Biogeochemical impact of a model western iron source in the Pacific Equatorial Undercurrent

摘要

Trace element distributions in the source waters of the Pacific Equatorial Undercurrent (EUC) show the existence of elevated total acid-soluble iron concentrations. This region has been suggested to contribute enough bioavailable iron to regulate interannual and interglacial variability in biological productivity downstream in the high-nitrate low-chlorophyll upwelling zone of the eastern equatorial Pacific. We investigated the advection and first-order biogeochemical impact of an imposed, data-based iron maximum in the western Pacific EUC using an ecosystem model forced by a global dynamical model. We imposed two source profiles of iron constrained by total acid-soluble iron measurements. Though the data for total acid-soluble iron included both dissolved and acid-soluble particulate iron species, we treated all of the total acid-soluble iron as if it was dissolved and bioavailable. A deeper (270 m) source was centered in the density horizon of the observed iron maximum and a shallower (180 m) source was located in the core of our model's EUC, where a dissolved iron maximum has been frequently postulated. These source runs were compared with a control run that contained no specific source of iron associated with the EUC. In the source runs elevated iron concentrations were simulated in the EUC across its entire zonal span, evident as a subsurface plume of dissolved iron slightly below the core of the EUC. In the control run there was no iron maximum associated with the EUC. Upwelling of iron-replete water in the central and eastern equatorial Pacific increased integrated primary productivity in the Wyrtki box (180°W:90°W, 5°S:5°N, 0:200 m) by 41% and 66% for the deeper and shallower iron perturbation, respectively. The source runs increased the realism of the zonal extent of HNLC conditions and the meridional distributions of biological productivity, relative to the control run. However, in the source simulations surface chlorophyll concentrations were too high by a factor of two and maximum surface nitrate concentrations were too low, relative to climatologies. The relative abundance of diatoms roughly doubled upon the input of additional iron, exceeding field observations. Though biogeochemical data are limited and we did not adjust parameters to optimize the model fits to observations, these results suggest that acid-soluble paniculate iron supplied to the EUC in the western equatorial Pacific is unlikely to be entirely bioavailable.