LONDON - The mysterious disappearance of an Air France jet while flying over the Atlantic in fierce thunderstorms is stirring a debate about whether new technologies and procedures are needed to prevent similar tragedies.

Experts say it will be difficult to build a better plane than the Airbus A330 that plunged into the ocean last Sunday, killing all 228 aboard. But they see room for improvement in other technological areas that could help boost safety.

One idea is to move from radar to satellite surveillance systems that would allow air-traffic controllers to track a plane's progress on flights across the ocean. Currently, planes go out of radar range after 200 miles from land.

Another focus is on how to make black boxes - the flight data and cockpit voice recorders - more easily recoverable in an ocean crash.

One solution being discussed in aviation circles is wiring the black box to make it stream data to help air-traffic officials locate the box and the wreckage.

"The black box could be set up to send an immediate message that could give the parameters of the plane," in a similar way that Flight 447 put out a burst of automated messages detailing mechanical failures, said Michael Boyd, an airline analyst in Colorado.

The Air France flight from Rio de Janeiro to Paris disappeared nearly four hours after takeoff, in the world's worst commercial air accident since 2001.

The disaster has prompted calls for the United States and other countries to hasten the move to GPS-based networks that would pinpoint planes and enable controllers to monitor them as they cross the ocean outside radar-range.

"It does seem a little disconcerting for the public who have not been familiar with the lack of surveillance in oceans," said Bill Voss, president and chief executive officer of the Flight Safety Foundation in Virginia.

Nearly 70 percent of the world's airspace is not radar-controlled, and the existing radar system is likely to remain for at least a decade.

While some European and Asian countries are moving toward satellite systems, which would reduce travel times and fuel use by helping the pilot find the most efficient route, a huge obstacle is expense.

In the United States, technology for such a system is being tested, but full implementation - estimated to cost $35 billion - has languished amid funding delays and disputes over technical complexities.

Some elements of these essentially GPS-controlled systems already exist but are not in widespread use.

Major carriers are already capable of using automatic dependent surveillance-broadcast (ADS-B) technology, in which the plane emits data that show up on the controller's screen. But the overall infrastructure is not yet in place to allow its general use.

Voss believes that being able to better communicate with aircraft is more important than surveillance for safety.

Passengers may be able to use cell phones on a flight, but the pilot may be relaying information via VHF - which has been standard in aviation for at least 60 years. When crossing oceans, pilots communicate with air-traffic control if necessary via high-frequency radio, which is prone to interference from sun spots and lightning and can be difficult to hear.

"This crash may put more pressure on international organizations to advance the use of satellite voice communications," said Voss.

One key factor in figuring out what went wrong on Air France Flight 447 is finding the black boxes. But they could be scattered nearly anywhere across a vast undersea mountain range, throwing retrieval efforts into doubt.

Boyd said the black box might be configured to automatically send messages out every 10 minutes or so, he added. "I think we're going to go in that direction now."

But new black box technology may be held up by cost, the rarity of ocean crashes, and the aviation industry's culture.

A plane crash over the sea is "very rare," said Boyd, noting the last such accidents to happen were in the 1980s.

"Most airline accidents happen on landing or takeoff. You always find a black box there," said Boyd.

Furthermore, "the aviation industry is extremely conservative in accepting new technology," according to Voss.

Chris Yates, Jane's aviation security editor, rules out black boxes that would float, saying the shell would have to be so flimsy that "in the event of a crash it would automatically break up and we would lose that data."