Historical and Geological Evidence for Seismic Origin of Newly.pdf

ENVIRONMENTAL ASSESSMENT DOI: 10.1007/s00267-001-0045-8 Historical and Geological Evidence for Seismic Origin of Newly Recognized Landslides in Southeastern Sicily, and Its Significance in Terms of Hazard FRANCESCA GRINGERI PANTANO Accademia di Belle Arti Via Malta 17 96100 Siracusa, Italy PIER GIORGIO NICOLETTI* Consiglio Nazionale delle Ricerche-IRPI Via C. B. Cavour 87030 Roges di Rende (CS), Italy MARIO PARISE Consiglio Nazionale delle Ricerche-CERIST Via G. Orabona 4 70125 Bari, Italy ABSTRACT / Old, large, and dormant landslides were unexpectedly found in southeastern Sicily, a territory of known seismicity but commonly considered as landslide-free or almost so. Purposely undertaken investigations revealed that: (1) these landslides are scarcely compatible with the local geoclimatic environment; (2) they usually show low-angle

Unawareness of a danger, or the long return period of its occurrence, which translates into loss of memory of the hazard, can greatly increase the damage related to a new phase of activity of the hazardous process. This is the present situation in southeastern Sicily as concerns earthquake-induced slope instability. In fact, while this 4300-km2 territory is considered hazardous because of its seismicity, it is also commonly thought of as nearly landslide-free. On the contrary, recent investigations have revealed that at least 146 old, large, and dormant landslides, previously unknown, exist and are attributable to seismic triggers (Figures 1 and 2) (Nicoletti and others 1999a,b, 2000; Nicoletti and Catalano 2000). In practically all these landslide sites a degree of KEY WORDS: Earthquake-triggered landslides; Historical evidence; Geological evidence; Sicily; Landslide hazard *Author to whom correspondence should be addressed; e-mail: [email protected]

Environmental Management Vol. 29, No. 1, pp. 116 –131

basal shear surfaces, despite the fact that the properties of the forming material are generally good; (3) they fulfill the known relationships between earthquake magnitude and epicenter–landslide distance; (4) sources coeval with high-energy historical earthquakes occurred in 1169, 1542 and 1693 testify to the occurrence of earthquake-triggered landsliding; and (5) documentary material (presented here for the first time) correlates with certainty a specific landslide to the 1693 earthquake. This geological and historical evidence, accompanied by the absence of contrasting elements, leads us to conclude that these landslides are earthquake-triggered. Because of their typological and geometrical characteristics, nearly all landslides can be reactivated, which has serious implications in terms of hazard, particularly with respect to lines of communication. Obviously, every action aimed at preventing or mitigating risks must start from the awareness of the causative processes, a condition substantially unsatisfied at the moment in SE Sicily. The paper concludes by emphasizing the opportunity not to trust excessively beliefs that, although shared, have never been really checked.

hazard is present, either due to possible reactivation of movement or to new failures. Since awareness is an indispensable premise of mitigation, the aim of this paper will be to illustrate the seismic origin of these landslides, and therefore the need to take into account slope instability when planning to prevent and mitigate the effects of future earthquakes in the area. In other words, it is aimed at building up the “awareness” step of a desirable awarenessmitigation process. In addition, since earthquakeinduced slope instability may have been overlooked in other countries, the experience introduced here may fruitfully stimulate doubts and research elsewhere.

Newly Recognized Landslides in SE Sicily: Main Attributes Some landslides were first seen, unexpectedly, while interpreting air photographs of the area for other reasons. A quick photogeological exam revealed there ©

