Temporal variation of methane flares in the ocean above Hydrate Ridge, Oregon

This is an author's peer-reviewed final manuscript, as accepted by the publisher. The published article is copyrighted by Elsevier and can be found at: http://www.journals.elsevier.com/earth-and-planetary-science-letters/.While bubble plumes have been acoustically imaged in the water column above marine gas\ud hydrate deposits in many studies, little is known about the temporal variation in plume intensity. In\ud July 2008, we conducted surveys using 3.5 and 12 kHz echosounders and a 75 kHz acoustic Doppler\ud current profiler (ADCP) over the northern and southern summits of Hydrate Ridge, on the Cascadia\ud continental margin. Our study included multiple surveys at both sites, including a survey of the\ud northern summit that was repeated 16 times in 19 hours. Seafloor depth at the northern summit is\ud ~600 m, well within the hydrate stability zone (HSZ), which was below ~510 m during our survey\ud based on CTD data. Three distinct flares (a term used to denote the acoustic signature of bubble\ud plumes) were detected at Northern Hydrate Ridge (NHR) and one was detected at Southern Hydrate\ud Ridge (SHR), coincident with where flares were observed a decade ago, indicating that the supply of\ud gas is stable on this time scale. High-resolution bathymetric surveys of NHR and SHR acquired with an\ud Autonomous Underwater Vehicle (AUV) flown ~50 m above the seafloor indicate that flare locations\ud are correlated with a distinctive pattern of short-wavelength seafloor roughness, supporting the\ud inference of long-term stability in the location of bubble expulsion sites. As in previous studies at\ud Hydrate Ridge, flares were not detected with the 3.5 kHz echosounder but were clearly imaged at 12\ud kHz. By reprocessing routine shipboard ADCP data, we show that they are also observed in at 75 kHz,\ud indicating that a wide range of bubble sizes is present. The intensity of the flares varied strongly with\ud time. Two primary sources for flares were observed. One, located on the regional topographic high,\ud showed continuous activity, with two times periods of particularly strong flares that are not correlated\ud with tidal height. The other, located on a local topographic high, shows a pulse of increased\ud backscatter that occurred on a falling tide. While the time period of observation is not enough to\ud constrain the effect of tidal changes in seafloor pressure on venting, the data suggest that tides are not\ud the dominant factor controlling release of bubbles from the seafloor. The data support previously\ud suggested models in which temporary sealing of vents by gas hydrate formation and breaking of these\ud barriers as gas pressure builds up is responsible for "burp-like" pulses of gas expulsion. We also report\ud the first observations of flares originating within the HSZ that extend well above the HSZ with little\ud loss of backscatter intensity. These represent a new source of methane injection into the upper ocean\ud and possibly the atmosphere. Flare extension above the HSZ may be due to coating of some bubbles by\ud oil or biofilms or to inclusion of particulate matter (possibly including floating pieces of hydrate) in the\ud plumes. Although this study provides tantalizing new information on both short-term variability in gas\ud expulsion rate and long-term stability of vent sites, longer, well-calibrated observations that integrate\ud bubble flux over entire vent fields as a function of time are needed to develop accurate models for the\ud flux of methane into the ocean and atmosphere from seafloor methane vents