The Possible Effects of Santorini Tephra Fall on Minoan Crete

Although Santorini tephra in any quantity has not been positively identified on Crete, numerous authors have attributed wholesale destruction of Minoan sites to either eruption-associated earthquakes, tsunamis generated by caldera collapse or tephra fall (Page 1970; Marinatos 1939; Hood 1971; Chadwick 1976). The recent paper by Watkins et al. (1978) demonstrating that only 0 - 5 cm of uncompacted tephra reached the eastern half of Crete, allows a more accurate appraisal of the likely effects of tephra fall on the economy and people of the island.

Although detailed accounts of the effects of known thickness of tephra on Mediterranean economies seem not to have been published despite the excellent documentation of the eruption histories of Vesuvius and Etna, considerable pertinent information is available. It should be remembered, however, that the literature tends to overemphasise the hazardness of tephra falls, as a lack of effects is not generally considered newsworthy.

EFFECTS ON PLANTS

The agricultural systems of Minoan Crete were based on the cultivation of wheat, barley, olives, vines, and fruits (Warren 1975; Platon 1966). At least two varieties of wheat were known (Hood 1971) and several varieties of grapes (Platon 1966). Olives were harvested in autumn or winter (Hood 1971; Clutton & Kenny 1976), and dried figs and raisins were probably available for winter consumption. A large variety of wild vegetables, including lentils and peas suitable for winter storage were also used.

The cultivable areas lie at different elevations; an important point as harvesting was permitted at varying times (Marinatos 1960). Thus, today, early table grapes are harvested near Iraklion in mid-July but on the higher slopes the vintage is not harvested until October (Clutton & Kenny 1976).

The annual rainfall in the hill zone is over 800 mm, the rainy season lasting from October to May. On the lowlands Iraklion's rainfall totals c. 490 mm with a wet season extending from November to March (Clutton & Kenny 1976). The rains of early spring and summer which would help crops in more elevated areas could be harmful to harvesting operations on the lowlands only a few kilometres away. Although Crete has few permanent streams, many springs run year round (Marinatos 1960). At Knossos several springs which evidently once supplied the city are now dry (Hood 1971).

Even within the limited area of Eastern Crete there is considerable geographic, climatic and agricultural diversity. A single tephra fall could thus be expected to produce widely differing effects even apart from the effects resulting from the 0 - 5 cm variation in tephra thickness.

Tephra fall adversely affects vegetation in four ways (Rees 1970):

1. complete burial of the plant,
2. partial burial restricting root access to oxygen,
3. defoliation and prolonged absence of leaves, and
4. tephra covering foliage surfaces, clogging stomata and blocking out sunlight.

Falls of 5 cm or less beat down plants and slightly injured crops at Tambora in 1815 (Anon 1816). At Soufrière, in 1902, 12 mm of tephra blighted and yellowed tender plants causing breadfruit to drop prematurely (Anderson & Flett 1903). Sweet potatoes, yams, tannias, and sugar cane were damaged, leaves were stripped off bananas and plantains and branches were broken off young breadfruit trees by falls of 25 - 30 mm (HMSO 1903). At Paricutin (1943 - 45) about 5 cm of fine tephra prevented pollination by entering avocado flowers (Rees 1970). At Taal (1911) 1 cm of tephra killed bananas and other delicate vegetation (Pratt 1911) and at Sakurazima in 1974 about 2 cm of tephra damaged mandarin oranges, vegetables and mulberry tree crops (SICSLP 1974). Evidently even thin tephra falls can affect plants in a variety of ways.

Some of the effects ascribed to very thin tephra falls possibly result from acid coatings on grains and/or acid rains. Some eruptions, such as those of Tambora (1815) and Krakatoa (1883) seem to have produced no effects resulting from acid coatings; on the other hand, the eruption of Katmai-Novarupta (1912) was particularly sulphorous, vegetation being severely burnt as much as 480 km from the vent (Erskine 1962).

Similarly, hydrochloric acid coatings on tephra from Mayon (1928) wilted and browned leaves, producing black spots on abaca plants (Faustino 1929). Other trees lost most of their leaves and grasses and bamboo turned burnt-brown. In some cases only those portions of the plants facing into the wind were damaged. Furthermore, the occurrence of acid rains may be sporadic and the threshold concentrations of deleterious gases and acids varies not only from one plant species to another, but also plant susceptibility may increase drastically during periods of active growth and/or high humidity (Wilcox 1959).

Indirect effects of tephra falls on vegetation through effects on insect populations are impossible to predict. For example, around Paricutin, elimination of a fruit fly by tephra fall produced good crops of mangos and guavas for several years but there was a widespread fall in the cane harvest due to a plague of cane borers following the annihilation of the borer's predator (Rees 1970).

