It is now a decade since the enormity of natural arsenic contamination in groundwater below the great plains of northern India and Bangladesh came to light. In 1995 the World Health Organisation announced that this waterborne arsenic was causing the world’s largest case of mass poisoning. Since then other areas at risk have emerged in East and Central Asia and South America. The tragedy is that groundwater generally presents the safest option for drinking water because sediments filter water and encourage biogenic oxidation that remove common pathogens. That tens of million people in West Bengal and Bangladesh face stealthy poisoning results from channels cut in the low-lying plains during the last glacial maximum being filled rapidly with sediment as sea level rose during climatic recovery. Sedimentation buried large amounts of organic debris to form anoxic conditions in the shallower sediments. Reducing conditions encourage breakdown of the common colorant in sediments, iron hydroxide grain coatings that, having adsorbed most arsenic and other ions from water, releases them when it dissolves. That this should occur was unsuspected during a massive programme of well sinking to relieve endemic ill health from waterborne disease, yet early signs that arsenic had replaced pathogens as a hazard was widely ignored, despite a few warning voices who discovered the unmistakable signs of arsenicosis in the 1980s. They include disfiguring pigmented skin spots and horny growths on hands and feet.

By 1995, the rest of the world took notice, pouring in funds to document occurrences and causes, and to remediate a clearly catastrophic situation. There are three main strategies: to remove arsenic from well water using chemical filters; to return to water from surface sources, though with careful processing to remove pathogens; to sink wells below the level known to encourage arsenic release from iron hydroxide dissolution. For two decades affected populations had been bombarded with encouragement to turn to groundwater: against their better judgement – they termed it the Devil’s water. Once using wells they saw that infant mortality plummeted, so they developed a new enthusiasm for water deemed safe. Caught on the horns of a dilemma, when arsenicosis appeared they were reluctant to return to what appeared to be the greater of two evils. In only a few places were wells deepened to safe depths, and the externally sponsored drive for a solution centred on arsenic removal techniques. Even that was not widespread: of millions of risky wells some 2000 were equipped with arsenic extracting devices, at around US$ 1500 each. It now emerges that the technologies chosen are not doing their intended job (Hossain, M.A. (and 10 others) 2005. Ineffectiveness and Poor Reliability of Arsenic Removal Plants in West Bengal, India. Environmental Science & Technology, v. 39, p. 4300-4306). The team, led by Depankar Chakraborti, who first spoke out about arsenicosis in 1983, tested the efficacy of 18 different devices installed in West Bengal. Only two reduced arsenic levels to the maximum of 50 parts per billion accepted by the Indian government, which is itself five times more than that deemed safe by the WHO. The teams view, supported by the agency that did most to encourage the massive well-driving programme since the 1970s (UNICEF), is that the only realistic solution is a return to rainwater harvesting and purification.

See also: Ball, P. 2005. Arsenic-free water still a pipedream. Nature, v. 436, p. 313.

Legendary events at the Gibraltar Straits

Everyone has heard of Atlantis, but few would care either to point to its former position, or to accept its existence without a shed-full of salt. Nevertheless, no lesser an authority than Plato first described the legend of Atlantis in the 4^th century BC, following verbal accounts that originated in pharaonic Egypt. In the last decade a number of legends, if not their religious connotations, have received scientific support. Foremost among these is that of the biblical Flood, which Ryan and Pitman pursued relentlessly, using the Epic of Gilgamesh as a geographic and chronological guide. They discovered that the Black Sea had catastrophically filled through the Bosphorus once global sea level topped the level of its floor, following glacial melting. Their evidence now includes numerous examples of habitations now inundated by the Black Sea.

As with Ryan and Pitman’s work, one key to resolving a real basis for a legend is carefully puzzling out clues in the most detailed accounts of it. In the case of Atlantis, the clues come from Plato himself (Gutscher, M-A. 2005. Destruction of Atlantis by a great earthquake and tsunami? A geological analysis of the Spartle Bank hypothesis. Geology, v. 33, p. 685-688). Marc-André Gutscher and previous workers focused on Plato’s geographic description of Atlantis, as well as its fate. Plato clearly specified an island in the Atlantic beyond the Straits of Gibraltar, and an earthquake and flood that put paid to the Atlanteans in a single day. Indeed, bathymetry does show well-defined shallows (less than 100 m depth) in such a location, but only about 5 km across. This is the Spartel palaeo-island, on which Gutscher turns his focus. Until the final, decisive rise in sea level after around 12 ka, Spartel would have been a low island. Plato’s account is supported by the existence of a proto subduction zone on the Atlantic sea floor off the Straits of Gibratlar, a major earthquake on which devastated Cadiz in 1755, partly because of a 10 m tsunami. Offshore sediments include turbidites that indicate 8 tsunamis since 12 ka, suggesting a 1500- to 2000-year periodicity of large earthquakes at the entrance to the Mediterranean. Plato’s version of the events includes a rough chronology that suggests a time around 11.6 ka before the present. The thickest of the tsunami-driven turbidites is of roughly that age. Unfortunately for the hypothesis that Spartel was Atlantis, at that time only two tiny islets would have stood above the waves. Seismic destruction of coastal regions by tsunamis is something that might easily become legendary, the more so in the distant past. There is one other possibility that might revive the Spartle hypothesis, demonstrated by the great Indian Ocean tsunami of 26 December 2004. Very powerful earthquakes can also result in massive displacement of the crust, or the order of tens of metres. Spartle might have sunk repeatedly since 11.6 ka, as a result of later events.