or collective redistirbution of any portion of this article by photocopy machine, reposting, or other means is permitted only with the approval of The Oceanography Society. Send all correspondence to: [email protected] ofor Th e The to: [email protected] Oceanography approval Oceanography correspondence all portionthe Send Society. ofwith any permitted articleonly photocopy by Society, is of machine, reposting, this means or collective or other redistirbution This article has This been published in SPECI A L Iss U E O N O C E an E XPL O R ATI on Oceanography Archaeological journal of The 20, NumberOceanography 4, a quarterly Society. , Volume Oceanography B Y Rob E R T D . B A LL A R D F O R T H O U sands of Y E A R S , waters because compressed air becomes highly maneuverable submersibles in ancient mariners traversed our planet’s a deadly gas at less than 90-m depth. the 1960s, but there was still a lack of waters; unfortunately for them, many Recent advances in exotic gas mixtures precision manipulation and inabil- of their ships have been lost along the and the development of saturation div- ity to carry out mapping efforts to any C opyright 2007 by The Oceanography Society. All rights reserved. The 2007 by All rightsopyright Oceanography Society. way, carrying their precious cargo and ing extend and lengthen a diver’s capa- degree of sophistication. Though George the history it represents to the bot- bility. But, these advances have been Bass (1975) pioneered the use of this tom of the sea. For hundreds of years, primarily limited to use by the military class of small submersibles to photo- attempts have been made to recover and the professional diving communi- mosaic a Roman Byzantine shipwreck their contents. In Architettura Militare ties. Although in the last few years some off Yassi Ada, Turkey, his idea did not by Francesco de Marchi (1490–1574), archaeologists have begun using exotic catch on within the marine archaeol- for example, a device best described gas mixtures, even then they are limited ogy community, which preferred the as a diving bell was used in a series of to less than 100 m of water, which repre- dexterity of human hands to excavate attempts to raise a fleet of “pleasure gal- sents less than 5% of the seafloor. ancient shipwrecks. leys” from the floor of Lake Nemi, Italy, The oceanographic community, While marine archaeologists contin- in 1531. In Treatise on Artillery by Diego however, has perfected the technology ued to rely upon divers to carry out their P ermission is granted to copy this article for use in teaching and research. article use for research. and this copy in teaching to granted is ermission Ufano in the mid-1600s, a diver wearing needed to work in world ocean depths work, marine geologists were perfecting a roughly fashioned hood and air hose of that average 4,000 m (Ballard, 2001a). the use of manned submersibles. At first, cowhide is shown lifting a cannon from In the middle of the last century, ocean- they, like marine archaeologists, con- the ocean floor. But, these early efforts ographers developed one-atmosphere, centrated on the shallow depths of the to recover lost cargo from sunken ship- manned vehicles that were initially crude continental shelf. Francis Shepard (1964) wrecks were crude and highly destructive and had only the simplest manipulative and his students at Scripps Institution of salvage operations. capabilities. The bathyscaphe, for exam- Oceanography began exploring nearby Only relatively recently have archae- ple, conquered the deepest ocean depth submarine canyons using SCUBA diving ologists taken an interest in such ship- in 1960 when it descended 11,000 m techniques but later turned to Jacques P wrecks. Marine archaeology is a young into Challenger Deep off the Marianas Cousteau’s Souscoup to explore their O Box 1931, research field with its roots deeply Islands. But bathyscaphes had limited deeper depths. Like the bathyscaphe embedded in the technology of SCUBA horizontal mobility and poor visibility. before it, Souscoup was used primar- R ockville, MD 20849-1931, R epublication, systemmatic reproduction, reproduction, systemmatic epublication, (self-contained underwater breathing They were basically elevators to the bot- ily as an observational platform and apparatus) diving that emerged in the tom, capable only of returning shortly had limited manipulation and crude early 1950s (Ballard, 2001a). However, after they arrived. mapping capabilities. SCUBA is limited to relatively shallow Bathyscaphes were followed by small, A major breakthrough program U SA. 62 Oceanography Vol. 20, No. 4 in marine geology—one that would of a diver’s hands. precision, measured in centimeters every also ultimately have a major impact In the late 1980s, it became appar- few seconds, which made it possible to on marine archaeology—was Project ent to some oceanographers that much place the vehicle in automated “closed- FAMOUS, the 