Little information is available concerning the effects of tephra falls on plants extant in Crete in Minoan times. References to the effects on wheat and barley have not been found but corn was damaged in Usu in 1944 by minor (unspecified) amounts of tephra (Minakami et al. 1952). A witness to the 1707 eruption of Lesser Kammeni (Santorini) reported that ashes and dust proved detrimental to the corn on Thera which had barely come up (Bourguigon 1708). "Smoak" (fumes) from this eruption also had a deleterious effect on the Santorini vines: "when this wind carried all the Smoak on upon their vineyards, which burnt them up in such a Manner, that the Grapes (which were not yet ripe) turned in one Nights time like dried Raisons, so because of their sowerness" (Goree 1711). This eruption, incidentally, was observed from both Naxos and Crete.

EFFECTS ON BIRDS, FISH AND ANIMALS

The Minoan agriculturist kept cattle, sheep, pigs and goats (Warren 1975). Deer, rabbits, partridge, pigeons, and fish were hunted although Branigan has suggested that by the early Bronze Age hunting of wild animals had probably made them harder to find (Branigan 1970). The sea was rich in food - tunny, mullet, mackerel, octopus, lobster, crab and a variety of shellfish. Seafood must have formed an important element of the diet in coastal sites (Hood 1971).

Once again little information is available about the effects of tephra falls in Mediterranean conditions, although Rodwell (1878) notes that birds around Etna died during the 1329 eruption, and de Lorenzo (1906) reports that during the 1906 eruption of Vesuvius where the tephra fall was heaviest "dead birds bestrewed the ground with wide open beaks, as if asphyxiated". Following this eruption, Delme-Radcliffe (1906) noticed a curious absence of insect and bird life. Quail which habitually use Santorini as a resting spot on their migration to North Africa abandoned the island in 1925 because of the noise of the eruption, the smoke clouds, or the ash (Washington 1926).

An eruption of Vesuvius in the time of Vespasian is reported to have killed many fish (Varenius 1683). Similarly, the 1906 eruption of Vesuvius reportedly harmed much of the floating fauna and some of the bottom dwellers in the Bay of Naples (Lacroix 1906). Around Santorini in 1925 many fish were killed by high water temperatures and by poisoned gases (Washington 1926). Similar comments were made about the 1707 eruption of Santorini (Bourguigon 1708).

A better notion of the effects of tephra falls on birds, fish and animals can be gained from a consideration of the 1912 eruption of Katmai-Novarupta which deposited about 35 cm of uncompacted tephra at Kodiak, the 1886 eruption of Tarawera and the 1815 eruption of Tambora. Most of the small birds around Kodiak died, apparently gassed. Large birds flew around in the darkness in panic until they hit something. Afterwards, dead gulls, snipes, ptarmigans and ducks were found everywhere, buried in the ash, tangled in trees, and at sea. Nearly all domestic fowl were killed, either gassed or smothered in the ashes. Sea birds that did survive could not find open water on which to land or take off because of the massive sheets of floating pumice; they subsequently died of exhaustion or drowned (Erskine 1962).

Around Tarawera where less than about 30 cm of tephra fell many birds were exterminated but many also survived. Sparrows were found temporarily blinded with their eyelids gummed together by falling mud while wild pheasants had so much mud in their feathers they were unable to fly. Pheasants and quail came almost to house doors seeking food (Grayland & Grayland 1971); At Macassar in 1815, 400 km north of Tambora, an ash fall of only 3 cm was sufficient to kill small birds (Anon 1816).

The 3 cm fall of Tambora tephra at Macassar in 1815 was also sufficient to cause the death of many pond fish. Similarly, around the Tarawera lakes in 1886, numerous small fish were washed ashore dead or dying. Many, though living, were partially discoloured and had a bruised appearance (Grayland & Grayland 1971). Around Kodiak 35 cm of Katmai-Novarupra tephra made the streams turbid and lethal to spawning salmon so that it took some years for recovery. On the othe hand, halibut were unaffected except that pumice-covered water surfaces seemed to frighten the fish away (Erskine 1962). In Icelandic waters the 1947 Hekla tephra fall also seemed to cause cod to move into deeper waters for two days (Thorarinsson 1954).

Fresh water fish would seem to be more affected by tephra fall than are marine species. Nonetheless, drifting sheets of pumice may have adversely affected fishing for months after the Minoan eruption of Santorini.

