Analysis of the Space-time Correlation between Seismic Activity and Exit-fluid Temperature at the TAG Hydrothermal Mound


DOEI Funded Research: 2008


Within the past decade it has become apparent that earthquakes play an important role in hydrothermal processes.  We have known that faults may serve as high-permeability conduits for sub-surface fluid flow since deep-sea vent fields were first discovered, but the specific ability of earthquakes to perturb or modulate sub-surface flow fields in time and space has only recently become apparent.  Hydrologic records from volcanically-hosted vent fields on the East Pacific Rise (EPR) and the Juan de Fuca Ridge (JdFR) have been shown to exhibit perturbations in the immediate aftermath of seismicity episodes, and these observations have stimulated a great deal of research.  First and foremost we need to understand the causality underlying these observations, and then it may be possible to use parameters such as the time lag between seismicity and the ensuing flow perturbations to deduce sub-surface flow parameters, such as velocities and permeabilities.  Naturally occurring earthquakes thus provide a perturbation mechanism that may be used to constrain hard-to-measure sub-surface flow parameters that have eluded hydrothermal researchers for decades.

A concerted effort has recently been made within the international mid-ocean ridge (MOR) hydrothermal research community to conduct coordinated time-series experiments where exit-fluid properties are measured contemporaneously with ground motions from ocean bottom seismometer networks.  These studies have been conducted in a variety of settings along the global MOR system, providing the data needed to carefully assess the relationship between seismicity and fluid flow at MOR deep-sea vent fields.  These datasets have proven to be surprisingly challenging to interpret.  Mid-ocean ridges are high-seismicity-rate environments, and hydrothermal fields may experience hundreds, or even thousands, of small earthquakes each day.  At the same time, hydrothermal fluids discharge from large areas of the seafloor with continuously variable properties that are modulated by a host of factors, including tides and bottom currents that have nothing to do with earthquakes.  Upon closer inspection, it has become apparent that the problem of assigning causality between earthquakes and exit-fluid properties is a challenging statistical problem that must be solved before we can exploit the phenomenology to provide new constraints on sub-surface hydrothermal fluid flow.

We propose to address this fundamental problem in hydrothermal research by developing, fitting, and testing new statistical models to identify and quantify relationships between high-rate seismicity catalogues and complex hydrologic records obtained at MORs using the unique dataset acquired during the “Seismicity, structure, and fluid flow of TAG” (STAG) experiment conducted by WHOI scientists Reves-Sohn, Humphris, and Canales.  Exit-fluid temperature were continuously measured every 8-10 minutes at 21 sites on the active TAG mound from June, 2003 – November, 2004, and a complete seismicity catalog for all earthquakes with local magnitudes greater than -2 was acquired during approximately the same time period was acquired with a local network of 13 ocean bottom seismometers (more than 50,000 events).  The methods we develop will not only help unravel the complex sub-surface flow system at TAG, but will also be flexible enough to be applied at any hydrothermal site for which simultaneous earthquake and fluid flow data are available.