All forms of asbestos (various serpentines and some amphiboles), but especially the blue variety, are carcinogenic because their dusts consist of minute fibres. Most publicity about the hazard that this mineral presents is from cases that stem from its use as an insulator in housing, shipbuilding and other constructions in developed countries. Areas where it has been mined or outcrops naturally are equally risky if wind can pick up asbestos dust under dry conditions. A large proportion of this now banned industrial mineral was mined in South Africa and many cases of asbestosis and mesothelioma in former mining areas have come to light there since the fall of apartheid. The locations of former asbestos mines are well known, and some attempts are being made to bury the waste. The most tragic cases are where the mining companies have either folded or been engulfed by larger transnational corporations; several legal actions for compensation have been dragging through the courts for a decade or more. However, asbestos minerals are common at what were non-commercial levels in many ultramafic rocks. Such rocks occur in ophiolite complexes and Archaean greenstone belts on every continent, and although ultramafics are in a minority as regards rock outcroppings, they are far from rare. In its natural state such land can shed asbestos-rich dust when dry, and urban and communications developments expose the material to wind action.

Asbestos minerals fortunately have distinctive infrared spectra in the short-wave infrared (SWIR), preferentially absorbing photons at around 2.3 micrometres because of their abundance of magnesium-oxygen bonds that such wavelengths cause to vibrate. Remote sensing is therefore a potentially useful means of screening areas of human habitation for asbestos risks (Swayze, G.A. et al. 2009. Mapping potentially asbestos-bearing rocks using imaging spectroscopy. Geology, v. 37, p. 763-766). The authors, from the US geological Survey and the California Department of Conservation, used a sophisticated and costly form of aerial remote sensing that covers the visible and infrared part of the EM spectrum with hundreds of narrow-wavelength bands: so-called hyperspectral imaging. It is possible to highlight areas containing asbestos minerals by matching the measured and mapped surface spectra with laboratory standard spectra of the pure minerals. In the case of the test area in northern California, where suburban expansion is likely to occur or has done already, the geology is known in some detail and the expensive airborne hyperspectral surveys could be focused. The approach gave results sufficiently accurate for preventive measure to be taken; not only for asbestos-rich bare soils, but also the specific kind of vegetation that ultramafic soils encourage.

There is another, far cheaper means of assessing asbestos risks that is not so accurate, but capable of covering very large areas of poorly known geology, especially in less well-off parts of the world. This uses the satellite remote sensing conducted by the US-Japanese ASTER instrument carried on NASA’s Terra satellite. ASTER data include 5 narrow wavebands that bracket the 2.3-micrometre part of SWIR, so that it is capable of assessing the distribution of ultramafic rock outcrops using software similar to that for hyperspectral data. The USGS/California DoC survey could have tested ASTER data to see how effective it would be if more costly airborne data was unaffordable. Sadly the team didn’t foresee how a local test of concept might benefit a great many areas elsewhere by using an ASTER scene that would cover their entire study area, be free to USGS scientists and cost only US$85 for anyone working in the Third World.

Nuclear waste: planning blight writ large

The artificial radioactive isotopes generated in nuclear fission reactors have half lives that range from days (¹³¹I) to a few million years (¹³⁵Cs). They pose a thorny problem for disposal since the radiation that they emit collectively is likely to reach ‘safe’ levels only after tens to hundreds of thousand years, even if they were diluted by leakage into air or water or onto the land surface. They have to be contained, and that demands storage in rock. More over, underground disposal sites must ensure no leakage for geologically significant periods – a great many rare events, such as magnitude 9 earthquakes, large volcanic upheavals and rapid climate changes all become increasing likely the longer the delay time. Apart from Sweden and Finland, no country that uses nuclear energy has a deep disposal site. The focus has been on the temporary measure of reprocessing, and one major facility, that at Sellafield in the UK, is to close down.

In 1987 the US Congress designated only one potential site for investigation as a place for long term water storage in their vast, geologically diverse country: Yucca Mountain in Nevada. The reasoning was that the area is remote and arid, and not so far away from highly secure military sites, so it could be guarded unobtrusively. After 30 years of investigation, Yucca Mountain has been abandoned, with no equally-well researched fallback site (Ewing, R.C. & von Hippel, F.N. 2009. Nuclear waste management in the United States – starting over. Science, v. 325, 151-152). From a geological standpoint, that is not so surprising as Nevada is seismically active; there has been volcanism in the not-so-distant past, it does have groundwater, and that is present in the volcanic ash proposed for storage. Moreover, the water is oxidising and uranium in spent nuclear fuel easily dissolves under those conditions – storage was to be in titanium casks. Clay saturated in anoxic water is a better bet, while the Scandinavian approach seems safer still: galleries and boreholes in dry crystalline basement rock with canisters packed in clay.

Yucca Mountain has been wrangled over for 3 decades, and one component in its abandonment was a change in the proposed ‘regulatory period’ from 10 thousand to a million years. How compliance might be demonstrated for a period five time longer than our species has existed, and 500 time longer than the length of the Industrial Revolution is something of a problem for bureaucrats, as of course is judging the cost and time for decommissioning obsolescent nuclear plant. If nuclear energy is to play any role in cutting carbon emissions, the volume of nuclear waste is set to rise enormously, but this does not seem to concentrate the regulatory group mind wonderfully.

See also: Wald, M.L. 2009. What now for nuclear waste? Scientific American, v. 301 (August 2009), p. 40-47.

Methane: the dilemma of Lake Kivu

A massive discharge of carbon dioxide from the small but deep Lake Nyos in Cameroon in 1986 killed 1700 local people after a small earthquake and landslide disturbed the bottom water.  The lake is stagnant, and carbon dioxide released by exhalation from deep magma chambers beneath it had dissolved under pressure in in deepest levels. Once disturbed, the gas came out of solution to reduce bottom water density so a large volume rose to blurt out gas and deal silent death in the lake’s immediate surroundings.

Lake Kivu in the western branch of the East African Rift system borders the Democratic Republic of Congo (DRC) and Rwanda. With an area of 2700 km² and a depth of over 400 m it is far larger than Lake Nyos, but similar in having stagnant water below a depth of about 75 m, in which gases are dissolved under pressure. Lake Kivu contains an estimated 256 km³ of carbon dioxide derived from magmas beneath the Rift and 65 km³ of methane that probably arises by anoxic bacterial reduction of the CO₂. Cores into Lake Kivu’s sedimentary floor indicate massive biological die-offs at roughly millennial intervals, which probably result from magmatic destabilisation of the gas-rich lower waters. Experimental vent pipes have been installed in Lake Nyos and nearby Lake Monoun to remove gas from the deep water (see Taming Lake Nyos, Cameroon and Letting Cameroon’s soda-pop lakes go flat in EPN issues for April 2001 and March 2003, respectively), but such a solution for the much larger Lake Kivu would be far less predictable and extremely expensive (Nayar, A. 2009. A lakeful of trouble. Nature, v. 460, p. 321-323). Energy companies based in DRC and Rwanda are now starting to use the ‘soda siphon’ approach that relieved Cameroon’s deadly lakes to capture the methane potential in Lake Kivu. Perhaps that will dampen down the lake’s potential for explosive gas surges, but no one knows if it could instead destabilise its uneasy equilibrium. Furthermore, the deep cool water is nutrient rich and may set off planktonic blooms in Lake Kivu’s surface waters. DRC is notorious for bandit mining and politics and security even more unstable than the lake that it shares with its tiny neighbour Rwanda. Population density on the lake’s shore, always high because of the fisheries and agricultural potential, rose explosively in the aftermath of the Rwandan genocide of 1994.