2002 Springer-Verlag New York Inc.

Landslides of Seismic Origin in SE Sicily

117

Figure 1. In the northern part of the area, close to the town of Palagonia, three huge landslides are aligned on the right flank of a torrent. All involve volcanic and volcaniclastic rocks. No. 1 is a listric or biplanar rock slide, still in a relatively early stage. The ratio of crown–toe height difference to planimetric length (H/L) is 1:5.5, corresponding to an inclination of only

about 10°. No. 2 is a listric rock slide showing a clear head graben, about 1 km long; here, H/L ⫽ 1:3.3, i.e., ⬇17°. No. 3 is a well-developed circular rockslide, with H/L ⫽ 1:3.5, i.e. ⬇16°. Published by permission of the Italian Air Force (Concessione Aeronautica Militare-R.G.S. n. 0043 del 27/01/ 1999).

were many others. An extensive scrutiny of the literature showed they had never been studied and, in most cases, had not even been recognized. Then, a purposely undertaken investigation led to the identification, either with certainty or with reasonable probability, of 146 landslides whose main attributes are as follows [nomenclature after Cruden and Varnes (1996)]. Typology—Typology is slide in 90% of cases. Lateral spreads, falls, and flows account for the remaining 10% (Table 1). Style of activity—Landslides are multiple in most cases, complex or composite very rarely. Material—Material is well- to moderately lithified

rock in 128 cases (88%), poorly lithified rock in the remaining 18 (12%). Slip surface—Circular, generally low-curvature, slip surfaces are the most common, and are topographically expressed by the usual, although often subdued and rather flattened, slump morphology (Table 1). Listric surfaces are also common and have as surficial expression a horst-and-graben morphology. Landslides possessing a circular or listric slip surface are most susceptible to being reactivated. Planar slip surfaces are rare and may be steeply inclined; they are observed, among others, in the slides that have reached the position of minimum potential energy and can be considered as

118

F. Gringeri Pantano and others

Figure 2. A general sketch of SE Sicily (4300 km2) showing the towns mentioned in text and the main rivers.

Table 1.

Earthquake-triggered landslides in SE Sicily: Typological distributiona

Slide 132 77 43 12 a

(90), of which: (53) upon a circular surface, (29) upon a listric or planar or bi-planar surface, (8) upon a surface of complex or uncertain shape

Lateral spread

Fall

Flow

Complex or composite

Uncertain

TOTAL

3 (2.1)

2 (1.4)

1 (0.7)

6 (4.1)

2 (1.4)

146 (99.7)

Values are number (percent).

stabilized. Importantly, whatever shape the slip surface has, it very often shows a remarkably developed horizontal component. Size—Linear dimensions are commonly on the order of hundreds of meters and occasionally kilometers, areal extent is between 9 ⫻ 103 m2 and 3.2 km2. Volume ranges between about 5 ⫻ 104 and 100 ⫻ 106 m3. Age—Morphological and bibliographical evidence of recent movement are practically nonexistent, so these landslides can be assumed to be old. State of activity—Most landslides are dormant. In a few cases, when the failed mass has reached the position of minimum potential energy in the valley bottom

and is therefore unable to move further in its entirety, the landslide is naturally stabilized. Because of age and state of activity, morphology is frequently subdued (Wieczorek 1984). In short, this slope instability is mostly represented by large, deep-seated, rotational or transalational rock slides, which are generally dormant or, in very few cases, naturally stabilized. The area shows no active analogs. Although not strictly a landslide “attribute,” a notable effect has also to be mentioned: the formation of lacustrine basins by valley damming. There are 8 –20 such cases, the range depending on doubtful situations.

119


Table 2.

Main earthquakes in SE Sicily after 1000 ADa

Date 4 Feb 1169 10 Dec 1542 9 Jan 1693 11 Jan 1693 a

Magnitudeb

Epicenter intensity (MCS scale)

Epicenter area

5.6 (0.8)c 6.6 (0.4) 6.1 (0.3) 7.4 (0.1)d

X X VIII–IX XI

Eastern Sicily Syracuse area SE Sicily Eastern Sicily

Data after Boschi and others (1995, 1997).

b

“Equivalent macroseismic magnitude” as defined by Gasperini and Ferrari (1997). Between brackets, the error in the estimate.

c

As concerns this magnitude figure, Gasperini and Ferrari (1997, p. 62) state that “probably results [are] highly underestimated” as a consequence of the method employed. In the former edition of the same catalog (Boschi and others 1995), this magnitude had been estimated at 8.3, which made it the strongest earthquake in the Italian area in historical time. In a very recent parametric catalog (Gruppo di Lavoro CPTI 1999), it has been raised to 6.6. d

This shock lasted probably 4 minutes.

