快猫短视频

ON 14 January 1993, Stanley Williams led a party of 12 scientists to the summit of the Galeras volcano in Colombia. Although active, the volcano was quiet that day: seismic activity was minimal and the crater was releasing little gas. But as they stood inside the crater and on the rim, the volcano erupted, belching hot gases and debris into the atmosphere for several minutes. The blast killed six of his party and three tourists. Williams survived, but suffered burns, a shattered leg and a serious head injury from flying debris.

The scientists were in Colombia as part of a UN programme to improve monitoring of the world鈥檚 most dangerous volcanoes. By a cruel coincidence, Williams, a volcanologist from Arizona State University, was planning to test a prototype that he hoped would help warn of impending eruptions and perhaps prevent such tragedies. Known as the VGM or volcanic gas monitor, the device is designed to automatically measure the concentrations of certain gases in the atmosphere. The instrument could take readings inside the craters of active volcanoes 鈥� readings that scientists risk their lives for. Williams is also developing an infrared telescope that can measure the concentrations of carbon dioxide in the column of air above a volcano. He believes these devices could help turn the haphazard art of eruption forecasting into a more precise science.

More accurate forecasts are urgently needed: scientists at the US Geological Survey in Menlo Park, California, estimate that, by the turn of the century, 500 million people will be at risk from active volcanoes, most of them in the developing world. The difficulty is that scientists do not fully understand what makes volcanoes erupt. However, they think they know how they are formed. Hundreds of kilometres beneath Galeras, the Nazca plate under the Pacific is sliding beneath the South American continent 鈥� a process known as subduction. According to the most widely accepted theory, the Nazca plate is melting as it thrusts into the rock beneath. The resulting molten rock or magma is less dense and so percolates up through the crust creating volcanoes along the Pacific coast of South America.

Subduction-zone volcanoes make up about 400 of the world鈥檚 500 active volcanoes and are particularly prone to violent eruptions. 快猫短视频s believe this is because they contain magma which is highly viscous and solidifies easily, blocking the path to the surface. Behind such a blockage, the pressure builds up until the solidified magma gives way. The result is a sudden, violent eruption. Other types of volcano contain magma that is less viscous and are therefore less likely to erupt explosively.

The reason why scientists think they can forecast eruptions is because the magma releases gases such as water vapour, carbon dioxide and sulphur dioxide. The huge pressure keeps the gases in solution at great depths but as the molten rock races towards the surface before an eruption, the pressure is reduced and the escaping gas reaches the surface first.

According to this theory, the types and quantity of gas reaching the surface are important. Some gases dissolve more easily than others and will be released from the magma at different pressures. When a large quantity of magma is rising towards the surface the least soluble gases should begin to escape from the magma first. CO₂, for example, is less soluble than sulphur dioxide so it escapes at a higher pressure and therefore a greater depth. In theory, if magma begins to rise from a chamber 15 to 30 kilometres beneath the volcano, scientists would detect a rise in CO₂ emissions. As the magma nears the surface, however, SO₂ should begin to boil off. 鈥淚n theory, the ratio between CO₂ and SO₂ emissions ought to change in the weeks before an eruption,鈥� says Williams.

Rising gas

Volcanologists say that changes in the output of a single gas have signalled eruptions in the past. During a series of eruptions at Mount Etna in Italy in 1977, daily emissions of SO₂ rose from 1000 to 10 000 tonnes 48 hours before each eruption, says Williams. To make better predictions and to fine-tune their theoretical models, volcanologists need to study reliable data on several gases that have been gathered over long periods of time. This is where the VGM can help.

In 1991, Williams began to develop the device with Transducer Research Inc (TRI) an industrial research company based in NaperviIle, Illinois with a grant of over $300 000 from the US National Science Foundation. According to William Penrose, the engineer in charge of the project, TRI faced two challenges in designing the instrument. First, the device had to be lightweight and portable so that scientists could take it into remote areas and carry it up to a crater. Second, it had to be rugged and reliable: an active crater is hot and filled with highly corrosive gases such as hydrogen chloride which forms hydrochloric acid when mixed with water. The instrument would have to survive in this environment for months at a time.

