Leg 3 began promptly at 0800 as we cast off from the pier on Sand Island, and proceeded to exit Honolulu harbor and head round the eastern end of Oahu towards the windward side of the island. 

We began our first ROV dive around noon in a submarine canyon on the northeast side of Oahu. The purpose of this dive, and most of the dives on this leg, was to investigate the processes associated with the formation and evolution of the submarine canyons around Hawaii. Since these canyons cut into the island margins, we can also learn about the structure and history of the islands by looking at and sampling the rocks exposed in the canyon walls. We are also looking for evidence of fresh water from the islands seeping out of the seafloor. This sapping can be an important cause of erosion, and is often associated with the formation of submarine canyons. We can detect the presence of even small amounts of fresh water seepage because water that filters through land contains trace amounts of radium, while seawater contains essentially no detectable radium. On every dive the ROV pumps water through filters designed to capture radium. These filters are then analyzed overnight. Although our primary goals are geological, we do also take note of the creatures we observe on and near the seafloor, and we follow and observe particularly interesting specimens.

On the first dive, we explored the upper part, or head, of a submarine canyon on the windward (rainy) side of Oahu. We found that this canyon head is largely eroded into a thick section of mud. The canyon walls are nearly vertical in places, and have regularly spaced grooves every few meters. We did not see any clear evidence for fresh water seepage, and the radium measurements later confirmed that little groundwater is seeping out in this canyon. At one point, we found that water would flow out of cracks in the canyon wall every time the ROV pressed up against the wall. It is likely that this section of the canyon wall is near failure, and a landslide may occur soon. The water in the canyon contained more particles, called marine snow, than on any of the dives of the last leg. Hawaiian deep waters contain much less organic material than is usual in Monterey Bay. Submarine canyons may tend to concentrate marine snow because material moving downslope toward the abyss tends to be funneled through canyons.

This picture shows a screen dump of the computer program used to navigate our ROV dives. This program displays the location of the ship (red) and ROV (blue) on a map of the seafloor. Here the contours show the water depth. The contours are separated by ten meters in depth (about thirty feet). The canyon walls are steep where the contours are close together.

We took a number of sediment cores and rock samples. This picture shows one of the more interesting rock samples. This is a cemented sandstone that initially appeared to be coral because it is narrow and branching. It turned out to be a hollow tube of sandstone. The sandstone may have become sedimented around the burrow of some creature living in the seafloor.

This picture shows an octopus (Berrya) living 200 meters deep in the canyon. Fine layering is observed in some of the rocks at the base of the canyon walls.

This shark was inordinately curious about our ROV, and came in close to give the camera a bump.

Here we see prawns feeding on some seaweed that has been transported down the canyon. Canyons may be an important avenue for nutrients to move from coastal waters to the abyss.

This is my first research cruise on board the Western Flyer, and in fact my first real experience in using ROV’s like Tiburon to do geology on the seafloor. I have spent months at sea, but its generally involved collecting seismic or seafloor mapping sonar data. Those sorts of geophysical cruises entail hours of standing watch while instruments collect data, and then months afterward processing gigabytes of data. This cruise involves a much more interactive (well, geological) approach to science. Here we get up early in the morning, spend all day exploring a specially chosen patch of seafloor in great detail from the dark confines of the ROV control room, and then spend the evening cataloging and preparing the samples (rocks and sediment cores) brought back by the ROV. I am astounded by how much our efforts resemble traditional, subaerial geological mapping. We use the ROV to traverse up and down exposures. When we need to know what kind of rock or sediment is exposed, we sample it. If we just want to know if the bottom is hard or soft, we ask the ROV pilots to poke it with the ROV’s manipulator arm. If we see something of interest, we go look at it close up with the video cameras. If we want to know what rocks are exposed somewhere else, we ask the pilots to take the ROV over there, and they do. The ROV brings us face to face with the seafloor, and yet every decision reflects the collective knowledge on board, because we’re all there in the control room with access to all of the previously collected data. In short, the ROV really is the most efficient tool imaginable for doing submarine geology.

I would like to send a special greeting to kids in Mrs. Fletcher’s K-1 class and Mrs. Apis’ 4th grade class at the Spreckels School in Spreckels, CA.

Cheers,
-David Caress

21 April 2001

Dive 299 was located in the lower (400-meter deep) headward part of the main submarine canyon that heads at the mouth of Koneohe Bay, just north of Makupu Point. The objective of this dive was to determine the origin of the canyon, to seek evidence on the process that created the canyon. This canyon is one of many canyons that head along a bathymetric amphitheater that formed when a large submarine landslide occurred approximately 1.5 million years ago. These canyons are also located on the wet side of the island of Oahu and offshore of a heavily dissected mountain slope eroded by the large amount of runoff that occurs in this area. We speculate that groundwater may play a role in the creation of offshore landslide and canyon development. Four rock samples and seven push cores were collected, water was continuously sampled for radon and two Golflow bottle samples were collected.

Observations and sample collection indicate that the canyon is eroded into basalt and volcanoclastic rocks. In places along the lower wall, cracks were observed in a well-layered mud wall and, when bumped by the ROV, water would squirt out along the crack. This hydraulic pumping action may be a major process in canyon wall erosion. A stimulant such as an earthquake could create a pumping action that could lead to failure of material along the crack thereby reducing support for the wall material and creating a local landslide.

Although evidence of recent down-canyon transport of sediment was not found, the sharp well-exposed and well-layered walls indicate that recent erosional events had taken place in the past. One indication of recent, although minor, sediment transport in the canyon was the presence of grooves cut into a soft mud-covered slope along one wall of the canyon. These grooves appear to have formed from sands and coarser materials sliding down the slope. In one groove we collected a sample of what appears to be a cemented bioclastic rock that may have formed from fluid venting. A core located in one of the grooves recovered stiff mud.

-Gary Greene (Chief Scientist)