Methods for measuring benthic nutrient flux on the California Margin: Comparing shipboard core incubations to in situ lander results

The objective of this study was to compare two techniques for estimating benthic fluxes of nutrients (nitrate, phosphate, and silicic acid) and Ge/Si flux ratios. In situ flux chambers were deployed, and cores were collected and incubated at 9 sites along the California margin in July 2001. Both techniques were successful at 8 sites, at depths from 100 to 3300 m. Flux chambers were deployed for 1 to 2 d, and cores were incubated for slightly longer on board the ship in a cold room. In some cases, core incubation flux temperature varied by up to 5°C from in situ temperature, and core incubation results were adjusted for this factor based on the effects of temperature on diffusivities and the adsorption of silicic acid. Sites studied had a range in nutrient fluxes of more than an order of magnitude, based on in situ chambers. The temperature-adjusted core incubation fluxes showed a similar, but slightly smaller range. Both methods had similar precision based on replicates, with uncertainties for high flux stations that were 5% to 20% of the mean. Only phosphate showed significant (95% confidence level) spatial variability in replicate cores; the larger in situ flux chambers had less spatial variability. The two techniques did show some systematic differences that are attributed to several artifacts created by core recovery. Silicic acid fluxes from cores were significantly lower than in situ fluxes at 2 sites; overall averages were about 80% of those for in situ chambers. The differences are attributed to reduced macrofaunal irrigation in incubated cores. Nitrate uptake in core incubations at 5 of 8 stations was significantly lower than in situ uptake; shipboard rates for all sites averaged about 66% of in situ chamber rates. This difference is attributed primarily to decreased denitrification rates in recovered cores in response to altered temperature and pressure. Phosphate fluxes from cores were significantly lower at only one site; overall, results for the two techniques were indistinguishable. Only one site had a significantly different Ge/Si flux ratio