By 1992, engineers had designed and built the first VGM prototype. The instrument contains an electric pump which draws a sample of the crater atmosphere into the device at preprogrammed time intervals. The machine first measures the amount of SO₂ and HCl with two low-power electrochemical sensors that create a current proportional to the amount of gas present. The gas is then pumped through a charcoal filter that absorbs the SO₂ and HCl to minimise corrosion to other parts of the machine. Next, the sample passes through a miniature infrared spectrometer that measures CO₂ concentrations and finally, through sensors that measure the oxygen content and atmospheric pressure. The entire measurement cycle takes 16 minutes. The machine has a double casing made of chemically resistant plastic. The inner case contains the sensors, electronics and data recorder while the power supply and the charcoal filter fill the gap in between.

While Williams was recovering from his accident, one of his students, Tobias Fischer, tested the prototype at Purac茅, another Colombian volcano. The machine was no match for the harsh environment inside the crater, however. Sulphur particles settled on the instrument鈥檚 casing, covering the solar panel that powered the device and preventing measurements from being taken. Also, a pinched O-ring used to seal the casing allowed hydrochloric acid to seep through and eat into metal components. So TRI replaced the solar cells with batteries and coated all metal components between the casings with an acid-resistant epoxy resin.

Lost data

At first, the improved version worked well. In 1993, Fischer collected several hours of data on SO₂ output from the active craters of Merapi and Papandajan volcanoes in Indonesia. But in August last year, disaster struck. Fischer planted the VGM in the Galeras crater and set the device to sample gases 48 times a day, intending to return the next day. In the event, the risk of an eruption kept him away for almost a month. By the time it was retrieved, the batteries had run down and the data had been lost.

TRI has since fitted the device with a radio transmitter that will beam the data out of the crater as they are collected. The 2-watt transmitter, however, has a range of no more than two kilometres so the machine requires a solar-powered relay station on the rim of the volcano to retransmit the signal to researchers.

Penrose says the instrument is designed more like a space probe than an Earthbound device. 鈥淚t has to diagnose any problems and be capable of switching to back up equipment,鈥� he says. For example, the pump that sucks gases from the crater atmosphere is particularly vulnerable to corrosion, so the VGM has a backup. The device now monitors the battery levels and the operation of the pumps, and transmits a status report back to scientists along with the data on gas measurements. The system has yet to be tested but given the challenging environment inside the crater, Penrose is prepared for more failures: 鈥淲e鈥檒l probably lose a few during the testing process.鈥�

If they succeed, the VGM will not be cheap: the cost of the current device is about $30 000 per unit. But it would be cheaper if it was simpler, and measured only SO₂and CO₂ concentrations, for example, or used a less expensive transmitting system which alone adds $12 000 to the cost. Penrose speculates that if the VGM was made in large numbers, its price could drop to as little as $2000. The demand will depend on whether Williams can prove that VGMs can be used to forecast eruptions. But even without this proof TRI believes the technology could be used in other applications: to monitor gases at sites where toxic waste is being processed, for example.

Williams is confident that gases hold the key to eruption forecasting. He is also developing a new way of measuring the amount of CO₂ in the plume of gases above a volcano. Since the early 1970s, volcanologists have measured concentrations of SO₂ in this plume with a device called COSPEC or correlation spectrometer, originally developed to monitor the SO₂ emissions from chimneys. COSPEC is essentially a telescope with a spectrometer that measures SO₂ concentrations in the column of air directly above it. As sunlight passes through the volcanic gas cloud, SO₂ molecules absorb certain wavelengths of ultraviolet light. By measuring the amount of absorption, scientists can estimate the concentration of SO₂ molecules in the cloud. With a number of readings at different sites beneath the plume, volcanologists can work out the mass of SO₂ belching out of the volcano.

This technique has been partially successful. In the weeks before the eruption of Mount Pinatabo in the Philippines in 1992, COSPEC measurements showed that SO₂ emissions had risen to unprecedented levels of up to 15 000 tonnes per day and that seismic activity was higher than usual. 快猫短视频s were able to evacuate a nearby American naval base which was later destroyed by the eruption.

