On May 31, 2010, a direct acoustic measurement method was used to quantify fluid leakage rate from the Deepwater Horizon Macondo well prior to removal of its broken riser. This method utilized an acoustic imaging sonar and acoustic Doppler sonar operating onboard a remotely operated vehicle for noncontact measurement of flow cross-section and velocity from the well's two leak sites. Over 2,500 sonar cross-sections and over 85,000 Doppler velocity measurements were recorded during the acquisition process. These data were then applied to turbulent jet and plume flow models to account for entrained water and calculate a combined hydrocarbon flow rate from the two leak sites at seafloor conditions. Based on the chemical composition of end-member samples collected from within the well, this bulk volumetric rate was then normalized to account for contributions from gases and condensates at initial leak source conditions. Results from this investigation indicate that on May 31, 2010, the well's oil flow rate was approximately 0.10 AE 0.017 m 3 s −1 at seafloor conditions, or approximately 85AE 15 kg s −1 (7.4 AE 1.3 Gg d −1 ), equivalent to approximately 57,000AE 9,800 barrels of oil per day at surface conditions. End-member chemical composition indicates that this oil release rate was accompanied by approximately an additional 24 AE 4.2 kg s −1 (2.1AE 0.37 Gg d −1 ) of natural gas (methane through pentanes), yielding a total hydrocarbon release rate of 110 AE 19 kg s −1 (9.5 AE 1.6 Gg d −1 ).Gulf of Mexico | oil spill | buoyant plume | buoyant jet | subsurface A ccurate assessment of the hydrocarbon fluid release rate from well blowouts such as that of the Deepwater Horizon Macondo well provides fundamental information for evaluating intervention options to regain well control, properly scaling oil collection and containment operations, estimating the total spill volume, assessing environmental damage, and investigating well casing or blowout preventer (BOP) failure modes. On May 31, 2010, a direct acoustic technique was used to measure the volumetric flow rate of fluids (liquid and gas) emitted from the Deepwater Horizon Macondo well. This method, which adapts acoustic techniques previously developed for deep-sea hydrothermal vent research (1), enabled observation from a remotely operated vehicle (ROV) (Fig. S1) at horizontal standoff distances of between 2 and 7 m, providing a noncontact method of measurement wherein the sensors did not disturb the flow or become fouled with oil or gas hydrate accretions. Despite the optical opacity of the fluid, this acoustic technique enabled quantitatively detailed measurement of the leaks' cross-sectional areas and velocity profiles. This acoustic flow rate assessment was conducted on a "not to interfere basis" due to the well containment operations being carried out from late April through mid-July. Data were thus collected on an opportunistic basis during short time intervals between containment procedures.Acoustic measurement commenced immediately following the unsuccessful "...