Among the eight certain cases, six lacustrine basins were already infilled by sediments (and silting up deposits are now reincised more or less in depth), while two basins are still being silted up.

Seismicity of the Area Since 1000 AD there have been many seismic events in SE Sicily. Three of them (Table 2) stand out and, according to the historical record, played a significant morphogenetic role (Boschi and others 1995, 1997). Unfortunately, no operationally useful information seems to be available about the number and strength of earthquakes in the previous millennia (Finley 1968, Guidoboni and others 1994, Boschi and others 1995, 1997), but the magnitude and frequencies that overall are comparable with those in Table 2 are geologically obvious over the long period. As a consequence, the earthquake landforms now visible are the cumulative result of many events, of which only the last three are documented historically. As concerns the seismogenic structures responsible for these earthquakes, and the epicenters, no general agreement exists among scholars (Barbano and Cosentino 1981, Barbano 1985a,b, Lombardo 1985, D’Addezio and Valensise 1993, Achilli and others 1993, Boschi and others 1995, 1997, Gruppo di Lavoro CPTI 1999, Azzaro and Barbano 2000). In any case the indications usually provided place both structures and epicenters in the eastern part of the area, either off- or onshore. The black polygons in the inset of Figure 3 “envelope” all suggested possibilities.

Seismic Origin of Landslides The seismic origin of these landslides is supported by five arguments, three of which are geological and two historical in nature. The geological arguments are:

(1) low compatibility with the local geoclimatic environment, (2) slip surface geometry, and (3) fulfilment of the known earthquakes-landslides relationships. The historical arguments are: (4) witness by coeval sources, and (5) documentary dating of a landslide. Each of these arguments will now be illustrated. It is worth noting that no contrary evidence was found, although it was carefully sought. In addition, we will also present information on malaria recrudescence, which might also be related to slope instability. Low Compatibility with Local Geoclimatic Environment Southeastern Sicily is part of a foreland area (Avampaese Pelagiano) (Lentini 1984, Finetti and others 1996, Lentini and others 1996) and consists basically of a plateau surrounded by lower, smoothly undulated to flat, lands (Figure 4). The central plateau, the Hyblaean Mountains, is dissected by canyons, which are often deeply incised, and by some wider, more open valleys. Although slopes may be steeply inclined to vertical, they are seldom higher than 200 –250 m and often much lower. The maximum plateau altitude, 986 m, is attained some 40 km inland. Average relief ratio of the drainage basins is as low as 0.021 (Nicoletti and others 1999a). Therefore, as these features indicate, relief energy is generally low. The Hyblaean Mountains, as well as the surrounding areas, are mostly underlain by subhorizontally stratified sedimentary rocks (Lentini 1984) (Figure 3). In the plateau, rock composition is more frequently clastic carbonate, with marlstone and other marly terms also represented. These materials are moderately to well lithified. In mechanical terms, field identification tests indicate that these rocks are moderately strong to very strong (GCO 1988). The thickness of the strata ranges typically between 1 or few dm and about 10 m. The attitude of the strata is normally subhorizontal (0°–

120



121

Figure 4. This topographic sketch shows the overall low, smooth relief characterizing the area. Note here and in Figure 2 the radial– centrifuge drainage pattern. A comparison with Figure 3 enables one to see the influence of lithology upon

relief. Of 146 observed landslides, 128 took place in the core area, mostly consisting of calcareous rocks with volcanic covers in the north part, and 18 in the outer belt, mostly consisting of poorly lithified sedimentary rocks.

