Physical ocean model

Any ecosystem model needs a physical model input. The physical model can be very simple, like vertical mixing or vertical velocity, or can be as complicate as a full 3D circulation model.

The baseline of the 3D model is Module Ocean Model (MOM) developed at the GFDL/NOAA, Princeton. The model covers the entire Pacific except the Southern Ocean section. The resolution is 2 deg. in longitude, 0.5 deg near the equator (10S to 10N), and 2 deg. close the north and south boundaries. There are 40 layers with 10m resolution within the euphotic zone (120m). The surface forcing uses COADS monthly averaged fields from 1950 to 1993, with a restoring surface salinity process back to Levitus climatological value.

NPZD model

The physical processes bring up nutrients in the surface euphotic layer. Let start with nitrate. The small phytoplankton group takes up nitrate through the photosynthesis, while micro-zooplankton graze on small phytoplankton, then excrete NH4 and detritus-nitrogen. At this stage, we have a typical NPZ model with NH4 and detritus, 5 components model. If we add DON and bacteria, it will return back to Fasham’s original 7 components model.

The physical processes also bring up silicate, another major nutrient, into the euphotical zone. The silicate supports diatoms grow, which will feed the meso-zooplankton. Again, the meso-zooplankton excrete NH4 and detritus-silicon pool. The silicate loop is described by following the red lines. The white lines show the nitrogen pathway. The carbon removal will be linked to the biological processes in the upper ocean, but also by the air-sea exchange of CO2 at the surface.

This map shows the modeled surface phytoplankton biomass (small phytoplankton and diatoms combined).

The model simulates the equatorial region very well with elevated high biomass in the cold tongue region, as well as the subarctic area. The modeled simulated spatial pattern agrees well with the SeaWiFs satellite observed data. From now on, all the figures will show the time series of different variables averaged over the Nino3 box.

The model simulates all El Niño events well with higher temperature anomaly during the El Niño, and cooler temperature during the La Niña. The temperature simulation indicates the physical model and surface forcing are reasonably setup.

The total phytoplankton biomass for the top 120m decreases during all five El Niño events, not just near the surface but throughout the water column. The strongest reduction of the phytoplankton biomass occurred during the 1982-83 El Niño, and followed by 1972-73 El Niño. The highest biomass periods are 1987-88 and 1974.

The primary reason for the phytoplankton biomass decrease is due to the decrease in silicate concentration during the El Niño events.

Beside the strong interannual variability, nitrate concentration tends to be higher before 1975-76 comparing to the 1980s. That is related to the Pacific decadal oscillation (PDO). In order to show the decadal variation of the modeled nitrate, 5 year running mean has been applied to the nitrate anomaly field. Nitrate concentration is about 2 mmol/m³ higher before 1975-76 (1965-1976) than the period after (1977-992). This is primarily due to the wind change in this region (not shown). Some limit data suggested that there was a decrease in the trade wind after 1975-76 in the tropics, which resulted less upwelling for the Nino3 box. There are not enough direct measurements of nitrate concentration to compare with on the decadal time scale. But fluctuation in fish stock may provide indirect evidence that the primary production in the equatorial Pacific decreased after 1975-76 (see more on climate and fisheries…).