ORIGINAL STUDY OF THE EQUATORIAL PACIFIC OCEAN FERTILIZATION BASED ON A LAGRANGIAN SIMULATION OF THE CIRCULATION COUPLED TO ND ISOTOPIC COMPOSITION AND RARE EARTH ELEMENT CONCENTRATION DATA
Abstract
The Equatorial Pacific Ocean thermocline is mostly fed and fertilized by the Equatorial UnderCurrent (EUC), which is carrying thermocline waters originating from the Southwest Tropical Pacific Ocean. Circulation in the Coral, Solomon and Bismark Seas is particularly complex, mostly due to the topography that concentrate the broad currents into narrow and swift jets. Here we used the lagrangian tool ARIANE (www.univbrest.fr/lpo/ariane/), that integrates ouputs of a ¼° global OGCM, allowing us to reconstruct pathways of “virtual tracers”. Seasonal and interanual variations of this complex circulation have been investigated. The numerical results are confronted to in situ observations of lithogenic tracers, Nd isotopic composition (IC) and REEs concentration in the same area and at locations considered by the model. ARIANE simulations show that in situ stations in the Coral Sea and in the Bismark Sea are connected by the thermocline currents, which flow from the southernmost station towards the northern ones. The Solomon Sea station seems to be on a different streamline. Using the simulated transports allowed us to refine the rates of input of the lithogenic tracers into the EUC. New Nd IC and REE data obtained in the Coral Sea confirm the occurrence of lithogenic element supplies and scavenging in Coral and Bismark Seas, initially suggested by Lacan and Jeandel (2001) and Cros (2008), although these authors used a very rough approach of the circulation. Furthermore, lithogenic supplies appear to occur between South-Eastern Pacific and the Coral Sea. We also noticed a progressive enrichment in the Coral Sea, then in the Bismark Sea. These results stand in contrast with past interpretations which located most of enrichments in the Bismark Sea. These conclusions drawn for thermocline waters bound to the EUC also hold for surface and intermediate waters. This work underlines that associating virtual and real tracer analysis could be a good track for the future of the data-model synergy in GEOTRACES.