Multiple timescale processes drive ecosystem metabolism in eelgrass (Zostera marina) meadows

Author Posting. © Inter-Research, 2014. This article is posted here by permission of Inter-Research for personal use, not for redistribution. The definitive version was published in Marine Ecology Progress Series 507 (2014): 1-13, doi:10.3354/meps10843.The oxygen flux between benthic ecosystems and the overlying water column is a measure of metabolic status and a commonly used proxy for carbon cycling. In this study, oxygen flux was measured seasonally using the eddy correlation technique in a restored eelgrass (Zostera marina L.) meadow in the Virginia coastal bays (USA). In 5 intensive field campaigns, we covered seasonal variation in oxygen metabolism and biomass with overlap in late summer to observe interannual variability. The high-resolution measurements allowed identification of the drivers of metabolism at multiple timescales: minute to hourly, daily, and monthly to seasonally. There was a strong correlation between nighttime hourly fluxes and current velocity that varied seasonally with seagrass shoot density and temperature. No similar relationship was observed during the day. A hysteresis effect in oxygen flux throughout the day was observed during October and August that was most likely due to increased respiration (R) in the afternoon. In October, net community production was 90% lower in the afternoon than in the morning at the same irradiance. From this hysteresis, we calculated that daytime R may be up to 2.5-fold larger than nighttime R. The magnitudes of daily gross primary production (GPP) and R were well correlated throughout the year with close to a 1:1 ratio that reflected a tight coupling between GPP and R on daily to seasonal timescales. Our results document the dynamic nature of oxygen fluxes that, when integrated over time, translate into highly variable rates of ecosystem metabolism over daily to seasonal timescales. This variation must be incorporated to accurately determine trophic status.Support for this study was provided by the University of Virginia and the National Science Foundation through grants from the Chemical Oceanography program (OCE-0536431 and OCE- 1061364) and the Division of Environmental Biology to the Virginia Coast Reserve Long Term Ecological Research Program (DEB-0621014 and DEB-1237733)