Fig. 1

Experiment of active source oceanic bottom seismometer (OBS) and velocity modeling at the Dragon Flag Hydrothermal Field (DFHF). (a) Multi-beam bathymetry map and active-source seismic survey at the DFHF. Origin of local coordinate system is at (49#cod#x000b0;39#cod#x02032;E, 37#cod#x000b0;46#cod#x02032;S), with the X-axis oriented along 82#cod#x000b0;. Red star shows the location of active vents. Black circles with white numbers and thin black dotted lines show the locations of OBSs and shots, respectively. Red line of Y3Y4 shows the location of the profile in (b). Thick white lines with #28 (in red) and #29 (in black) indicate the neo-volcanic ridges (NVRs) for segments 28 and 29 (defined by Cannat et al, 1999). Thick dashed lines indicate the non-transform discontinuities (NTDs). Red star in the inset at the top-right corner shows the location of the seismic survey area. The inset at the bottom-right corner shows the corrugated surface of Dragon Flag oceanic core complex (OCC). (b) The final model of P-wave crustal structure along Profile Y3Y4. Red star is the horizontal projection of active vents. The inverted triangles indicate locations of OBS. Red lines on the Moho mean there are PmP rays being covered. (c) The interpreted geological model. BA: Breakaway; T: Termination; and ODF: Oceanic Detachment Fault (Zhang et al, 2013)"

Fig. 2

Experiment of passive source OBS and records of microseismicities at the DFHF. (a) Red, pink and white triangles represent the locations of imported, domestic and lost OBS, respectively. Red star shows the location of active vents. (b) Filtered four-channel seismic records of OBS B65 and B71 for vertical channel (labeled V), two horizontal (H1 and H2) seismometer channels and hydrophone channel (H). The traces are scaled to equal maximum amplitude and band-pass filtered of 3~20 Hz. The start time of trace was 23:55:15.015 on Feb 18th, 2015. P waves, S waves and PwP phase are indicated by solid red line, dashed red line and dotted gray line, respectively"

Fig. 3

Travel-time residuals plotted against their source-receiver azimuths. (a) A total of 65634 pair travel-time residuals plotted against their source-receiver azimuths. The 0#cod#x000b0; azimuth corresponds to the general trend of NVR (#28), and n is the number of travel-time residuals. (b) Travel-time residuals binned and plotted against their source-receiver azimuth. The black line shows best-fitting curve of cos2#cod#x003b8;. The circles show the mean residual in each 20#cod#x000b0; bin, and the bars indicate the ranges of a standard deviation. Fast direction is indicated by the triangle with standard error bar located below the upper horizontal axis, and n is the number of bins"

Fig. 4

Theoretical diagram of anisotropy inverted from travel-time analysis. (a) Top view for anisotropic medium. Black dashed lines represent dry cracks, and thin dashed gray lines represent saturated cracks. Black lines and arrows represent axis and spreading direction. (b) Side view for anisotropic medium. Thin dashed gray lines represent saturated cracks. According to different penetrating depths of shot rays, the anisotropic characteristics from different depths can be obtained. (c) The relationship between travel-time residuals and corresponding azimuths. Black, blue and red curves indicate the relationship of cos2#cod#x003b8;, cos4#cod#x003b8; and simultaneous cos2#cod#x003b8; and cos4#cod#x003b8;, respectively. Black line means larger percentage of dry cracks, and shows closer relationship with cos2#cod#x003b8;. Blue dashed line means larger percentage of saturated cracks, and shows closer relationship with cos4#cod#x003b8;. Red curve means equivalent effects of dry and saturated cracks"

Fig. 5

Result of azimuthally seismic anisotropy technique on the western limb of the 9#cod#x000b0;N overlapping spreading center on the East Pacific Rise (EPR) (Tong et al, 2004). (a) Cosine function relationship between travel-time residuals and their azimuths; (b) estimated percentage anisotropy; (c) geometric configuration of along-axis hydrothermal circulation (arrow) consistent with crack structures inferred from anisotropic study"

Fig. 6

Shear-wave splitting occurred when a shear wave traveled through an anisotropic medium (W#cod#x000fc;stefeld et al, 2008). When incident shear wave arrived at an anisotropic medium, it split into two shear waves of perpendicular polarization along seismic fast and slow directions, respectively. Traveling through anisotropic medium, the two waves accumulated a delay time #cod#x003b4;t. Shear-wave splitting techniques inverted for #cod#x003b4;t and the fast polarization direction"

Fig. 7

Result of shear-wave splitting at the axis of Mid-Atlantic Ridge (MAR) near 35#cod#x000b0;N (Barclay et al, 2003). (a) Rose histograms of fast polarization directions for six OBSs. (b) Examples of shear-wave splitting in horizontal particle motions. The shear wave arrivals are shown for earthquakes at four different OBSs (52, 56, 60, and 63), with fast wave moving in a nearly NS direction and slow wave moving in a nearly EW direction. Numbers in parentheses are the earthquake epicenters in minutes of latitude and longitude, respectively. Each trace was 203ms long, with sample points (dots) every 7.8125ms. The open circle is the origin of particle motion"