An acoustic subsea blowout-preventer control system must distinguish between transmitted signals and ambient noise. Noise levels taken near an underwater gas vent are presented. These data are extrapolated to determine a limit for designing an acoustic control system. Scaling criteria and hydrophone placement are discussed, and an acoustic control system designed to operate under severe blowout conditions is described. Introduction In floating drilling operations, it is necessary to be able to close the blowout preventers (BOP's) on wells that are being drilled if the redundant primary control systems fail. In shallow water this is usually done with the aid of divers. Where diving is more difficult, as in deeper water or where severe environmental conditions are anticipated, it has been common to use a drop in hydraulic pressure to automatically close the annular preventer. It is better, however, to selectively close preventer. It is better, however, to selectively close the major functions on the BOP stack or to release the riser remotely; this can be done with an acoustic backup system. To design an acoustic system, the maximum background noises at the time the system is to be used should be estimated. A blowout at a gas flow rate of about 100 MMscf/D probably will generate the maximum noise level anticipated at the BOP. Available noise data dealing with gas turbulence in air and cavitation noise are not directly applicable for estimating the noise of gas being released underwater. To our knowledge, no noise data regarding gas vented subsea have been published. Consequently, representative data were taken using an inactive dry gas vent. However, venting gas at the rate of 100 MMscf/D through existing underwater gas vents is impractical, so lower rates had to be used in actual field measurements. This paper presents noise levels taken near an underwater gas vent flowing at controlled rates up to 11.4 MMscf/D. These data are extrapolated to 100 MMscf/D to determine a limit for designing an acoustic control system. Scaling criteria and hydrophone placement are discussed, and an acoustic control system designed to operate under severe blowout conditions is described. Subsea Noise Tests Test Situation The data were taken from a vent in 84 ft of water. Gas was vented vertically through a 5.5-in.-ID pipe with no restrictions at the outlet (see Fig. 1). A calibrated hydrophone with a preamplifier was suspended 6 ft from the flare opening, at about 120 degrees from the vertical center line of the gas vent. The remaining sound measuring equipment was located on a vessel moored near the vent. Hard copies of each spectral analysis were obtained on site and the sound data were recorded on magnetic tape for further evaluation. All ship's power was shut down while taking data, and instrument power was furnished by a small generator set on styrofoam pads for sound insulation. The data were taken during pads for sound insulation. The data were taken during short sessions over 3 days, and ambient noise conditions were measured during each session. The data were corrected for hydrophone sensitivity, band-pass, and spreading loss. Data taken on the first 2 days were converted to reflect the ambient noise on the third day when the highest ambient noise was read (see Appendix A.) Test Results Noise spectra in the 1- to 50-kHz range from the subsea vent at rates varying from 0 (background) to 11.4 MMscf/D are presented in Fig. 2. JPT P. 259