Magnetic anomalies at spreading centers - from global sea-surface data compilations to local deep-sea investigations
Jérôme Dyment, Yujin Choi (both at IPGP and CNRS, Paris, France; [email protected]), and Florent Szitkar (GEOMAR, Kiel, Germany)
Marine magnetics have been instrumental in the construction and general acceptance of Plate Tectonics in the 60s and 70s. The recognition of geomagnetic reversal sequences recorded by the oceanic crust allows dating the seafloor and reconstructing the plate motions through time at divergent plate boundaries. Beyond this well-known exercise of pattern recognition, marine magnetics deliver precious information on the magnetic source geometry and properties, and therefore on seafloor spreading processes and the evolution of the geomagnetic field. In this presentation, I will show (1) how we can learn further on these aspects from a global compilation of sea-surface anomalies, and (2) how deep-sea magnetic anomaly data, although sparse and more costly to acquire, provide invaluable high-resolution information.
As a by-product of the construction of a new World Digital Magnetic Anomaly Map over oceanic areas (available at wdmam.org), we use an original approach based on the global forward modeling of seafloor spreading magnetic anomalies and their comparison to the available sea-surface marine magnetic data to derive the first map of the equivalent magnetization over the World's ocean. This map reveals consistent patterns related to the age of the oceanic lithosphere, the spreading rate at which it was formed, and the presence of mantle thermal anomalies which affects seafloor spreading and the resulting lithosphere. As for the age, the equivalent magnetization decreases significantly during the first 10-15 Myr after its formation, probably due to the alteration of crustal magnetic minerals under pervasive hydrothermal alteration, then increases regularly between 20 and 70 Ma, reflecting variations in the field strength or source effects such as the acquisition of a secondary magnetization. As for the spreading rate, the equivalent magnetization is twice as strong in areas formed at fast rate than in those formed at slow rate, with a threshold at ~40 km/Myr, in agreement with an independent global analysis of the amplitude of Anomaly 25. This result, combined with those from the study of the anomalous skewness of marine magnetic anomalies, allows building a unified model for the magnetic structure of normal oceanic lithosphere as a function of spreading rate. Finally, specific areas affected by thermal mantle anomalies at the time of their formation exhibit peculiar equivalent magnetization signatures, such as the cold Australia-Antarctic Discordance, marked by a lower magnetization, and several hotspots, marked by a higher magnetization.
Although more difficult and costly to acquire, near-seafloor magnetics can significantly increase the signal resolution. Either acquired with a magnetometer towed at depth from the ship or mounted on a deep-sea exploration vehicle (AUV, ROV, manned submersible), they can help, for instance, to decipher geomagnetic intensity variations and date the seafloor at an accuracy of 10 to 100 kyr, unravel the presence of hot axial dykes or hydrothermal sites at spreading centers. Focusing on the last point, the magnetic signature of hydrothermal sites strongly depends on the basement and the vent temperature. New examples from active and inactive sites on mid-ocean ridges and back-arc basins confirm that basalt-hosted sites show a magnetic low which results from the presence of low magnetic sulfide deposit, the alteration of basalt, and possible thermal demagnetization of basalt. The shape of the anomaly can be used to infer tectonic rotations. Active and inactive high-temperature ultramafic-hosted sites exhibit a magnetic high that results from the formation of magnetite as a result of high-temperature serpentinization. The strength and diversity of these anomalies helps to detect and characterize seafloor massive sulfide deposits and potential deep-sea mining targets.
