Sensors working overtime

17 Jun 2002

Living in the ominous shadow of Mount Vesuvius, the citizens of Naples know all about volcanoes. Michael Hatcher reports from the southern Italian city on the development of optical sensors that could help to predict future eruptions and possibly even earthquakes.

Volcanologists would benefit hugely from the continuous monitoring of the physical processes occurring in volcanic regions. However, suitably reliable equipment to measure effects such as changing gas compositions has never been available. As a result, monitoring has generally been limited to intermittent field campaigns. Now, with a new generation of optical sensing components emerging, continuous monitoring is becoming a realistic possibility.

That should be good news for people living in and near Naples, because it's not a question of if Vesuvius next erupts but rather when. The next time Vesuvius explodes angrily, Naples will be buried like Pompeii and Herculaneum in AD 79. So Bacoli, a town sitting inside an ancient volcanic crater just west of Naples, was an appropriate setting for optics specialists and volcanologists to meet and discuss ideas at the first international workshop on optical methods in Earth sciences (OMES).The gases that volcanoes spew into the atmosphere are an indication of the processes that are going on deep inside the Earth. Changes in these gas compositions, or an increase in the concentration of an unusual isotope, can, in theory, herald a forthcoming eruption. Unfortunately, however, the chances of the gaseous changes being picked up are very limited because there is no continuous monitoring of volcanoes. Gas measurements are typically made using techniques that require somebody to go out and take a sample of gas from the crater under observation.

Dario Tedesco, a geochemist from Naples' second university, summed this up: "Continuous monitoring is the future," he said in his OMES paper. According to Tedesco, when an eruption is on the cards, unusual isotopes of gases typically found deep inside the Earth start to appear in higher concentrations up on the surface. For example, helium-3 comes directly from the mantle.

Earthquake prediction might also be possible - measurements made by researchers in Japan showed that, before the 1995 Kobe earthquake, there was an increase in chloride ion concentration in the groundwater.

The development of a variety of optical sensing techniques is now making the continuous monitoring of volcanoes and seismic areas a more serious option. Methods with potential for continuous monitoring that were discussed in Bacoli included diode spectroscopy techniques, in situ evanescent-field fibre sensors, quantum cascade lasers and fibre Bragg grating sensors to monitor seismic movements.

However, the volcanic environment is as tough as they come - hot, corrosive, toxic and unstable. The meeting's aim was to increase volcanologists' awareness of the potential that novel optical sensors have, and to give the optics researchers a clearer understanding of the data that both volcanologists and geologists require.

Although accurate predictions of volcano eruptions and earthquakes remain an unlikely goal, continuous monitoring would give Earth scientists far more data to work on than has ever been available before. This should help to improve their understanding of fundamental processes.

Various optical techniques are used in volcano monitoring. These include correlation spectroscopy, Fourier-transform infrared (FT-IR), lidar and passive methods that use the Sun as a light source. Useful though they undoubtedly are - and they set the standard in current monitoring schemes - they are not suited to long-term, continuous monitoring owing to their dependence on clear weather and low humidity. As a result, they tend to be used for finite periods of time on field campaigns.

So many of the techniques discussed in Bacoli were looking to get round these problems. Components originally developed for telecommunications applications have great potential. Frank Tittel from Rice University in the US is a leading exponent of diode-laser spectroscopy, and he and his team have used both diode lasers and fibre amplifiers in developing mid-infrared difference-frequency generation (DFG) sources. Using lasers made by Altitun of Sweden, Tittel measured gases emitted by the Masaya volcano in Nicaragua.

Tittel's set-up includes two fibre-coupled diode lasers, one emitting at 1083 and the other tunable from 814 to 870 nm. An ytterbium-doped fibre amplifier increases the power generated by the fixed-wavelength source. After difference-frequency mixing the two beams in a 19 mm long periodically poled lithium niobate crystal, 2.9 µW of light is emitted at around 3.5 µm. By tuning the diode source wavelength, the emission is widely tunable in the mid-infrared, so it can be used to monitor a variety of gases.The real-time, automated sensor has a spectral selectivity of about 40 MHz and is tunable in the 3.3 to 4.4 µm range. It can detect carbon dioxide, sulphur dioxide, hydrochloric acid (resolving both chlorine isotopes), water and methane - all of which are important volcanic gases - within a response time of 5 to 10 s. Tittel carried out his measurements simultaneously with FT-IR measurements made by volcanologist Clive Oppenheimer of Cambridge University. The techniques are complementary because FT-IR has a lower resolution but a wider spectral coverage and the DFG source's higher sensitivity gives concentrations of the less-abundant isotopes.

Wolfgang Schade from the Technical University in Clausthal, Germany, has developed an evanescent-field fibre sensor for measuring hydrogen sulphide. The evanescent field is the light that does not undergo total internal reflection in the fibre and "escapes" outside. Unlike conventional spectroscopic techniques, using the evanescent field allows absorption measurements to be made in optically thick and highly scattering media, and the small sensors can be placed in awkward spots. Schade has made in situ measurements (figure 1) of hydrogen sulphide with the device at the Solfatara fumarole (the simmering crater left after major volcanic activity) near Bacoli (figure 2).

Owing to the extremely toxic environment, Schade first stripped the polymer cladding from his fibre so that just the core material remained. Sitting in the fumerole, the sensor withstood 10 m/s gas flow streams at about 150 ºC.

Livio Gianfrani from the second university of Naples has also studied the Solfatara gases, using a diode spectrometer to measure the ratio of carbon-13 to carbon-12 isotopes in the emitted carbon dioxide. Diode laser systems might not be necessary for carbon isotope analysis, says Peter Werle of the Fraunhofer Institute for Atmospheric Environmental Research in Germany. He says that German company Wagner Analysen Technik has developed a breath analyser called Iris, which is used for the same purpose in medical applications. This device has a drawback, however - it still requires the manual collection of a gas sample and would be awkward to use in hostile environments.

As ever, research funding is tricky to come by - even Steven Love of Los Alamos Laboratory in the US, who is world-renowned for his FT-IR volcano-monitoring work, says that he is "between projects" at the moment. As he points out, what volcano watchers could probably do with is a major, high-profile eruption to remind the funding bodies that such projects are well worth supporting.

Paolo De Natale of the Istituto Nazionale di Ottica in Florence, who co-chaired OMES with Livio Gianfrani, said after the workshop that bringing together the geologists, optical engineers and physicists to transfer cutting-edge technology to the Earth sciences was a major achievement in itself. This summer, De Natale will be collaborating with Tittel, Oppenheimer and Gianfrani on the first joint campaign under an international project funded by Italy's Gruppo Nazionale per la Vulcanologia.

"The need for a fully integrated volcano-monitoring system emerged from OMES," said De Natale. This will demand a European-funded, multidisciplinary project - an appropriate legacy for the OMES workshop.

OMES workshop www.ino.it/ino/omes