MBARI Ridges 2005 Expedition
Juan de Fuca Leg: August 7–18, 2005
Gorda Leg: August 22–September 2, 2005
August 24 update - from David Clague
Tiburon dive 885 - 1996 lava flow on Gorda Ridge
Today we dove with ROV Tiburon on a 1996 lava flow on the northern Gorda Ridge. This lava flow was created during a seafloor eruption that was initially detected by the Navy SOSUS hydrophone array as a swarm of earthquakes beginning on February 28, 1996. (The SOSUS array also detected recent eruptions at several of the volcanic sites that we visited earlier on this cruise, during the Juan de Fuca leg). Ten days after the earthquake swarm began, the NOAA ship McArthur made a rapid-response cruise to the area. Using water-column sampling, they located a plume of warm, gas- and particle-rich water. About five weeks later, the UNOLS ship R/V Wecoma returned to the area and continued water-column chemical surveys to locate the source of the plume.
A few months later, a third event-response cruise on the NOAA ship Discoverer recovered a float that had been inserted into the eruption plume to track its movement through the water. The following summer, the R/V Laney Chouest used the ROV ATV and the R/V Thompson used the ROV ROPOS respectively to map the lava flow. They also collected samples, which demonstrated that an eruption had indeed taken place at this site. The site has not been revisited since, although we (MBARI) dove in this general area using the Western Flyer and ROV Tiburon in 2002, in an unsuccessful attempt to locate a second hypothesized eruption site about 10 km to the south.
Our dive today was aimed at the northern part of the 1996 lava flow, where a photograph from a towed-camera used on one of the previous cruises showed black sand on old pillow talus. It was our hope that this deposit might be pyroclastic in origin, showing that the eruption was more energetic than generally thought.
We started to the east of where the 1996 flow had been mapped, and landed on a young flow with few animals and little sediment, which in retrospect was actually the 1996 flow itself. We crossed to the northwest, eventually encountering older flows of pillow basalts that were rich in animal life and covered with thick sediment. Later we crossed back over the 1996 flow several times, and eventually located the previously reported gravel and sand deposits. They lie beyond an extensive talus slope at the base of a 90-m-tall overhanging cliff composed of elongate and truncated lava pillows. The gravel and sand deposits proved to be mainly glass rinds chipped from the pillows that cascaded down the talus slope. The vertical cliff of pillows and the talus slope at its base are constructional features that formed during the initial eruption, rather than by subsequent tectonic faulting or landsliding.
One of the most exciting parts of the dive was when we encountered a deposit of elongate log-shaped lava fragments, some up to at least 3 meters long, lying on the pillow talus. The pilots skillfully collected some of these "lava logs". We concluded that they were created when lava cascaded off the overhanging cliff and formed lava stalactites, which then tumbled to the talus slope below. For reasons we are still trying to understand, these "stalactites" remained intact rather than smashing when they hit the talus below.
The construction of this tall, steep stack of pillow lava apparently resulted from extremely low eruption rates, and also apparently produced few limu o Pele bubble fragments, although most of our samples remain to be examined.
We have been using two techniques to collect samples of millimeter-sized bubble fragments. In places where the sediment is thick enough, we use sediment push-cores. In areas of thin sediment cover, we use a small, glass suction sampler. The suction sampler was originally built to quickly sample glass rinds on large bulbous pillow lavas, where it is time-consuming to search for a rock the manipulator can handle. But we have found that we can also use this suction sampler to "vacuum" thin sediment deposits to collect samples of limo o Pele fragments.
After taking a quick look at samples from the 1996 eruption site, we know that they contain some limu fragments. However, we won't know if this limu formed actually formed during the 1996 eruption until we chemically analyze the glass fragments using an electron microprobe.
We have already learned that the boundaries of the 1996 flow are only roughly located and that the eruption produced some novel lava forms, although few pyroclastic particles of glass. Tomorrow, we will dive on the southern part of the flow, which is virtually unexplored.