10°), as already noted; higher dips are only locally present. Jointing is overall of moderate density, and with an almost exclusively subvertical attitude (Ghisetti and Vezzani 1980). Thus, in the Hyblaean Plateau both materials and structural configuration are generally favorable to stability. In the northern Hyblaean Mountains, extrusive basic rocks, mostly of basaltic composition, cover an ample sector and are locally interbedded with the sedimentary materials. Clastic rocks, more commonly medium- to finegrained and unlithified to moderately lithified, predominate in the areas around the plateau. Thickness of strata here ranges from decimeters to meters; attitude is also subhorizontal. In the NW corner of the area, heavily tectonized sedimentary terranes belonging to an allochtonous nappe crop out. Both in lithologic and structural terms,

they differ considerably from those described above and therefore are of no interest in the present context. Because of its location, SE Sicily has a typical Mediterranean climate— hot temperate with a dry summer. At present, annual rainfall is of about 400 mm in some coastal areas and increases to 600 mm along the outer belt of the Hyblaean Plateau and then to 1000 –1100 mm in two small top zones (Adorni and Aureli 1989). Such a geoclimatic environment does not really seem suitable to justify a significant instability of slopes. As a matter of fact, there is actually no news of significant instability as long as seismic triggers are not taken into account. This will now be shown by the preexisting geological and/or engineering literature (Agnesi and Lucchesi 1987). Early investigations on the land of Sicily (Ferrara 1810, Afan De Rivera 1832/42, Baldacci 1886), al-

4™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™™ Figure 3. Geologic sketch of SE Sicily (modified after Lentini, 1984, 1986, Grasso, 1997). In the inset, the black areas indicate the seismic epicenters and structures suggested by the various authors for the historical earthquakes.

122


Figure 5. Angular diagram of H/L ratios (in parentheses) for the 141 cases in which the figure is available (of which 129 slides). Angular values (␣ ⫽ arctan H/L) show the actual gradients. Note that inclination is smaller than 20° and 30° in 70% and 90% of cases, respectively.

though mentioning slope instability problems, never hint at them with reference to the southeastern part of the island. Later on, slope instability in this area was dealt with in several papers or maps, either as the main subject (Crino` 1921, Passerini 1922, Montanari 1941, Ministero dei Lavori Pubblici 1964, Liguori 1977, 1988, Coltro and others 1978, Jappelli and Musso 1981) or only marginally (Dal Piaz 1948, ESE 1951, Campisi 1961, Ministero dell’Agricoltura e delle Foreste 1976, Di Grande and Grasso 1977, Lentini 1984, 1986, Beccaluva and others 1993, Grasso 1997). Overall, this literature provided the picture described below. 1.

In SE Sicily slopes are basically stable (Crino` 1921, Montanari 1941, Ministero dei Lavori Pubblici 1964, Ministero dell’Agricoltura e delle Foreste 1976, Liguori 1977, Coltro and others 1978). 2. Instability, when occurring, is mostly determined by local factors, such as high slope angle, lithology, jointing, structural relationship between materials differing in permeability, and so on (Crino` 1921; Passerini 1922; Montanari 1941; Coltro and others 1978; Jappelli and Musso 1981). As a result, identified instability processes are mostly shallow rock, debris or earth slides, small rock falls, and small debris or earth flows. Only in four or five cases were conspicuous earth flows involved. These are the only recognized active or recently active processes, and they differ strikingly both in typomorphological attributes and size from the processes described above (see “Newly Recognized Landslides. . .”). 3. Some kind of recent interference between mass movement and anthropic structures or interests is often known (Passerini 1922, Ministero dei Lavori Pubblici 1964, Liguori 1988, Coltro and others 1978, Jappelli and Musso 1981). 4. Landslides more or less coinciding with cases dealt with in this paper are mentioned or shown only sporadically (Dal Piaz 1948, ESE 1951, Campisi 1961, Di Grande and Grasso 1977, Lentini 1984,

