A nutrient relay sustains subtropical ocean productivity.

The expansive gyres of the subtropical ocean account for a significant fraction of global organic carbon export from the upper ocean. In the gyre interior, vertical mixing and the heaving of nutrient-rich waters into the euphotic layer sustain local productivity, in turn depleting the layers below. However, the nutrient pathways by which these subeuphotic layers are themselves replenished remain unclear. Using a global, eddy-permitting simulation of ocean physics and biogeochemistry, we quantify nutrient resupply mechanisms along and across density surfaces, including the contribution of eddy-scale motions that are challenging to observe. We find that mesoscale eddies (10 to 100 km) flux nutrients from the shallow flanks of the gyre into the recirculating interior, through time-varying motions along density surfaces. The subeuphotic layers are ultimately replenished in approximately equal contributions by this mesoscale eddy transport and the remineralization of sinking particles. The mesoscale eddy resupply is most important in the lower thermocline for the whole subtropical region but is dominant at all depths within the gyre interior. Subtropical gyre productivity may therefore be sustained by a nutrient relay, where the lateral transport resupplies nutrients to the thermocline and allows vertical exchanges to maintain surface biological production and carbon export.

Attribution of Space-Time Variability in Global-Ocean Dissolved Inorganic Carbon.

The inventory and variability of oceanic dissolved inorganic carbon (DIC) is driven by the interplay of physical, chemical, and biological processes. Quantifying the spatiotemporal variability of these drivers is crucial for a mechanistic understanding of the ocean carbon sink and its future trajectory. Here, we use the Estimating the Circulation and Climate of the Ocean-Darwin ocean biogeochemistry state estimate to generate a global-ocean, data-constrained DIC budget and investigate how spatial and seasonal-to-interannual variability in three-dimensional circulation, air-sea CO2 flux, and biological processes have modulated the ocean sink for 1995-2018. Our results demonstrate substantial compensation between budget terms, resulting in distinct upper-ocean carbon regimes. For example, boundary current regions have strong contributions from vertical diffusion while equatorial regions exhibit compensation between upwelling and biological processes. When integrated across the full ocean depth, the 24-year DIC mass increase of 64 Pg C (2.7 Pg C year-1) primarily tracks the anthropogenic CO2 growth rate, with biological processes providing a small contribution of 2% (1.4 Pg C). In the upper 100 m, which stores roughly 13% (8.1 Pg C) of the global increase, we find that circulation provides the largest DIC gain (6.3 Pg C year-1) and biological processes are the largest loss (8.6 Pg C year-1). Interannual variability is dominated by vertical advection in equatorial regions, with the 1997-1998 El Niño-Southern Oscillation causing the largest year-to-year change in upper-ocean DIC (2.1 Pg C). Our results provide a novel, data-constrained framework for an improved mechanistic understanding of natural and anthropogenic perturbations to the ocean sink.

Stable aerobic and anaerobic coexistence in anoxic marine zones.

Mechanistic description of the transition from aerobic to anaerobic metabolism is necessary for diagnostic and predictive modeling of fixed nitrogen loss in anoxic marine zones (AMZs). In a metabolic model where diverse oxygen- and nitrogen-cycling microbial metabolisms are described by underlying redox chemical reactions, we predict a transition from strictly aerobic to predominantly anaerobic regimes as the outcome of ecological interactions along an oxygen gradient, obviating the need for prescribed critical oxygen concentrations. Competing aerobic and anaerobic metabolisms can coexist in anoxic conditions whether these metabolisms represent obligate or facultative populations. In the coexistence regime, relative rates of aerobic and anaerobic activity are determined by the ratio of oxygen to electron donor supply. The model simulates key characteristics of AMZs, such as the accumulation of nitrite and the sustainability of anammox at higher oxygen concentrations than denitrification, and articulates how microbial biomass concentrations relate to associated water column transformation rates as a function of redox stoichiometry and energetics. Incorporating the metabolic model into an idealized two-dimensional ocean circulation results in a simulated AMZ, in which a secondary chlorophyll maximum emerges from oxygen-limited grazing, and where vertical mixing and dispersal in the oxycline also contribute to metabolic co-occurrence. The modeling approach is mechanistic yet computationally economical and suitable for global change applications.

Towards seasonal forecasting of malaria in India.

BACKGROUND: Malaria presents public health challenge despite extensive intervention campaigns. A 30-year hindcast of the climatic suitability for malaria transmission in India is presented, using meteorological variables from a state of the art seasonal forecast model to drive a process-based, dynamic disease model. METHODS: The spatial distribution and seasonal cycles of temperature and precipitation from the forecast model are compared to three observationally-based meteorological datasets. These time series are then used to drive the disease model, producing a simulated forecast of malaria and three synthetic malaria time series that are qualitatively compared to contemporary and pre-intervention malaria estimates. The area under the Relative Operator Characteristic (ROC) curve is calculated as a quantitative metric of forecast skill, comparing the forecast to the meteorologically-driven synthetic malaria time series. RESULTS AND DISCUSSION: The forecast shows probabilistic skill in predicting the spatial distribution of Plasmodium falciparum incidence when compared to the simulated meteorologically-driven malaria time series, particularly where modelled incidence shows high seasonal and interannual variability such as in Orissa, West Bengal, and Jharkhand (North-east India), and Gujarat, Rajastan, Madhya Pradesh and Maharashtra (North-west India). Focusing on these two regions, the malaria forecast is able to distinguish between years of "high", "above average" and "low" malaria incidence in the peak malaria transmission seasons, with more than 70% sensitivity and a statistically significant area under the ROC curve. These results are encouraging given that the three month forecast lead time used is well in excess of the target for early warning systems adopted by the World Health Organization. This approach could form the basis of an operational system to identify the probability of regional malaria epidemics, allowing advanced and targeted allocation of resources for combatting malaria in India.