This is the cloudiest and windiest place on the planet," said oceanographer Jim Bishop, after the robotic floats were deployed from the research vessel Revelle as part of SOFeX, the Southern Ocean Iron Experiment. "The float in the center of the fertilized patch just keeps sending--its antenna is spectacular. I call it 'the little float that could.'"

SOFeX, a multi-ship project led by Moss Landing Marine Laboratory and the Monterey Bay Aquarium Research Institute, fertilized plankton by adding scarce iron to otherwise nutrient-rich waters. Because phytoplankton grow using carbon dioxide, ocean fertilization has been touted as a way to control global warming.

	Jim Bishop and his colleagues devised instruments that SOLOs could use to measure both organic carbon particles, like plankton, and inorganic, like calcite, the most common carbon mineral in seawater. The map shows SOLO following the iron-fertilized water.

Berkeley Lab researchers had earlier led a collaboration to equip SOLO floats (Sounding Oceanic Lagrangian Observers), designed at the Scripps Institution of Oceanography to measure temperature and salinity at various depths, with particulate organic carbon detectors plus global positioning and fast communications systems. Remotely programmable Carbon Explorers were the result.

During SOFeX the Revelle, with Berkeley Lab researchers Todd Wood, Christopher Guay, and Phoebe Lam aboard, pumped iron into two areas along 170 degrees west longitude. In the "howling 50s" of seagoing lore, the ocean is poor in the silicates some phytoplankton need to form shells, so plankton was expected to bloom sluggishly there even after fertilization. A more vigorous bloom was expected after fertilization of the silicate-rich waters farther south in the 60s.

The northern fertilized region promptly divided into two patches. Revelle launched three floats, the last programmed to stay in the center of the main patch. While the research vessel Melville and the U.S. Coast Guard icebreaker Polar Star later performed follow-up studies, the Carbon Explorers operated continuously and independently, diving and resurfacing each day to report. When high winds interfered with communication, data were saved and played back later.

"The floats were diving to depths as great as 1,000 meters every day, but we programmed them to 'sleep' at 100 meters so that they would follow surface waters best," Bishop explains. There was a risk, however: "By staying shallow in the middle of the plankton bloom, the instruments--which measure particulate matter by how much light is transmitted through the water--might become fouled."

Revelle headed south to fertilize the second patch of ocean; there the final Carbon Explorer experienced similar storms but much-reduced satellite coverage. Heard from only once in the next three weeks, it finally reestablished contact and played back most of its data.

Meanwhile Bishop, using his laptop back in Berkeley, was in daily contact with the northern floats. "Our guess at how to program the floats to stay with the patch, plus a simple change in configuration to reduce biofouling, paid off in a big way," says Bishop. The two floats stayed with the patch, one near its center, the other just outside, serving as a control.

	Deep-diving SOLO floats like the one shown here have been specially equipped to report on biological activity in the sea. A flotilla of SOLOs faithfully tracked a plankton bloom through the stormy Southern Ocean as part of SOFeX, an experiment to test the "iron hypothesis," the proposal that phytoplankton blooms can lower global temperature by removing carbon from the atmosphere.

Just before Revelle returned to the northern patch in early February, guided by the floats' accurate GPS positions, the skies cleared and the ship received the first color image of the area in over three weeks from NASA's SeaWIFs satellite. Despite expectations that growth would be poor, Bishop reported, "we can confidently say that particulate organic carbon in the patch has grown to be four or five times that outside the patch."

Revelle returned to port, but the northern Carbon Explorers kept reporting, despite days when winds exceeded 50 miles per hour and swells averaged 40 feet from trough to peak. By June all three northern floats had traveled over a thousand kilometers.

The lone Carbon Explorer in the south reported periodically until late in the Antarctic autumn, when it was overtaken by advancing pack ice. Its data allowed glimpses of life at 1.5 degrees Celsius below zero. "That's about as cold as the ocean can get, and now it's almost perpetually dark," Bishop remarked.

In mid-June, Bishop reported that after a month of repeated "bonkings" under the ice, the indomitable Explorer reappeared, probably surfacing through leads in the ice. It reported its position and transmitted data stored from over two weeks of diving in the freezing dark.

Long after all SOFeX ships had returned to port, the Carbon Explorers were still on the job, more than half a year since their launch. In July, 2002, Bishop reported "We still have direct control of the Explorers, even though they are half a world away and operating in conditions akin to those experienced by Shackleton and his men." No other approach to investigating the ocean's carbon budget could have matched the performance of these intrepid robots.

The Carbon Explorers were developed with support from the National Oceanographic Partnership Program. Their deployment during SOFeX was supported by the Department of Energy's Office of Science, Ocean Carbon Sequestration Program.

-- Paul Preuss