Following archaeological digs, analysis of the lead pollution in sediments in the antique port of Naples proves that the region’s vast water supply network, based on the Augustan Aqueduct, was profoundly affected by the infamous eruption of Vesuvius that destroyed Pompeii.
In the year 79, the eruption of Vesuvius buried a large swathe of the Bay of Naples in a cloud of ash. It entirely destroyed Pompeii and Herculanum, and killed more than 3,000 people. Historians and archaeologists have long wondered about the consequences of the eruption on the water supply of the bay, a popular holiday destination for the Roman aristocracy.
Towns on the Bay of Naples were supplied with fresh water by the Aqua Augusta, a network of galleries, canals, reservoirs and bridges built during the reign of the emperor Augustus between 27 BC and 10 AD. Around 140 km long, it conveyed fresh water from a source in the Apennines, supplied around ten towns (including Neapolis – ancient Naples - Pouzzoles, Cumes, Baïes and Nola), and ended in the Piscina Mirabilis, a huge Roman reservoir of drinking water overlooking the port of Misenum. The route took it along the northern and north-eastern flanks of Vesuvius. Although it is more than likely that the water supply was hard hit by the eruption, no traces of destruction or repair have ever been found. It should be said that few vestiges of the aqueduct remain. Our team from the Archéorient Laboratory and the Lyon Laboratory of Geology, in collaboration with scientists from the Universities of Glasgow and Naples, have just identified a first indirect clue to the damage suffered by the water supply: lead contained in the sediment of the Neapolis harbour basin (1).
In 2014, as part of the very first study of its type, we showed that the lead accumulated in the port sediment could be a reliable clue to the development of an urban water supply over a huge time scale (2). Then we analysed the lead contained in the sediment of the Imperial Rome’s port (Portus), which not only included the lead naturally present in the rock (“natural” lead), but also lead from human activities (“anthropogenic” lead). We revealed that this lead had the same signature as the lead used in some of the Roman pipes that distributed fresh water to the capital.
Contrary to the hypothesis sometimes put forward, the lead pipes were not protected from corrosion by a layer of limescale. The water dissolved some of the lead in the pipes, and discharged into natural water courses before depositing it in the harbour sediments.
Analysis of the harbour sediment enabled us to determine that the levels of lead contamination in the drinking water were not sufficient to poison the inhabitants. Saturnism (*) cannot therefore be used to explain the decline of the Roman Empire at the end of Antiquity.
Based on this Roman experience, we followed the dig in the ancient port of Naples with interest. In 2004, it was located in an ancient bay, called Echia, separating the two former urban centres of Neapolis and Parthenope, during preventive archaeological operations started as part of the construction of two new metro lines. The work gave access to the sediment deposited in successive layers by progressive silting up of the port basin. Analysis of the lead in this sediment provided useful information, both about the history of lead craft in Naples and the geographic origin of the sources of lead ore used in metal-working workshops over the centuries.
In 2011, we obtained authorisation from the archaeological office of Naples and Pompeii to sample sediments for geochemical and sedimentological analyses. Between 2011 and 2013, we visited the dig site four times to sample the stratigraphic layers gradually being uncovered. The sampling conditions were exceptional: usually, we work on sediment columns (core samples) taken by systems that drill through several metres depth of ground. This time, the dig threw up several thousand cubic metres of soil, enabling more samples to be taken and at several places on the site, thus giving us a much more accurate idea of the lead content of the sediment. In 2013, the archaeological site reached a depth of 6 metres under the current sea level. At around 4.5 metres deep, a lighter layer measuring around thirty centimetres could be seen. This was the sign of the eruption of Vesuvius in 79 AD, comprised of pumice carried by the resulting tsunami (3). Once the samples were brought back to the Lyon Laboratory of Geology, we commenced the chemical analysis.
