Undersea mountains provide crucial piece in climate prediction puzzle

A mystery in the ocean near Antarctica has been solved by researchers who have long puzzled over how deep and mid-depth ocean waters are mixed. They found that sea water mixes dramatically as it rushes over undersea mountains in Drake Passage - the channel between the southern tip of South America and the Antarctic continent. Mixing of water layers in the oceans is crucial in regulating the Earth's climate and ocean currents.

Satellites could spot volcanic eruptions in advance

It could be possible to get advance warning of volcanic eruptions by observing them from space.

Seismologists - not poor building standards - blamed for earthquake deaths

Six scientists and an ex-government official have been found guilty of manslaughter in Italy.

New system warns of impending solar flares

Researchers believe they've found a way of predicting solar flares more than a day before they occur, possibly helping to protect satellites, power grids and astronauts from dangerous radiation.

Chemical analysis can predict volcanic eruptions

Examining crystals formed deep within volcanoes could give a year's warning of impending eruptions, say scientists.

Team predicts future regional temperatures for North America

The first detailed predictions of future temperatures across North America have been produced, using a combination of climate models and some complicated statistics.

Gene mapping often fails to predict disease

Whole genome sequencing as a method of predicting disease risks isn't as informative as thought, a study of thousands of identical twins has found.

Twitter use reveals more than you think

Your friends' tweets may be giving away more information about you than you realise, say sneaky computer scientists at the University of Rochester.

Algorithm predicts pop chart success

A team of British scientists reckon they can predict the success of a pop song with abut 60 percent accuracy using a machine learning algorithm.

Night lights used to track diseases

Satellite images of city lights can be used to monitor outbreaks of disease, says a team led by Princeton University.

Seismologists in court over Italian earthquake

Six Italian scientists and former government officials are to appear in court today over charges that their failure to predict an earthquake led to hundreds of deaths.

Supercomputer forecasts political unrest

A supercomputer at the University of Tennessee could have predicted the location of Osama Bin Laden to within 125 miles through analysis of the mood of international news stories and the geographical locations they mentioned.

Team drills to the heart of San Andreas Fault

An international team of scientists and engineers has drilled down a mile and a half into the epicenter of an earthquake.

Acoustic monitoring predicts dangerous solar flares

Predicting solar flares is going to become increasingly important as the sun hits the most active phase of its 11-year cycle.

Scientists move closer to mind reading

Researchers at the University of Western Ontario say they can tell what action a person is planning, moments before they actually do it.

New tool helps game designers predict player behavior

A team at North Carolina State University says it can tell what gamers are about to do next, allowing games designers to steer players to the parts of games they'll like most.

Team dramatically improves US flood and drought prediction

A Columbia Engineering School team says it's made a major step in improving forecasts of extreme weather events in the US.

Google searches predict dengue outbreaks

After discovering that search queries about flu could predict outbreaks effectively, Google's launched a similar tool to help health officials respond more quickly to outbreaks of dengue.

Ford teams with Google for cars that can read your mind

Ford is using Google's Prediction API to try and predict driver behavior and optimize performance.

Twitter analysis predicts stock prices a day ahead

Playing the stock market just got a little easier, thanks to economists at the Technical University of Munich (TUM) who have developed a website that predicts individual stock trends.