Around Tarawera rats and mice "made bold by hunger" were seen hunting for food in the ruined buildings (Grayland & Grayland 1971). On the other hand, small animals (marmots and ermine) at Kodiak were almost totally annihilated. At Kodiak livestock survived, obtaining water from rain and ice, some food from shoots that poked through on steep slopes, and salt from kelp beds on the shore. The cattle suffered from severe nasal infections and running eyes but they survived a tephra fall of about 35 cm (Erskine 1962). East of Tarawera even where the 1886 tephra fall was only 8 cm or less in thickness (Pullar 1967) the cattle moaned with hunger, suffered parched tongues, bloodshot eyes, and nostrils filled with mud (Grayland & Grayland 1971). But evidently the great majority survived. A 20 cm thick tephra fall 400 km west of Tambora caused the death of 212 horses and cattle chiefly from want of forage during the month following the eruption (Raffles 1830). Although the number of animals killed seems large, it should be emphasised that the survival rate is not known but was probably very high.

Chemical effects, notably those resulting from fluorine toxicity have been reported from a number of eruptions, almost exclusively in Iceland. Following the Hekla 1970 eruption poisoning of sheep occurred across very wide areas where the fluorine content of dried grass exceeded 250 ppm (Thorarinsson & Sigvaldason 1972). Contents of up to 4000 ppm were recorded and sheep died in their thousands although many had been in poor condition before the eruption. Fluorine poisoning occurred even where only 1 mm of tephra fell. Similar effects were noted after the 1783 Laki fissure eruption, the 1934 Grimsvotn outburst (Nielsen 1937) and the 1693, 1766 and 1845 eruptions of Hekla (Thorarinsson 1970). A case of poisoning also occurred during the 1767 eruption of Vesuvius; a peasant lost eight hogs when tephra fell into their trough. The hogs grew giddy and died in a few hours (Hamilton 1772).

EFFECTS ON BUILDINGS

The walls of Minoan buildings were constructed of stone or rammed earth (Warren 1975) with ceilings made of clay laid on a cane or reed framework on wooden beams (Branigan 1970). Many houses were three storied with a penthouse on a slightly sloping terrace roof. House fronts had broad airy windows which may have been fitted with thin transparent skins (Marinatos 1960). Some houses had cisterns, others had a reticulated water supply.

Although freshly fallen tephra has a unit weight of only 0.7 - 1.0 g cm^-3 there are a number of records in the literature of house roofs collapsing under falls of less than 10 cm. In fact, in two cases roofs collapsed under the weight of 3.0 cm of tephra or less: Soufrière, 1812 - 1.5 to 2.5 cm (Anderson & Flett 1903); and Vesuvius, 1906 - 2.5 cm (Hobbs, 1906). In the latter case the roof of the market of Monte Oliveto collapsed together with some poorly constructed house roofs. Further evidence that roof collapse can occur under quite small loads is provided by a recent United States example where the roof of a four year old sports stadium costing $US 70 milion collapsed under the weight of 11.2 cm of wet snow (TIME 1978). Despite these extreme examples, the percentage of buildings which collapse under such small tephra loads is very low indeed. Although usually only old or poorly constructed buildings are affected dwellings with flat roofs are more susceptible to collapse than are those with steeply pitched roofs (Jagger 1956). After the 1906 eruption of Vesuvius, Perret reported that 10 cm of tephra was usually enough to cause the collapse of flat roofs (Perret 1924). A fall of 6 - 8 cm on flat roofs in the towns around Stromboli (1912) produced no observable effects (Perret 1912).

Drainage systems can easily become clogged during and after tephra fall thus causing local flooding as at Irazú in 1964 (Bolt et al. 1975). Furthermore, a 3 - 6 mm fall of Mt Spurr tephra at Anchorage in 1953 caused the pH of the public water supply to fall to 4.5 and then to return to 7.0 after a few hours (Wilcox 1959). Turbidity of the water rose from 5 ppm to 290 ppm and took 6 days to return to normal levels. In the Cretan situation the effects would vary depending on whether individual water supplies were from streams, springs, cisterns or wells.

Mudflows following rainfalls which wash tephra from hillslopes can also be a serious source of damage to dwellings but they are not likely to be of much consequence with falls of less than 5 cm of tephra.

 EFFECTS ON PEOPLE

Although it is possible, perhaps even probable, that some buildings collapsed and that a few people died in the wreckage the direct effects of the Bronze Age tephra fall on people on Crete were certainly minor. Even where exposure to tephra fall has been long continued such as at San Jose during the 1963 - 64 eruption of Irazú, local physicians argued that the effects were not severe enough to produce any deaths (Horton & McCaldin 1964). Most people suffered acute conjunctivitis, throat irritation and discharge but these problems cleared up very quickly after exposure to the falling tephra ceased. People with pre-existing chest complaints developed severe bronchitic symptoms which lasted some days beyond exposure to the tephra fall.