The first step consisted of separating the natural lead – present in the rock formations from whence the sediments come – from the anthropogenic lead - which fixed in the sediment during a pollution, specifically due to contaminated water coming from the towns’ supply network. This separation is possible because natural lead and anthropogenic lead are connected to the sediments differently. Natural lead is contained within the crystal structure of the minerals, whereas anthropogenic lead is adsorbed, i.e. it adheres on the surface of sedimentary particles thanks to stronger or weaker chemical interactions. Anthropogenic lead can be collected by washing with a suitable cocktail of acids and solvents. The operation is conducted in a clean room, where the atmosphere is carefully controlled and free of particles so as not to contaminate the samples. This step allows numerous metal pollutants to be recovered, including the lead. Several chemical purification steps are then used to isolate the anthropogenic lead.
But where did this contamination come from? Our experience in Rome indicated that it could have resulted from the town’s water supply system. To verify this, we had to determine the isotopic signature of the anthropogenic lead. Each lead ore differs in the relative proportion of the lead isotopes (*) it contains, giving clues as to the history of the rocks containing it. In nature, there are four stable isotopes of lead: lead 204, lead 206, lead 207 and lead 208. From one rock to another, the proportion of the isotope lead 204 does not change. It is considered as primitive, as its quantity has not changed since the formation of the Earth.
The proportion of heavier isotopes however, known as radiogenic, varies as it is enriched over time. Lead 206, lead 207 and lead 208 are the result of the decay of radioactive elements, respectively isotopes of uranium (uranium 238 and uranium 235) and thorium (thorium 232). The proportion of radiogenic lead isotopes increases continuously, at least as long as these elements are free, i.e. as long as the mineral crystallisation process is not complete and the rock is not formed. Then, exchanges with the outside cease, the production of radiogenic lead is halted and the isotopic composition of the lead in the newly formed rock is “frozen” - it remains stable. The proportions of lead isotopes thus form an isotopic signature, specific to each metal-bearing deposit. To know this signature is to be able to deduce the age of the rock formation from which the lead was extracted. It is a sort of geological clock. This, incidentally, is the type of clock that enabled the Earth to be accurately dated for the first time in the last century.
We therefore determined the isotopic signature of the lead in our samples, using the mass spectrometry technique. To know if the lead was indeed from the Naples water supply, we needed to compare this signature with that of the lead circulating in the Aqua Augusta. We were unable to access the current, rare vestiges of the Naples lead pipes used in this structure, but rather a clue that told us just as much: the limescale made up of very fine layers of calcium carbonate deposited every year on the internal walls of the aqueduct’s canal. This anthropogenic “travertine” is formed by precipitation of the carbonate dissolved in the water running through the water supply infrastructure. In total, around ten samples were taken at different points along the aqueduct, specifically in the towns of Pompeii, Naples and Miseno.
The first result from these analyses shows that the urban water from the city of Naples was severely contaminated with lead during the first six centuries of our millennium (Fig. 1). In fact, the lead content was on average 3.5 times higher than that of “clean” water. The second result is that the isotopic signature of the anthropogenic lead in the harbour sediments of Neapolis is indeed that of the lead trapped in the carbonate deposits in the aqueduct. Like in Rome, the water in the ancient port of Neapolis was full of lead due to the dissolution of parts of the pipes making up the Aqua Augusta and possibly from the urban water distribution system in Naples.
The common footprint of the lead in the pipes and port sediments indicates that the rocks containing these metal-bearing ores were very ancient; these were “Hercynian” rocks, formed between 250 and 400 million years ago. These types of rocks do not occur anywhere in the Italian peninsula. The local volcanic rock, such as in the Phlegraean Fields or the sedimentary rock in the Apennines, formed recently, less than 20 million years ago. This lead from human activity was therefore imported by the Romans from far-off sources. The isotopic composition of the lead in the Naples deposits is very similar to those in the lead pipes in Rome and Pompeii (4). The metal was therefore imported from similar sources. In order to identify these sources, we compared the isotopic signature of the lead in the pipes in these three southern Italian cities with that of the first mineral deposits mined in the Mediterranean during the Antiquity. It is close to several signatures from mining areas in western Europe, active during the Imperial Roman Empire[AG1] : that of the Sierra Morena in Spain, the southern Cévennes and the Alps in France, the Eifel in Germany and the Pennines in England. The shared origin of the lead ore contained in all these pipes is not surprising. Naples, Rome and Pompeii kept close cultural and trading ties with each other, but also with Western Europe, whose territories fell under Roman control during this period.
