Petrographic Description of Marine Sediments from Santorini: Evidence for a New Submarine Hot-Spring Field in the Bottom of the Caldera

In this study we will present the results of a petrographic study of the coarse fraction of sediments in the caldera and its surroundings. This study is a complement to a study by Boström et al. (1990).

The results presented here as well as those by Boström et al. (1990) show the existence of a previously unknown hydrothermal centre in the north-east part of the caldera depression.

BOTTOM PHYSIOGRAPHY

The morphology of the sea floor round the Santorini islands has been described in more detail in Perissoratis (1990). The major islands Thera and Therasia form a ring-formed structure defining the outer rim of a caldera which obtained its last features in the Minoan eruption about 1500 BC (see Fig. 1).

In the area outside these major islands the sea floor steepens abruptly outside the shore line down to the 10 metre isobath, but further down the bottom slope diminishes. A very steep area exists in the north, however, in the strait between Thera and Therasia, where bathymetric studies reveal a deep furrow in the sea floor with depths exceeding 300 m; south of Thera there is a fairly shallow ridge.

Within the caldera there are four separate basins (Fig. 1). These include the northern basin north of the Kameni islands, with maximum depths reaching below the 390 m isobath, and the eastern, southern and western basins, all located south of the Kameni islands. These basins have maximum depths of about 290, 290 and 320 m respectively. All four basins are surrounded by precipitous slopes, but have flat bottom areas in their central parts.

FIELD AND LABORATORY METHODS

A total of 32 bottom surface samples were obtained outside the main islands of Thera and Therasia, and 14 within the caldera; the distribution of the samples is shown in Fig. 1. The samples were procured by means of a Dietz la fond grab sampler during a cruise on R/V lrini undertaken by the Marine Geology Department of IGME from May 30 to June 1, 1987. The determination of ship tracks and station locations was done by means of Loran-C and radar.

In the laboratory the initial samples were treated with oxygen peroxide to remove organic matter, after which a grain-size analysis was done. The pipette method was used to determine the fine fractions, i.e. those with grain-size diameters of less than 63 μm, whereas the coarse fractions were determined by means of sieving at one phi intervals. All sub-fractions of the coarse fractions were examined under a binocular microscope in order to determine the constituents of the sediments.

SEDIMENT DESCRIPTIONS AND PETROGRAPHIC RESULTS

The 32 samples from the area outside the caldera were all obtained at water depths between 14 and 280 metres. Their proportions of grain-size components, gravel, sand, silt and clay are shown in Fig. 1 and 2. The colours range from pale yellowish brown (10YR 2/2) to brownish black (5YR 2/1). The gravel is more abundant to the west, north and south of Santorini, while sand is predominant to the east. The silt and clay contents increase in the distal, deeper samples. The coarse fraction is composed of terrigenous, biogenous and authigenic (chemogenic) components. The terrigenous matter consists of pumice and other volcanic rock fragments. Also, quartz and feldspars occur to the south, apparently deriving from the erosion of the metamorphic Thera Basement formations that crop out at the adjacent parts of Thera. The biogenic fraction is composed mainly of shell fragments of benthonic and planktonic foraminifera, echinoid spines and sponge spicules. The authigenic components consist of iron oxides, which occur as 'aggregates' that have the appearance of distinct reddish grains within the coarse fraction. In most samples the terrigenous component makes up the largest constituent in the coarse fraction, but no particular regional trend is evident. The sediment samples outside the caldera basin contain between nil and 14% oxides, the highest contents being observed in samples 17 and 18, north-east of Thera (see also Fig. 2).

In the caldera the samples all derive from flat areas of the basin floors. The petrography of the sediment is markedly different from that in the sediments outside the caldera. The sediments are generally rich in sand and silt and the colour of most samples is moderately brown (5YR 3/4), apparently due to the high content of iron oxides. Thus, the authigenic mineral distribution in the coarse fraction is of particular interest. In the basins south of the Kameni islands and in most of the samples from the northern basin, the authigenic iron oxide aggregates constitute some 2 to 11% of the total coarse matter. However, in samples 138 and 139 the authigenic compound makes up 22% and 94% respectively. In sample 138 these grains have the common appearance of distinct reddish grains, but in sample 139 almost all of the coarse fraction consists of these aggregates. Even more, some of them occur in the gravel or even still coarser fractions (see Fig. 2, 4). The largest particles are in fact crusts with layered structures, with a dark reddish brown colour (10R 3/4) to a very dusky red (10R 2/2) and the individual layers have a thickness of 50 to 200 μm, and resemble the ore accumulations described at Nea Kameni by Butuzova (1978). Geochemical analyses of this sample showed a content of 40% Fe₂O₃, as compared to 5.8 to 7.3% for other caldera samples (Boström et al. 1990; see also our Fig. 3, 4).

DISCUSSION

The large amount of iron in the sediments in the northern part of the caldera has been reported by several researchers (Butuzova 1978; Petersen and Müller 1978; Smith and Cronan 1978), who thought that the iron derived from the Kameni islands or had been redistributed from the southern basins by the action of currents. However, the great water depth, of about 350 m, where the crusts are found, and the distance from the Kameni islands make the role of redistribution very unlikely. It is equally hard to visualize a long-distance transport from the Kameni islands that results in such exceptionally iron-enriched sediments but does not cause similar deposition closer to the presumed source region.

These relations suggest that another source must exist, most probably at or very close to sites 139 and 138. The simplest explanation is that a hydrothermal vent is debouching on the sea floor in the north-east part of the northern basin, or has been active there in the recent past. Station 139, in particular, is close to a line that connects the Kameni islands and the submarine volcano Kolombos, suggesting that a weakness zone exists along this line, as was also suggested by Heiken and McCoy (1984). Perhaps it is indicative that the sediments outside the caldera that are richest in ferric hydroxides (nos. 17 and 18) occur near this suggested weakness zone. However, further studies will be needed to corroborate the value of this indication.

These results are corroborated by findings in Boström et al. (1990), who likewise deduced the existence of a new hydrothermal vent in the north-east caldera. Their study demonstrated that the authigenic fraction of the sediments from site 139 showed remarkable similarities with exhalative-sedimentary deposits presently forming at hot springs on Nea Kameni, with regard to both major and trace element abundances. Their mixing models suggest that about 90% of all iron, 75% of all copper, and 67% of all vanadium in the sediment derive from such local hot-spring solutions.

CONCLUSIONS

Sediments in the caldera of Santorini are rich in authigenic iron oxides, most probably largely deriving from metalliferous hot springs at the Kameni islands, but some exceptionally iron-rich sediments occurring in the north-east part of the caldera depression are most probably formed by local hot springs on the caldera floor.

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

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