Williams believes that better forecasts could be made if CO₂ levels in the plume are also known. Until recently, however, they could only be measured by analysing samples of gas taken from airborne devices. But flying through the plume is expensive and dangerous, and balloons can be impractical to carry and launch in remote areas. So in 1992, with a grant of $250 000 from the US National Science Foundation, Williams approached the manufacturers of COSPEC, Barringer Research in Toronto, with an idea for a version of the machine that could measure CO₂ concentrations from the ground by looking at infrared light emitted by the gas.

Increased radiation

Measuring CO₂ is more difficult than measuring SO₂ emissions because the atmosphere already contains CO₂ which can mask the readings. Volcanic emissions increase the level of CO₂ only at high altitude and then only by two or three per cent. 鈥淭he problem is that gas molecules absorb radiation at the same frequencies at which they emit it,鈥� says Robert Dick, the engineer in charge of the project. So any extra emissions from volcanic CO₂ at these altitudes can be absorbed by terrestrial CO₂ at lower altitudes. This makes changes in the concentration difficult to detect. The trick is to choose a wavelength that CO₂ emits and absorbs very weakly so that any radiation produced at high altitude is less likely to be absorbed as it travels through the atmosphere. Infrared radiation at wavelengths of about 10 micrometres is the best choice.

But other gases in the atmosphere also emit infrared radiation, says Dick. Engineers at Barringer got around this with a technique known as spectral filtering which splits the light entering the instrument into two beams. The first goes to a detector which measures the total amount of infrared light emitted by the atmosphere. The second beam passes through a chamber of pure CO₂ that absorbs infrared radiation at precisely the frequency at which CO₂ emits it in the atmosphere. The machine then measures the remaining infrared light. Subtracting the second measurement from the first, leaves the amount of infrared radation emitted by CO₂. A similar technique can be used to measure the level of other gases in the atmosphere such as carbon monoxide and various hydrocarbons. But GASPEC, as Barringer has called it, has an important limitation: low clouds also emit infrared radiation so the device can only be used in clear conditions.

Barringer hopes to start testing GASPEC early this year by measuring the CO₂ emissions from fossil fuel power stations near Toronto. CO₂ emissions are closely monitored by other methods at Canadian generating plants so that scientists will be able to compare results.

Batteries included

The machine is also portable enough to carry into remote regions. The first prototype is the size of a large suitcase and can run for up to 30 hours on a 12-volt car battery. It contains an IR sensor which must be cooled by a miniature refrigerator, and a built-in computer which stores the data and compares measurements. But it won鈥檛 be cheap. The parts alone cost $50 000 but the final cost will depend on how many units are built.

Ground-based sensors are not the only way to monitor volcanoes. An international team of volcanologists headed by Peter Mouginis-Mark of the University of Hawaii hope to monitor volcanic gas emissions using the Earth Observing Satellite which is set for launch in 1998. The satellite is designed to measure environmental processes such as ozone depletion and global warming but it will also gather data that are useful to volcanologists, such as the concentration of CO₂ and SO₂. Joy Crisp, a member of the team and a researcher at the Jet Propulsion Laboratory in Pasadena, California, believes that satellite images could revolutionise volcanology. She says previous satellite images of Earth have not been suitable for identifying volcanic rock. The high resolution images of large areas taken by EOS, however, will allow scientists to map the types and amounts of rock ejected in previous eruptions. This gives an indication of the frequency and severity of past eruptions which scientists can use to assess how dangerous a volcano may be. 鈥淭hese images could one day be a standard tool for volcanologists,鈥� she says.

In addition, satellite images offer the chance to conduct wide-ranging surveys of many volcanoes instead of visiting them individually which is a lengthy, labour intensive and painstaking process says William Rose, a veteran volcanologist increasingly involved in remote sensing. Satellite and ground-based monitoring systems are complementary he says. For example: ground-based instruments can get far more accurate measurements of CO₂ output near the ground but satellites have a much better view of volcanic plumes that reach high into the atmosphere.

Williams insists, however, that groundbased technologies such as GASPEC and the VGM have clear advantages over satellites. He points out that most of the world鈥檚 most dangerous volcanoes lie in developing countries which cannot afford to participate in satellite programmes and will not have the expertise to analyse EOS images. Portable gas-monitoring equipment and more basic instruments such as seismometers and tiltmeters, on the other hand, are within the budgets of international aid agencies and some developing nations. As forecasting techniques improve, Williams says it is equipment that could save many lives.