Dohm's team has calculated that enough water could have come from the Tharsis basin to account for the theoretical oceans. But they note several additional potential sources that could have contributed to either an ocean or large lakes.

Equally important, they don't think all water was pushed out of the "hat." What remains within the central Tharsis basin has been dubbed in previous studies a potentially convenient "watering hole" for future explorers, who could drill into Mars to reach the water.

"The southern part of the hat is still depressed and may hold a huge underground aquifer," Dohm said.

The top of the potential aquifer is suspected of being just 360 feet (110 meters) under the surface, based on a separate investigation of asteroid impacts in the region by the University of Central Florida's Barlow.

Other researchers have questioned whether any water remains this close to the surface.

Philip Christensen, an Arizona State University geologist, says it's more likely that if there is any water it would be half a kilometer (1,600 feet) down.

Odyssey's thermal imager is designed to spot minute temperature differences in the landscape, says Christensen, who is the lead scientist for the camera. If an underground aquifer were heated from below by volcanic activity, it might show up. But spotting evidence for anything 100 meters below the surface would stretch the camera's ability.

"I think it would be pretty fortuitous for us to see it that deep," Christensen said. "It could be down there and we wouldn't see it."

The volcanic activity on ancient Mars came in spurts of time and in geographic pockets, the new animation shows, but the global effects may have been huge.

"Violent volcanic episodes triggered short-lived global climatic responses lasting tens of thousands years, including fueling raging floods of water, mud and boulders that ponded to form an ocean and/or lakes, ice and glaciers," Dohm said.

The volcanic activity would have brought heat from the planet's insides to the surface, possibly warming the atmosphere. Dohm doubts the weather was ever tropical, however. He agrees with a view put forth by his colleague Victor Baker that the climate might have been somewhere between a London fog and an Arctic snow.

Though impossible to know for sure, it's unlikely that there were ever placid, long-lived rivers like the Mississippi.

But the researchers note that determining precise dates for the emergence or disappearance of any feature or process on Mars is not possible with current data. The geologic history of Mars can, however, be divided into three major periods, defined by the number of impact craters. Older surfaces have more craters, from back when the solar system was a more chaotic place.

"Until we have sample rock returns from known locations, we can only surmise that the Noachian Period roughly runs from the birth of Mars to between 3.8 and 3.5 billion years ago," Dohm said.

The Hesperian Epoch ended sometime around 1.8 billion years ago. The Amazonian Epoch extends to the present day.

Around the end of the Noachian period and the beginning of the Hesperian, lakes or oceans started forming around much of the rising Tharsis basin, the thinking goes.

Around the same time, a small dent in the landscape presaged what was to become the great, expansive Valles Marineris, which if dug into Earth would stretch nearly from Los Angeles to New York. This region of Tharsis rose too, but not as much as the area to the west, and so it became a gateway for the great floods of an outflow channel system known as circum-Chryse.

"Many planetary researchers have documented floods that resulted in the formation of water and ice bodies in the northern plains, perhaps as far as the north pole, though the northern ice cap and surrounding circum-polar sand seas mask older activities," Dohm said.

Calm periods followed the floods. The surface cooled. Water froze. Dust covered ample frozen lakes or oceans. Then the whole cycle began again as more lava welled up under the ever-growing Tharsis complex.

Across much of Mars, from below the equator to well into its northern plains, the floods fueled by periodic volcanism in the Tharsis region left layers of wet sediment that were then covered by hard layers of lava, which again were covered by subsequent floods in a continuous cycle of destruction and infilling.

Water became trapped in the ground, locked in aquifers composed of loose sand, rock and soil and capped by hard but brittle layers of cooled lava. Cracks in the lava eventually released some of the water again, a process that may still occur today, Dohm and his colleagues speculate.

Nonetheless, Mars appears dry and dusty today. Many scientists expect that the atmosphere is too thin and cold to allow water to remain on the surface. If it gushes out now and then -- and that is not known for sure -- then it would likely evaporate away or freeze solid almost instantly.