Closing the oxygen mass balance in shallow coastal ecosystems

Matthew H Long,David J Burdige,Jennie E Rheuban,Daniel C Mccorkle,Richard C Zimmerman

doi:10.1002/lno.11248

Matthew H Long, David J Burdige + Show 3 more

Open Access

https://doi.org/10.1002/lno.11248

Copy DOI

Abstract

AbstractThe oxygen concentration in marine ecosystems is influenced by production and consumption in the water column and fluxes across both the atmosphere–water and benthic–water boundaries. Each of these fluxes has the potential to be significant in shallow ecosystems due to high fluxes and low water volumes. This study evaluated the contributions of these three fluxes to the oxygen budget in two contrasting ecosystems, a Zostera marina (eelgrass) meadow in Virginia, U.S.A., and a coral reef in Bermuda. Benthic oxygen fluxes were evaluated by eddy covariance. Water column oxygen production and consumption were measured using an automated water incubation system. Atmosphere–water oxygen fluxes were estimated by parameterizations based on wind speed or turbulent kinetic energy dissipation rates. We observed significant contributions of both benthic fluxes and water column processes to the oxygen mass balance, despite the often‐assumed dominance of the benthic communities. Water column rates accounted for 45% and 58% of the total oxygen rate, and benthic fluxes accounted for 23% and 39% of the total oxygen rate in the shallow (~ 1.5 m) eelgrass meadow and deeper (~ 7.5 m) reef site, respectively. Atmosphere–water fluxes were a minor component at the deeper reef site (3%) but a major component at the shallow eelgrass meadow (32%), driven by diel changes in the sign and strength of atmosphere–water gradient. When summed, the measured benthic, atmosphere–water, and water column rates predicted, with 85–90% confidence, the observed time rate of change of oxygen in the water column and provided an accurate, high temporal resolution closure of the oxygen mass balance.

Highlights

IntroductionOnedimensional formulations based on O2 time-series measurements (i.e., the left-hand side of Eq 1) were pioneered by Odum (1956), and variations of this method are still commonly applied today (Caffrey et al 2014; Howarth et al 2014; Qin and Shen 2019; Tassone and Bukaveckas 2019)
Pnet indicated a net balance between autotrophy and heterotrophy (Pnet = 1.7 Æ 2.5 mmol m−2 d−1) at the Virginia eelgrass site in July
This is consistent with a shift from net autotrophy to net heterotrophy during July found by Rheuban et al (2014)

Summary

Introduction

Onedimensional formulations based on O2 time-series measurements (i.e., the left-hand side of Eq 1) were pioneered by Odum (1956), and variations of this method are still commonly applied today (Caffrey et al 2014; Howarth et al 2014; Qin and Shen 2019; Tassone and Bukaveckas 2019) These formulations are simple to implement and are scaled due to the prevalence of O2 and depth measurements but provide limited information regarding the different components of the system responsible for the observed flux (Kemp and Boynton 1980; Ziegler and Benner 1998; Gazeau et al 2005; Demars et al 2015) and can be biased by physical processes such as advection and mixing (Swaney et al 1999; Beck et al 2015). Each of these different methods has specific biases and ideal conditions for their use, which has been discussed elsewhere (e.g., Tengberg et al 1995; Berg et al 2003; Collins et al 2018; Long and Nicholson 2018)

Methods

Results

Discussion

Conclusion