While our analyses point towards imported lead, they also indicate that the composition of the Naples pipes changed over time. However, there is a break in this development; the isotopic composition of the anthropogenic lead trapped in the port sediment changes drastically after the infamous eruption of Vesuvius in 79 AD. In other words, the isotopic signature of the lead pollution differs between the lower, lighter layer (hence older and before the eruption) and the upper layer (after the eruption). This difference points to the use of different lead ore in the pipes used in the second generation of the region’s water supply. The source of the lead ore thus changed between the initial construction period of the aqueduct and its reconstruction, about one century later. This double lead isotopic signature in the port sediments is proof that the vast water supply system for the city and bay of Naples was destroyed during the eruption of Vesuvius in 79 AD, and underwent extensive repairs to get it working again.
This is hardly surprising. Prior to the eruption, bulging of the flanks of the volcano deformed the slope in the section of the Aqua Augusta channel located on the northern slope of Vesuvius. This could have broken the channel. Traces of such an event do remain at Ponte Tirone, a site where the land lifted and fell more than 30 centimetres before and after the eruption (5). The earthquakes that shook the bay of Naples just a few days before the major eruption of Vesuvius could also have caused widespread damage to the aqueduct. Finally, the ash spewed out during the eruption is a third possible cause of damage, as it could have obstructed the pipes after getting into the system through open access wells designed for maintenance.
The change in the isotopic signature of the anthropogenic lead in the port sediments occurred around fifteen years after the volcanic eruption, suggesting that the “pre-eruption” water supply was replaced with the “post-eruption” system in a relatively short period, probably less than a generation. Nevertheless, for the inhabitants of the bay, it must have seemed a long time. Some would have had to get their fresh water in the same way as the Greek colonists did five centuries earlier, from underground tanks around Naples.
(1) H. Delile et al., PNAS, 113, 22, 2016.
(2) H. Delile et al., PNAS, 111, 18, 2014.
(3) H. Delile et al., Quat. Sci. Rev., 150, 84, 2016.
(4) M. Boni et al., Archaeometry, 42, 201, 2000.
(5) D. Keenan-Jones, Am. J. Archaeol., 119, 191, 2015.
(*) Saturnism is a neurological disease caused by lead poisoning.
(*) An isotope is a variant of a chemical element different in its number of neutrons (and hence mass).
The eruption of Vesuvius in the year 79 AD is well documented. However, its impact on the water supply had not yet been demonstrated. The study of the lead isotopes contained in uncovered sediments revealed that the Aqua Augusta, which supplied water to cities in the Bay of Naples, was destroyed by the eruption and must have undergone extensive repairs.
4.5 METRES, this is the depth at which a light-coloured layer, thirty centimetres thick, was discovered in the archaeological dig in the ancient port of Naples. It is a mark of the eruption of Vesuvius in 79 AD, as it is comprised of pumice carried after the resulting tsunami.
30 CENTIMETRES is the height of the lift and fall of the land, before and after the eruption of Vesuvius, at Ponte Tirone, on the northern flank of the volcano, which could have broken the canal in the Aqua Augusta aqueduct.
> FOR MORE INFORMATION
This blog explains the activities of members of the Archéorient Laboratory to the general public, and more specifically current archaeological and historical research being conducted in the Mediterranean, southeast Asia and Africa.
Hugo focuses his research on the development of metal paleo-pollution in ancient harbour basins, the reconstruction of coastal and river delta paleo-environments in these sites in the Mediterranean and the hydroclimatic variability and the impact of human activities over a long period of time.