SEISMIC WHILE DRILLING

Hammer Drill Observations with an Ocean-Bottom Seismometer at Site 1104

At Hole 1104E tests were made of an SDS Digger Tools 12-in fluid hammer drill using an SDS concentric winged bit from 2140 to 2320 on 1 May. Two ocean-bottom seismometers (OBSs) were deployed at ~90 m (A4) and ~300 m (A8) from the test drilling site. Because of its high signal level, the drill-string accelerometer was not placed on the drill string, but recordings were made for 1.5 hr on the downhole measurement laboratory deck.

Figure F82 shows the 80-min interval from 2141 to 2300 covering the test, as recorded on the vertical component of OBS A4. The energy coupled to the ground varied greatly, depending on the fluid pressure (pumping rate) applied to the drill. The initial 10-min trace shows a gradual intensity increase as pumping pressure was increased to 1650 psi. At several points there were short, more intense intervals followed by the drill stalling. The increase in amplitude followed by stalling may have been caused by the heave of the ship. When not on bottom, the drill goes into bypass mode and resumes hammering when sufficiently loaded to close the bypass valve. The lower levels on the sixth trace (2231 to 2240) and the beginning of the eighth trace (2251 to 2254) may represent intervals in which the weight on bit was possibly too little or too great or drill orientation was incorrect for optimal coupling.

Figure F83 shows details of the waveform at 22:06:20 when the hammer drill stalled. The signal tapered off over 10 or more cycles. The signal appears to be richer in the second harmonic than the primary frequency (21 Hz) of the bit, or first harmonic.

SWD Experiments at Site 1105

While drilling at Hole 1105A, concurrent measurements of the accelerations of the drill string and of ground and water-borne seismic energy were made over four intervals totaling about 39 hr.

The hole was drilled with a four-roller-cone coring bit from 15 to 158 m; gabbro, olivine gabbro, and oxide gabbro were recovered. The majority of the drilled section was massive, but there were intervals showing deformation and/or alteration. A four-arm caliper log of the hole showed portions of it to have ellipticity of up to 25%. Observations were made for 54 m (34%) of the 158 m drilled. The data acquisition particulars are described in "Seismic While Drilling" (in "Seismic Drilling") in the "Explanatory Notes" chapter. The accelerometer data are arbitrarily grouped in 30-min 4.464 MB files. Each 1-s record consists of 2480 bytes (400 samples per second, two bytes/sample, three channels, and an 80-byte ASCII header).

The two OBSs were deployed at ~1000 on 3 May and recovered at ~1330 on 10 May. The OBS data are recorded in one continuous 732-MB file (200 samples per second, two bytes/sample, three channels) for 7 days. The three channels consisted of a vertical geophone, a horizontal geophone, and a hydrophone. The OBS data concurrent with the accelerometer data were extracted from the two complete OBS data files, and converted into 40-s traces (SEGY format) for plotting and correlation purposes. The trace length was arbitrarily chosen. This produced working files consisting of 860, 908, 540, and 1215 records for each OBS (a total of 7046 traces).

Sample portions of these data time series and spectra were examined to (1) determine recording performance, (2) compare the signals between intervals of actual drilling and while the bit was raised during core retrieval, and (3) look for the anticipated spectral components relating to

  1. Drill-string rotation rate (50 to 55 rpm with four roller cones);
  2. Circulating pump (typically 80 strokes/min);
  3. Drilling fluid circulation; and
  4. Ship-generated noise (notably 15 Hz and harmonics from the 900 rpm of the ship's four running diesel engines).

Figure F84 compares samples of drill-string acceleration spectra during drilling and while the bit is raised during core recovery. The 0.68-Hz peak in the upper plot corresponds to the 41-rpm drill speed during core recovery. The 2.9-Hz peak in the lower plot appears to be the drill-string rotation rate (44 rpm) times the number of roller cones (four).

Figures F85 and F86 are the time series and spectra for two 31-s samples, the first while not drilling and the second while drilling. Note that the level while drilling is 1.5 to 3 times greater, depending on the component, than the level while not drilling. The cutting action of the bit produces more energy in the 8-to 40-Hz band. The spectral spikes at 15 Hz, the doublet at 29/30 Hz is ship engine noise. The 5- to 7-Hz peak in the seismometer traces, but not the hydrophone, is presently unexplained.

A comparison of the spectral responses between seismometer A4 and A8, both in deployment and then on the lab bench, indicates that seismometer A4 has less of the lower frequency response than A8.

Locating the OBSs on the Seafloor

The precise positioning capability of the ship allowed us to determine the bottom location of the OBSs with greater accuracy than is done in their customary use for refraction profiles. The ship is positioned with respect to a seafloor acoustic beacon (Datasonics model UAB-354M) using an array of three hydrophones, spaced about 24 and 30 m apart and lowered on rods 6 m below the ship. The OBSs descended at 0.8 m/s. After being released from their 45-kg anchor, the ascent rate of the OBSs was measured to be 1 m/s. Their positions on the seafloor were determined by using the dynamic positioning system to locate the ship 200 m east, north, west, and south of the hole and at the hole. At each position, we made slant-ranging measurements to the OBSs' transponder with a Benthos model DS-7000 acoustic deck unit. These ranges, together with the water depth at each OBS, allowed us to calculate the horizontal locations. The 200-m offsets were determined by the amount that the dynamic positioning operators felt to be reliable.

At the second deployment in 700 m depth, OBS A4 was dropped 100 m north of the hole and landed at 60 m, 060º from the hole, 107 m to the southeast. OBS A8 was dropped 300 m north of the hole and landed 230 m north of the hole, 70 m to the south of the drop point. Drifts were both southerly and 15% and 12% of the water depth, respectively. The estimated precision is 20 m.

NEXT