1986, Liguori 1988, Beccaluva and others 1993, Grasso 1997), and never as active phenomena. 5. Only in two cases are old river blockages recognized (Dal Piaz 1948, ESE 1951, Campisi 1961). 6. Only in one case, and only incidentally in a paper devoted to a geological survey, is the earthquake explicitly suggested as the trigger of a landslide in 1693 (Campisi 1961). The landslide cited is the same we describe later (see “Documentary Datingѧ”), and although this author did not know of the documents we present, his insight was correct. Such a picture is fairly exhaustive as concerns landslides of nonseismic origin, whose incidence appears remarkably limited, particularly in a region like Sicily, which is altogether highly landslide-susceptible. What is most important here, however, is that such a picture appears in good agreement with local physiographic and climatic context. Therefore, other landslides, which are both unknown and hard to explain within that context, must have another origin. Slip Surface Geometry As already noted, independent of their shape, slip surfaces have often a remarkably developed horizontal component. Figure 5 shows the ratio of crown–toe height difference to planimetric length (H/L), as well as the corresponding slope, of 141 landslides. It may be noted that 70% of them slope less than 20°, and 90° less than 30°. Such low slopes are significant in themselves, but their importance comes fully to light when they are considered alongside the lithologic and structural features of the rock masses, as sketched above (see “Low Compatibility. . .”). These low-slope slip surfaces provide a strong indication of horizontal seismic acceleration, because only in this way can masses of rock having fairly good mechanical properties and a structural configuration favorable to stability overcome all resisting forces and be set in motion (Hansen 1965, Jibson and Keefer 1988, 1993, Jibson 1996).


123

Figure 7. Diagram relating earthquake magnitude and area affected by landslides (modified after Keefer 1984). The thick segment is representative of SE Sicily data, and is compatible with the literature data. Figure 6. Diagram relating earthquake magnitude and distance from epicenter (modified after Keefer 1984). Curve A is the envelope for “disrupted slides or falls,” curve B for “coherent slides” (the most common typology in SE Sicily), and curve C for “lateral spreads and flows”. The dashed area shows the field of SE Sicily data, which are fully compatible with Keefer’s envelopes.

Such geometrical features of the slip surfaces also justify a kind of activity that is both episodic, because it occurs only as a consequence of a sufficiently energetic earthquake, and causing very long lasting landforms, because such earthquakes are relatively rare. In this context, repeated reactivations were and will be the rule. This has great importance in terms of hazard prevention. Fulfillment of Known Earthquake–Landslide Relationships Several diagrams linking earthquake magnitude or intensity with landslide distribution are available from the literature. Figure 6 shows the envelope magnitude– epicentral distance obtained by Keefer (1984) for various categories of landslides: disrupted slides and falls in curve A, coherent slides (the most common type in our population) in curve B, lateral spreads and flows in curve C. The dashed area is representative of conditions in SE Sicily: magnitude comprised between 5.6 and 7.4 and distance of epicenter not greater than 100 km. Most of this area falls within the envelopes, thus clearly indicating that seismic trigger is possible. The

same applies to the diagram in Figure 7 (also by Keefer 1984), which links magnitude and area: the 4300 km2 of SE Sicily fall largely within the envelope. It has to be noted that although coherent slides are the most represented category in our population, disrupted slides and falls were certainly triggered in great numbers by the earthquakes, as normally occurs (Keefer 1984) and was well noted by contemporary observers (see below). However, after centuries they can no longer be detected because they are easily eroded or otherwise obliterated due to their small size and constituent materials. Ground deformations are also known to have repeatedly occurred due to soil liquefaction (see next section). Field evidence related to the cases mentioned in the literature has not been traced yet, but two unknown cases have been identified, although they are doubtful (Nicoletti and others 2000). In any case, the diagrams available from the literature show full compatibility of seismicity and liquefactions in SE Sicily (Figure 8). Witness by Coeval Sources Sources coeval with important earthquakes were handed down through centuries by seismic catalogs (Mongitore 1743, Bonaiuti 1793, Perrey 1848, Mercalli 1883, SciutoPatti 1896, Baratta 1901). The recent catalog by Boschi and others (1995, 1997) includes CDROMs containing the richest list of sources, both coeval and later, ever published on Italian earthquakes, and is well-endowed with a number of quotations from original texts. Scrutinizing this catalog enables one to con-

