Keck Expedition 2004
September 3, 2004 Day 5
Update for September 3, 2004—Endeavour Dive 1, Valley around Salty Dawg
By Deb Glickson on the Western Flyer
Scientists usually ask a lot of questions and the Tiburon pilots generally get asked the most annoying ones. On our cruise, the most common question is probably “We have time for one more sample, right?” followed closely by “We have room for one more sample, right?” Well, not me. The question I usually find myself asking is, “Hey! Are we really where we think we are?”
One of the goals of this cruise is to integrate the geochemistry of each Endeavour rock sample with its geological environment. This will answer questions about the nature of the magma chamber beneath the segment, the patterns of lava eruption on the seafloor, and how the axial valley and the segment may have evolved over time. Two components are needed in order to do this accurately – good navigation and great bathymetry.
The first component, good navigation, is generally pretty easy with the Tiburon. They use a navigational system called Ultra-Short Baseline, in which the ship uses acoustic signals to locate the ROV on the seafloor. At Endeavour it’s kind of hard to get good navigational positions because the seafloor is very rough. Lots of bare rock, ridges, fissures, and faults tend to get in the way of the acoustic navigation. Most of the time, this system works to within about 15 meters, which means that the ROV could be smack-dab on the place you want to be or up to 15m away. That’s pretty good, really.
This leads to the next important component, bathymetry. Historically it’s been hard to map the seafloor at high resolution because it requires the use of a submersible, ROV, AUV, or towed vehicle to map very near the bottom. It also requires a lot of time, because these systems can only map small pieces of the floor during each dive. It’s faster and easier to map the seafloor using a sonar system mounted on a ship, but the resolution is not as good since you are further from the seafloor. The whole Endeavour Segment has been mapped at the highest resolution possible with a ship, which means about 30-50m. This means that any seafloor feature smaller than 30-50m, like a rock outcrop or hydrothermal chimney, cannot be seen on the map. It would be like looking at a map of your neighborhood – all the houses would be distinguishable, but the trees, cars, and people wouldn’t show up. However, many parts of the axial valley at Endeavour have also been mapped using ROVs and AUVs, and those pieces of the seafloor have a resolution of 2-3m. Using the same analogy as before, trees and cars would start to appear on your neighborhood map, although people still wouldn’t. For the seafloor, this is pretty amazing!
For today’s dive we did four zig-zag’s across the
northern valley floor. For part of our dive we combined our good-to-15m
navigation with our 30-50m resolution data and had a pretty good idea of where
we were and what we were looking at. But
given the topographic roughness of this area, we were constantly getting
confused by connection between the smaller features such as lava lakes and
faults. For the other part of the dive, we had the 2-3m resolution data, and
were able to pick out these more interesting features that we couldn’t have
seen on the other map. This became
really clear on the transects today. Based on the lower-resolution bathymetry,
we thought there was a large, broad ridge in the middle of the valley. When it was examined more closely on the high-resolution
data, we saw that there were actually two peaks on the top of the ridge,
separated by a small saddle. This helped determine the position of 
the
ROV, as we were able to pinpoint our location even though steep topography
caused inaccurate navigation. This figure is an example of the two levels of
bathymetric data from the area crossed by today’s dive. The normal ship-based
bathymetric data is on the lower portion of the image and it accurately shows in
a golden brown the axial ridge. This is the data that we use to navigate most of
our dives and we think its great. However, look at the new data on the upper and
right-hand side of the image! The broad axial high has turned into two distinct
ridges. No wonder we were confused.
Brief
synopsis of today’s dive by Debra Stakes. Because so little of the Endeavour
geology has been examined in detail we had a hard time deciding on priorities.
However, based on the earlier data and the availability of the great bathymetry
from Deb Glickson, we decide that the first order of business is to look at the
spreading axis for the distribution of volcanic features. We crossed the
northern four kilometers of the spreading axis five times, picking up rocks
along the way. 
Everything
was much bigger in scale than what we had seen at Cleft. The faults were more
offset, the collapse pits were more extensive, and the hydrothermal deposits
more abundant. We actually stumbled across a new low temperature vent and two
inactive deposits in our one dive (see top
right). We were surprised by
how much of the deeper axis floor was filled with sheet flows and drained lava
lakes (see left). Such features
indicate large volumes of lava being erupted to pond and then drain away. These
drainage channels typically have tall pillars where steam or water was
discharged. This is a picture of a very small pillar that shows the layers (see
bottom right). 
Where
the lava lake cools more completely you see thick massive layers of lava. This
one seems to be surrounded by younger pillows. How do you think that it formed? 
The plan for tomorrow’s dive is to do one long transect across the entire axis including both flanks and the axial valley. We hope that this will let us see the transition from the valley to the flanks to understand the age relationships here also.