Chimney Identification Tool for Automated Detection of Hydrothermal Chimneys from High-Resolution Bathymetry Using Machine Learning

Identifying the locations of hydrothermal chimneys across mapped areas of seafloor spreading ridges unlocks the ability to research questions about their correlations to geology, the cooling of the lithosphere, and deep-sea biogeography. We developed a Chimney Identification Tool (CIT) that utilizes a Convolutional Neural Network (CNN) to classify 1 m gridded AUV bathymetry and identify the locations of hydrothermal vent chimneys. A CNN is a type of Machine-Learning model that is able to classify raster data based on the shapes and textures in the input, making it ideal for this task. The criteria that have been used in previous manual classifications of chimneys have focused on the round base and spire shape of the features, and are not easily quantifiable. Machine-Learning techniques have previously been implemented with sonar data to classify seafloor geology, but this is the first application of these methods to hydrothermal systems. In developing the CIT, we compiled the bathymetry data from two rasters from the Endeavor Ridge—each gridded at a 1 m resolution—containing 34 locations of known hydrothermal chimneys, and from the 92° W segment of the Galapagos Spreading Center (GSC) containing 14. The CIT produced a primary group of outputs with 96% agreement with the manual classification; moreover, it correctly caught 29 of the 34 known chimneys from Endeavor and 10 of the 14 from the GSC. The CIT is trained to identify features with the characteristic shape of a hydrothermal vent chimney; therefore, it is susceptible to the misclassification of unusually shaped cases, given the limited training data. As a result, to provide the option of having a more inclusive application, the CIT also produced a secondary group of output locations with 61% agreement with the manual classification; moreover, it caught three of the four additional known chimneys from the GSC and four of the five from Endeavor. The CIT will be used in future investigations where an inventory of individual chimneys is important, such as the cataloguing of off-axis hydrothermal venting and the investigation of chimney distribution in connection to seafloor eruptions.

Skew of mantle upwelling beneath the East Pacific Rise governs segmentation

Mantle upwelling is essential to the generation of new oceanic crust at mid-ocean ridges, and it is generally assumed that such upwelling is symmetric beneath active ridges. Here, however, we use seismic imaging to show that the isotropic and anisotropic structure of the mantle is rotated beneath the East Pacific Rise. The isotropic structure defines the pattern of magma delivery from the mantle to the crust. We find that the segmentation of the rise crest between transform faults correlates well with the distribution of mantle melt. The azimuth of seismic anisotropy constrains the direction of mantle flow, which is rotated nearly 10 degrees anticlockwise from the plate-spreading direction. The mismatch between the locus of mantle melt delivery and the morphologic ridge axis results in systematic differences between areas of on-axis and off-axis melt supply. We conclude that the skew of asthenospheric upwelling and transport governs segmentation of the East Pacific Rise and variations in the intensity of ridge crest processes.

Seismic reflection images of the Moho underlying melt sills at the East Pacific Rise

The determination of melt distribution in the crust and the nature of the crust-mantle boundary (the 'Moho') is fundamental to the understanding of crustal accretion processes at oceanic spreading centres. Upper-crustal magma chambers have been imaged beneath fast- and intermediate-spreading centres but it has been difficult to image structures beneath these magma sills. Using three-dimensional seismic reflection images, here we report the presence of Moho reflections beneath a crustal magma chamber at the 9 degrees 03' N overlapping spreading centre, East Pacific Rise. Our observations highlight the formation of the Moho at zero-aged crust. Over a distance of less than 7 km along the ridge crest, a rapid increase in two-way travel time of seismic waves between the magma chamber and Moho reflections is observed, which we suggest is due to a melt anomaly in the lower crust. The amplitude versus offset variation of reflections from the magma chamber shows a coincident region of higher melt fraction overlying this anomalous region, supporting the conclusion of additional melt at depth.