, respectively. The low-flow conditions in the shallow bays allowed for periodic stratification and long residence times within the Eulerian control volume that likely reduced its precision. Overall, the eddy covariance technique had the highest temporal resolution while producing accurate long-term flux rates that surpassed the capabilities of the biomass addition and Eulerian control volume techniques in these shallow coastal bays.

Surface gas exchange determined from an aquatic eddy covariance floating platform

“…S3) and not subject to a flow‐dependent response bias as demonstrated experimentally by Holtappels et al (). However, waves remain an unresolved issue for eddy covariance measurements as contributions can be observed at wave frequencies while still producing biologically constrained and expected flux values (Long et al 2015 a ). Whether this is a bias or the advection of turbulence by wave velocities past a fixed measuring point (Lumley and Terray, ; Gerbi et al, ) requires further study.…”

“…Further, the flux at height may not represent the interface of interest (Holtappels et al ), for example, when the flux has components of exchange across both the air‐water and water‐sediment interface. A correction factor that accounts for changes in the mean concentration at height can account for mass balance variability (Rheuban et al ; Long et al 2015 a ) but must be applied carefully at the air‐water interface as both benthic and water column sources can differentially influence this correction. Despite these challenges, this new application of air‐water eddy covariance, when combined with existing complementary measurements of benthic and water column production, now provide a framework for the comprehensive analysis of the oxygen mass‐balance between aquatic systems and the atmosphere.…”

“…The ADV (Nortek firmware version V1.21) sampled continuously on high power with a sampling volume of 14.9 mm and a transmit length of 4.0 mm, but we note that these settings are condition and site specific. The ADV also provided power to, and logged the signal from, a FirestingO 2 Mini fiber‐optic O 2 meter with a fast‐response (< 0.3 s) 430 μm diameter optode (Pyroscience, DE) (Long et al 2015 a ). The optode was located 0.5 cm from the ADV measuring volume to limit interference with the ADV acoustic signals (Donis et al ; Berg et al ) with the appropriate corrections applied for the time‐lag between the sensors (e.g., McGinnis et al ).…”

“…Temperature loggers (Hobo Pendant) were attached at 0.1 m and 1 m water depths on the frame to detect stratification. With the exception of the IMU sensor and floating platform, the described eddy covariance system is very similar to previous instruments used to examine benthic O 2 exchange and the specific instrument and sensor designs, capabilities, and configurations can be found in Long et al (2015 a ).…”

“…The eddy covariance O 2 exchange across the air‐sea interface was then determined over 0.25 h intervals from the time‐averaged product of the instantaneous variations in the vertical velocity and the O 2 concentration using customized Matlab routines (Long et al 2015 a ). The instantaneous variations in the vertical velocity and O 2 concentration were determined by Reynold's Decomposition with the means determined from linear detrending over each 0.25 h interval.…”