1974 French-American manned submersible work could be con- loop” control, driven by a computer Mid-Ocean Undersea Study (Ballard, ducted by more efficient, cost-effective instead of a human operator. A few years 2001a). FAMOUS marked the first research platforms. To reach 4,000 m, later, SHARPS was replaced by a wireless manned exploration of the mid-ocean scientists spend two and half hours in system called EXACT, which now makes ridge and the first use of manned sub- the morning to make their descent to the it possible to precisely control an ROV mersibles for comprehensive mapping bottom, followed at the end of the day in any water depth, including down to of complex geologic terrain. Prior to this with a similar ascent to the surface. Thus, 6,000 m, now a common ROV operat- program, submersibles had been used they have only three or so hours to work ing depth—a depth that encompasses as exploratory tools, but due to the lack on the bottom, typically covering one 98% of the world’s ocean floor. Equally of precise knowledge about the terrain nautical mile of underwater terrain. important, scientists not only know in which they were operating as well as The advent of fiber-optic cables, also where the ROV is every few seconds, they the lack of accurate navigation, their use in the 1980s, made it possible to develop know the position of its various sensors, had been ineffectual, diminishing their a remotely operated vehicle (ROV) that sonars, and cameras. value in mainstream oceanographic could transmit high-quality imagery to programs. During Project FAMOUS, the surface and create “tele-presence” HAMILTON and SCOURGE manned submersibles were used in a on the ship’s deck. With this technologi- PROJECT more surgical fashion. cal leap forward, scientists were able to The first major application of this inte- Before the three FAMOUS submers- deploy an ROV for long periods on the grated technology for marine archaeo- ibles began their dives, the region to bottom while controlling its operation logical purposes was the 1990 survey be studied—the rift valley of the Mid- onboard ship. Critical to the success of of the Hamilton and Scourge in Lake Atlantic Ridge—was intensely surveyed this technology, however, was the abil- Ontario. These two American ships using a broad array of mapping devices, ity to precisely control both the surface sank in a violent storm during the War including multi-narrowbeam sonar ship and the ROV thousands of meters of 1812. They were identified in 1975 systems, side-scan sonars, and a vari- below. Control of the ship was made and later given to the Canadian city of ety of imaging systems. By the time the possible through the development of Hamilton by President Carter; the city submersibles were deployed, the scien- dynamic positioning systems coupled wanted to know their precise condition tists had acquired detailed topographic with the use of GPS navigation. Precise and whether these ships should be raised maps of the rift valley and developed control of the ROV was more difficult or left where they were. significant understanding of its geol- because, unlike GPS that had a broad Answering these questions required ogy. The submersibles were able to work user base, there was little interest in such a detailed survey of the ships to a preci- within this accurate database, examin- precise vehicle control. sion not previously possible. Installed ing specific targets of importance to the In the early 1990s, major new technol- on a barge anchored over the wreck site, scientific team. Even so, their mapping ogy closed the gap between the marine the SHARPS system tracked ROV Jason technology and acoustic navigation sys- geologist and the marine archaeologist operating in “closed-loop” control to tem limited their resolving power to a when engineers interested in precision record data from which detailed opti- few meters at best, clearly short of the control of ROVs developed the SHARPS cal and acoustical mosaics and three- mapping precision of marine archae- tracking system (Ballard, 2001a). The dimensional images were constructed. ologists, who were accustomed to cen- fact that this system was wired severely About the same time that this coop- timeter accuracy. And, once again, the limited its initial use to test tanks or shal- erative program between marine archae- manned submersible manipulators were low water. What was important about ologists and oceanographers accom- crude when compared to the dexterity the SHARPS system was its tracking plished these advances in precision Oceanography December 2007 63 Over the next nine years, nine Roman trading ships (Figure 1) dating from the first century BCE to the fourth century CE were found within this corridor of ancient debris as well as two long lines of amphorae that were obviously thrown overboard by ships hoping to save them- selves in storms by jettisoning their cargo (Ballard, 1998; Ballard et al., 2000).