Measurements of convective heat flux are relatively poor, though not from lack of trying. The greatest concentration of scientific effort has been at the Endeavour Segment of the Juan de Fuca Ridge where many investigators have estimated convective heat flux in the past 15 years (Baker and Massoth, 1987; Rosenberg et al., 1988; Thomson et al., 1992; Schultz et al., 1992; Bemis et al., 1993; Ginster et al., 1994). While the reported values are of a reasonable magnitude (~700 to ~12000 megawatts), they vary widely, have uncertainties as large as a factor of five, and do not allow us to address fundamental questions such as whether the flux is steady or varying through time.
Flow Mow takes advantage of technological advances in deep sea autonomous vehicles, such as WHOI's Autonomous Benthic Explorer (ABE), to make precise measurements of convective heat flux. (Photo left of ABE during MAGIC cruise 1996.) We concentrated on the Main Endeavour Field (MEF) of the Juan de Fuca Ridge, anticipating to obtain a value with less than 20% uncertainty. ABE will "mow" a pattern over the rising "flow" of hydrothermal plumes while gathering data on many properties, including water velocity, temperature, salinity, depth, optical index of refraction, eletrochemical potential, and buoyancy. Combining these parameters with precise navigation will yield an accurate value of heat flux through all walls of a box surrounding the MEF. This will be repeated many times to examine fluctuations in tidal currents, vent-fluid output, and average out the variability due to sampling statistics. Simultaneous measurements of overall currents will be made with a current meter mooring located 500m away within the same valley and will be used to refine velocity measurements acquired by ABE.
Vertical Flux. The upper surface was surveyed with 20-meter line spacing 12 times, during seven of which the velocity sensor on the vehicle worked reliably. These seven replicate measurements range from 304 to 615 MW, with a mean of 483 MW, a standard deviation of 103 MW and a standard deviation of the mean of 43 MW. This level of precision was as expected from the analysis of synthetic surveys; accordingly we believe the source flux is steady and the variation seen is mainly statistical. A surprise, when compared to historical observations, was the magnitude of thermal contamination of fluids filling the axial valley, about 0.05 C and showing substantial spatial and temporal variability. This variable datum represents an additional significant contributor to variance in the flux estimate.
Horizontal Flows. We placed a mooring of five current meters ~1 km south of MEF, with meters 50, 100, 150, 200 and 250 meters above the valley floor. Within the proposal body we will discuss the results in greater detail, but for now the currents are tidally-driven, with the deeper meters below the axial valley walls showing net transport to the north and the shallower meters, above the valley walls, showing net transport to the southwest. Within the valley, subinertial flow was steady and northward. Horizontal Flux. We have integrated data from ABE transects on vertical walls around the field, from the current meter mooring and from precisely navigated, vertically oscillating, CTD casts and tows to put bounds on the corresponding horizontal flux of heat. This analysis of observational data has been complemented by development and application of a "puff" model for dispersion driven by oscillatory currents. As expected from this model, we observe a difference between the spatially and temporally averaged thermal anomaly north and south of the field, giving rise to a northward horizontal flux of order 10-100 MW.
Principal Conclusions. Because our measurements came after the 1999 earthquake swarm that resulted in significant changes in the behavior of the MEF, one must integrate these data with older data with some caution. From our data, in August 2000 the total flux of heat from MEF was ~500-600 MW, with 5/6 or more of this transported vertically into the overlying neutrally buoyant plume layer. Since the nature of our approach is to measure the flux across the surfaces of a control volume enclosing the vent field, this estimate integrates all heat emanating from the field, be it of focused, diffuse or conductive origin. As the smoker-style vents within MEF remain active, ~300 MW remains a reasonable estimate of the flux due to this mode of discharge [Bemis et al., 1993; Ginster et al., 1994]. Johnson et al.  estimate diffuse flow from the field to be ~150 MW. Thus we conclude that the partitioning between focused and diffuse flow is ~1:1-3:1, substantially different from the canonical 1:10 value generally attributed to Schultz et al. . Past estimates of heat flux in the area based on steady-state analysis of observations in the neutrally buoyant layer range from 1 to 3 GW [Baker and Massoth, 1987; Thomson et al., 1992; Rosenberg et al., 1988]. Either these include significant contribution from other fields along the Endeavour Segment, a possibility consistent with the measurement techniques employed, or alternatively the output from the MEF has decreased in the past decade.
An equally intriguing result is the magnitude, several cm/s, and steadiness of subinertial currents confined within the walls of the axial valley. We believe that this low frequency component of flow is most likely driven by entrainment of fluid into buoyant hydrothermal plumes.
Flow Mow was supported by the National Science Foundation, Grant Number
OCE-9872090. Any opinions, findings, and conclusions or recommendations expressed in
this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.