Given that even the minor 1707 eruption of Santorini was observed from Crete (Bourguigon 1708), the Minoans were probably forewarned and the population may have fled westward during or before the tephra fall. The 1906 eruption of Vesuvius which deposited 3 cm of tephra on Naples caused 100,000 people to leave the city during a five day period (Lacroix 1906). Perret reported that "the behaviour of these stricken folk was admirable, and a greater patience, resignation, and 'savoir faire' could hardly have been expected of any race. An interesting feature noted by the writer was that the children had, so to speak, taken charge; it was they, rather than their elders, who were directing the flight, suggesting destinations, urging on the beasts" (but see the contrary view held by Delme-Radcliffe 1906). Some people might not have returned to their homes following flight; for example, many people left Santorini during the 1925 eruption, so that a large school had to be closed (Washington 1925).

CONCLUSIONS

The best available information suggests that the Minoan tephra fall on Crete did not exceed 5 cm uncompacted thickness. In fact the isopach map of Watkins et al. (1978), indicates that much of eastern Crete received less than 2.0 cm tephra fall.

Perhaps people did flee from their homes and perhaps some dwellings did collapse, even killing a few of those who remained behind. Water supply from springs would have allowed a prompt return of those who fled, even while streams remained turbid. Throat, eye and chest inflammations should have ceased within a few days of the tephra fall.

Despite the catalogue of death and destruction presented here the available evidence suggests that domestic and wild animals are probably well equipped to survive falls of 5 cm of uncompacted tephra. Such a thin tephra layer is unlikely to cover or destroy all the vegetation so that many animals can continue to forage for food. This is particularly true if rainfall washes the tephra from leaf surfaces.

Present knowledge does not allow any conclusions to be reached regarding fluorine poisoning during the Bronze Age eruption of Santorini. The incidence of such poisoning outside Iceland would seem remarkably low; the probability is that it was not a factor in the destruction / abandonment of Minoan Crete. If fluorine poisoning did not occur the effects of the tephra fall on most animals would probably have been fairly minor, especially as only a fraction of the eastern end of the island received even 5 cm of fallout. If fluorine contents were high enough to be toxic to animals, particularly close-grazing animals such as sheep and goats, a very high percentage of the animal population could have been killed. On the other hand, if good rainfalls followed the tephra fall, adverse effects would probably have been short-lived. Thus a fluorine-rich tephra fall in summer would provide the most adverse conditions for animal survival. The probability of such an occurrence must be very low.

Although a wide range of effects on plants resulting from tephra fall are possible even where the maximum thickness of the tephra is so circumscribed, the severity of damage is dependent not only on thickness but also on particle size, the occurrence of rainfall during and after the tephra fallout, the erosional removal or wind drifting of the tephra and the season of the year. The last point is particularly important as Thorarinsson (1971) has previously noted.

Whatever the effects of the Minoan tephra fall on the vegetation and agriculture of Crete, the effects were probably only short-lived. For example, the fine ash which coyered the vegetation around Vesuvius in 1906 seemed at first to destroy everything, but the buds of the vines were still closed and the destruction was less than expected.

At the beginning of May, three weeks after the eruption, the countryside was again luxuriantly green (Lacroix 1906). Similarly, Varenius writing in 1683 about an eruption of Vesuvius in the time of Vespasian reported: "But the Conflagration ceasing, and the Showers watering the Sulphorous Embers and Ashes, in the Superficies of the Mountain here and there was great fertility of Wine". After the 1812 tephra fall from Soufrière, agriculturists on Barbados treasured the so-called "May dust" for its fertilising properties "utilizing every little deposit as it was discovered and carefully applying it to the cultivation" (Huggins 1902). Griggs (1922) recorded the recovery of vegetation after the 1912 eruption of Katmai-Novarupta in great detail. His volume includes a revealing picture of increased increments of tree growth after the tephra fall. Some Highland New Guineans regard tephra falls as generally beneficial because of subsequent improved crop yields (Glasse 1963). Icelandic farmers also reported improved grass growth on low-lying lands following the Katla 1919 and Hekla 1947 eruptions. In areas around Paricutin where 3 cm of tephra fell, crops of wheat and barley were reported to be excellent (Rees 1970). A great deal of evidence suggests then that not only would the adverse effects of a 5 cm tephra fall be short-lived, but also that short-lived conditions more beneficial to agriculture than previously might be attained very rapidly. The diversity of Minoan agricultural practice would also seem to preclude the possibility that the effects of a minor tephra fall could be catastrophic.

It seems likely then that the eruption of Santorini had few harmful effects on Minoan Crete at least as far as the tephra fall was concerned. Certainly the effects of associated atmospheric shock waves, earthquakes and tsunamis may have proved most deleterious to Cretan communities and it is possible that a tephra fall on Minoan outposts on Rhodes, Melos and elsewhere was more harmful that it was to eastern Crete. Nonetheless there seems to be a high probability that tephra fall on Crete played a very minor role if indeed any at all in the decline/destruction of Minoan Crete.

---------------------------------------