Constraining Ocean Biogeochemical Models with Dissolved Iron Observations

Iron is a limiting factor to biological activity in some ocean regions and plays a vital role in oceanic biogeochemistry. We analyze a global database of dissolved iron measurements to better understand marine iron cycling and its relation to atmospheric and continental sources. The observational data are heavily weighted towards the upper ocean, with 68% from the upper 103m and 88% from the upper 502m, and towards the Northern Hemisphere (75%). In the deep ocean iron is clustered together within a relatively narrow range (0.2-1.0 nM) when margin-influenced data are excluded. In the upper ocean (< 500m) iron concentrations are much more variable, but generally low (< 0.4 nM) except in the high dust input regions of the North Atlantic and North Indian oceans. The lowest concentrations are seen in the Southern Ocean and the Equatorial Pacific. Observed iron distributions are correlated with atmospheric dust deposition in a non- linear fashion reflecting variable particle scavenging and biological uptake. We compare the dissolved iron observations with output from a Biogeochemical Elemental Cycling (BEC) ocean model. Scavenging of iron by particles is parameterized as a function of particle concentration and ambient dissolved iron concentrations. The model output was in general agreement with the field data (r = 0.77, for 103-502 m depths). At lower iron concentrations (< 0.3 nM) the model was biased high relative to observations. Future work will focus on improving the BEC iron cycling parameterizations utilizing the observational database.