Abstract: The ratio of dissolved oxygen to argon in seawater is frequently employed to estimate rates of net community production (NCP) in the oceanic mixed layer. The in situ O2/Ar‐based method accounts for many physical factors that influence oxygen concentrations, permitting isolation of the biological oxygen signal produced by the balance of photosynthesis and respiration. However, this technique traditionally relies upon several assumptions when calculating the mixed‐layer O2/Ar budget, most notably the absence of vertical fluxes of O2/Ar and the principle that the air‐sea gas exchange of biological oxygen closely approximates net productivity rates. Employing a Lagrangian study design and leveraging data outputs from a regional physical oceanographic model, we conducted in situ measurements of O2/Ar in the California Current Ecosystem in spring 2016 and summer 2017 to evaluate these assumptions within a �worst‐case� field environment. Quantifying vertical fluxes, incorporating nonsteady state changes in O2/Ar, and comparing NCP estimates evaluated over several day versus longer timescales, we find differences in NCP metrics calculated over different time intervals to be considerable, also observing significant potential effects from vertical fluxes, particularly advection. Additionally, we observe strong diel variability in O2/Ar and NCP rates at multiple stations. Our results reemphasize the importance of accounting for vertical fluxes when interpreting O2/Ar‐derived NCP data and the potentially large effect of nonsteady state conditions on NCP evaluated over shorter timescales. In addition, diel cycles in surface O2/Ar can also bias interpretation of NCP data based on local productivity and the time of day when measurements were made.

Abstract: Estimating interannual variability in carbon export is a key goal of many marine biogeochemical studies. However, due to variations in export mechanisms between regions, generalized models used to estimate global patterns in export often fail when used for intra-regional analysis. We present here a region-specific model of export production for the California Current Ecosystem (CCE) parameterized using intensive Lagrangian process studies conducted during El Niño-Southern Oscillation (ENSO) warm and neutral phases by the CCE Long-Term Ecological Research (LTER) program. We find that, contrary to expectations from prominent global algorithms, export efficiency (e-ratio = export / primary productivity) is positively correlated with temperature and negatively correlated with net primary productivity (NPP). We attribute these results to the substantial horizontal advection found within the region, and verify this assumption by using a Lagrangian particle tracking model to estimate water mass age. We further suggest that sinking particles in the CCE are comprised of a recently-produced, rapidly-sinking component (likely mesozooplankton fecal pellets) and a longer-lived, slowly-sinking component that is likely advected long distances prior to export. We determine a new algorithm for estimating particle export in the CCE from NPP (Export = 0.08 · NPP + 72 mg C m-2 d-1). We apply this algorithm to a two-decade long time series of NPP in the CCE to estimate spatial and interannual variability across multiple ENSO phases. Reduced export during the warm anomaly of 2014-2015 and El Niño 2015-2016 resulted primarily from decreased export in the coastal upwelling region of the CCE; the oligotrophic offshore region exhibited comparatively low seasonal and interannual variability in flux. The model resolves intra-regional patterns of in situ export measurements, and provides a valuable contrast to global export models.

Abstract: Oceanographic field programs often use delta15N biogeochemical measurements and in situ rate measurements to investigate nitrogen cycling and planktonic ecosystem structure. However, integrative modeling approaches capable of synthesizing these distinct measurement types are lacking. We develop a novel approach for incorporating delta15N isotopic data into existing Markov Chain Monte Carlo (MCMC) random walk methods for solving linear inverse ecosystem models. We test the ability of this approach to recover food web indices (nitrate uptake, nitrogen fixation, zooplankton trophic level, and secondary production) derived from forward models simulating the planktonic ecosystems of the California Current and Amazon River Plume. We show that the MCMC with delta15N approach typically does a better job of recovering ecosystem structure than the standard MCMC or L2 minimum norm (L2MN) approaches, and also outperforms an L2MN with delta15N approach. Furthermore, we find that the MCMC with delta15N approach is robust to the removal of input equations and hence is well suited to typical pelagic ecosystem studies for which the system is usually vastly under-constrained. Our approach is easily extendable for use with delta13C isotopic measurements or variable carbon:nitrogen stoichiometry.