Fracture healing in a magma: An experimental approach and implications for volcanic seismicity and degassing

Abstract

The healing of magmatic fractures is considered essential to repetitive seismicity and the closure of degassing paths during emplacement of lavas. To estimate the healing time of magmatic fractures, we performed healing experiments on rhyolitic melts at 850°–1000°C and 1.6–3.2 MPa for 0.5–94 h. Two cylindrical obsidian cores were juxtaposed on surfaces prepared by cutting the cores both with and without polishing. These were annealed in an open‐system cell. The contact interface became coherent and finally disappeared. The water content across the contact initially decreased toward the interface via diffusive dehydration, but later homogenized. This change was interpreted to reflect atomic‐scale closure of the interface, probably by chemical bonding. We defined this closure interval as microscopic healing time and determined this by fitting the measured profiles with a diffusion model. The microscopic healing time was strongly dependent on temperature and roughness of the interface and was, for the nonpolished interfaces, 67–74, 4.0–4.9, and 0.36–0.38 h at 850°, 900°, and 950°C, respectively, whereas for the polished examples it was 1–3 and 0.5–0.6 h at 850° and 900°C, respectively. This microscopic healing time is consistent with the period of actual seismicity and is prolonged sufficiently to permit the formation of millimeter‐thick bubble‐free obsidian layers along fractures in vesicular lavas through bubble resorption due to diffusive degassing.