Accounting for variation in temperature and oxygen availability when quantifying marine ecosystem metabolism

Given human modification of Earth’s ecosystems, it is essential to understand how these changes are influencing ecosystem functioning, including net and gross community production and community respiration. These responses are often estimated by measuring oxygen production in the light (net community production) and consumption in the dark (community respiration). These values can then be combined to estimate gross community production. However, the method used to create “dark” conditions – either experimental darkening during the day or conducting measurements at night – could result in different estimates of respiration and production. These data were used to test this prediction and include measurements of oxygen concentrations under daytime “ambient” light conditions, in darkened tide pools during the day, and during nighttime low tides. We made measurements every 1-3 months over one year in southeastern Alaska. We found that daytime respiration rates were substantially higher than those measured at night. These differences were associated with higher temperature and oxygen levels during the day and led to major differences in estimates of gross community production calculated using daytime versus nighttime respiration measurements. Our results highlight the need to measure respiration rates during both day and night to account for effects of temperature and oxygen.

Methods
During each sampling event, two complementary sets of data were collected to evaluate changes in O2 concentrations in the pools. The first consisted of daytime and nighttime (Day-Night) sampling, where pools were left uncovered and light varied naturally on a diel cycle. During one daytime and one nighttime low tide event, we took in-situ measurements and collected water samples. Initial readings of O2 (DSS optical DO meter, YSI, Inc., Yellow Springs, Ohio, USA), temperature (YSI DSS), and light (MQ 210 Underwater Quantum Meter, Apogee Instruments, Logan, Utah, USA) were taken in each pool as soon as the pool was isolated by the receding tide. After a minimum of 1 hr (median = 1.5 hr, max = 3.5 hr), pools were resampled, and a second set of O2, temperature, and light measurements were taken. Daytime measurements were taken on days when irradiance levels were high enough to saturate photosynthetic rates of the most widespread and abundant seaweed species in the tide pools, Neorhodomela oregona (> 47 umol photons m-2 s-1; M. Bracken, unpublished data). Nighttime samples were collected on nights when the pools were isolated by the tide after dark (measured irradiance levels of 0 umol photons m-2 s-1).

The second set of measurements consisted of light and dark (Light-Dark) sampling during the daytime. Our protocols were similar to those used to calculate light-dark bottle estimates of productivity in oceanographic and limnological applications, modified for in-situ measurements of oxygen fluxes in tide pools to estimate net community production, community respiration, and gross community production. Measurements were started shortly after pools were isolated by the receding tide, when irradiance levels were sufficient to saturate photosynthesis by N. oregona (see above). Immediately after initial measurements of O2, irradiance, and temperature (see above), pools were covered with an opaque black plastic sheet (6-mil black polyurethane plastic sheeting, Film Gard, Berry Plastics, Evansville, Indiana, USA), which was anchored to the substratum with beach cobbles to ensure that no light entered the pool during the “Dark” incubation. Irradiance levels beneath the opaque plastic sheet were uniformly 0.0 mmol photons m-2 s-1, and the presence of the plastic sheet did not appreciably alter tide pool temperatures during the “Dark” incubation (mean ± s.e.m. = -0.04 ± 0.16 °C; t = 0.2, d.f. = 4, p = 0.832). After an incubation period of at least 30 min (median = 45 min, maximum = 80 min), a second set of measurements was made, and the plastic sheet was removed, initiating a “Light” incubation of at least 30 min. At the end of the “Light” incubation, a final set of O2, irradiance, and temperature measurements was made.