An episode of seafloor spreading and associated plate deformation inferred from crustal fluid pressure transients

Three-year records of crustal fluid pressures and temperatures at four Ocean Drilling Program (ODP) sites on the northern Juan de Fuca Ridge and eastern ridge flank reveal a broad range. of variations that include hydrologic transients that are contemporaneous with earthquakes along the ridge axis, the Nootka transform fault, and within the Juan de Fuca plate. One example of such a transient is the response to what is inferred to be a seafloor spreading event that triggered a swarm of earthquakes near the Endeavour ridge segment, beginning with a M-W = 4.6 event on June 8, 1999. The largest transients were observed at ODP Sites 1024 and 1025 located 25.6 and 33.5 km east of the Endeavour axis. Pressures rose coseismically with the first earthquake of the swarm by roughly 1.6 and 1 kPa, continued to rise to maxima of 3.2 and 2.0 kPa within hours, then decayed to half the peak levels in 1 and 2 days at the two respective sites. A small (similar to0.2 kPa) response of the same sign followed by a decay over 100 days was observed at Site 1027 situated 101 km east of the axis, and a similarly small response but of opposite sign was observed at Site 857 in Middle Valley, located along strike of the Endeavour segment roughly 50 km to the north of the earthquake swarm. The pressure transients are inferred to reflect a combination of the instantaneous internal plate deformation associated with extension at the ridge axis and lateral water flow in the crust following strain-induced fluid pressure gradients. The rate at which the transients dissipate constrains the regional-scale permeability of the upper igneous crust to be of the order of 10(-10)-10(-9) m(2). Instantaneous strain calculated from the initial amplitude of the transients ranges from 8 X 10(-9) at the most distal site to 1.7 X 10(-7) at the most proximal. The magnitude of regional strain is much larger than that which would result from the simultaneous earthquake, and we conclude that the first and all subsequent earthquakes of the swarm, and the crustal strain responsible for the hydrologic transients, are the consequence of a dominantly aseismic spreading "event" involving similar to 12 cm of dilatation at the ridge axis. There were no clear indications of associated magmatic activity; hence episodic spreading may take place outside the influence of either dike injection or seismic rupture. Given the excellent sensitivity of pore pressure to strain, we anticipate that this simple observational technique can be applied to the investigation of seismic and aseismic deformation in a variety of tectonically active settings.