124


Figure 8. Diagram relating epicenter distance with epicenter intensity for liquefactions (modified after Galli and Meloni 1993). SE Sicily data (shaded field) are compatible with data from the literature.

clude that although coeval sources are still available in great numbers, only a few of them paid attention to the earthquake-generated landforms, and then only with little detail. Evidently, society and culture had other priorities in those times, not to mention the limited scientific knowledge and technical possibilities, as well as the uncertain control that, for centuries, the authority had on the interior zones of Sicily (Mack Smith 1968). Based upon the material of Boschi and others (1995, 1997) and other sources, the picture of slope instability can be summarized as follows (translations ours, page numbers seldom available). ●

Earthquake of February 4, 1169: It is referred that the town of Lentini was “closed between two mountains”, that “farms . . . perished due to the mentioned earthquake” (Bernardo Maragone in Pertz 1866), and that in several places new springs were created and old springs clogged (Falcando in Siragusa 1897). ● Earthquake of December 10, 1542. It seems that damage to the towns of Sortino and Lentini was increased by slope instability. A later source refers that mills, bridges, aqueducts and roads went destroyed (Gaetani-Specchi 1695). Liquefaction phenomena took perhaps place in unidentified localities (ha-Kohen, 16th century). ● Earthquake of January 9/11, 1693. This case is better documented. A coeval, comprehensive account, both informed and interesting, is that provided by Boccone (1697), on which several later sources seem to be based. With reference to several localities (Figure 2), this author describes what can be recognized, or sometimes only interpreted, as slope instability of various type: Sortino (subsidence, falls, perhaps a slide), Melilli (ground openings), Noto

and its territory (subsidence, slides, falls), Syracuse (liquefaction), Lentini (liquefaction). The passage which follows (Boccone 1697, p. 19) seems particularly interesting, in part because it is sufficiently detailed as to permit the tentative identification of the site involved (compare with Figure 9): “Not far from the territory of Cassaro, two very great rocks detached from two spurs at the ends of two mountains, between which a river streamed through a long valley. Both rocks plunged down, and stopped in the defile of the canyon, or valley, and, coming in contact to each other, clogged the passage of the water to such an extent that it, being unable to flow out either underground or laterally, filled the valley to the elevation of the rocks accumulated, and then overflowed, leaving a lake 3 miles around and considerably deep.” The texts by Boccone (1697) and Tortora (1712), a somewhat later but local source, suggest the identification of a specific rock slide, whose reactivation in 1693 interrupted an important road and consequently had great relevance on the contested decision to move the ancient town of Noto to a new site (Figures 2 and 10). A coeval, anonymous witness seems particularly effective in describing processes and effects, although it makes no reference to specific sites (Anonymous 1693): “Many mountains were shaken, and while opening up were thrown in the air, and having then fallen down onto the plain, filled the beds of huge rivers, and these being unable to have their natural course, it is assumed, with good grounds, that by those waters it will soon befall a new sort of very notable damage.” Finally, “long crevices” and similar forms are often mentioned. To sum up, despite some “cultural limits,” these first-hand accounts outline the essentials of seismically induced processes and, specifically, document the occurrence of landslides and valley dammings. In particular: (1) In some circumstances coeval accounts may fruitfully be compared with field evidence. (2) Quite often the described landslides leave the impression of having been relatively “minor” with respect to what can be actually seen nowadays, and of having been in various degree obliterated in the meanwhile, which is consistent with the normal abundance of earthquake-triggered shallow instability (see previous section). (3) The “long crevices,” at least in several cases, testified to centimeter to meter displacements of major landslides along low-angle slip surfaces. They were normally the only symptoms


125

Figure 9. These are probably the sites referred to by Boccone (1697) in the quoted passage. The accumulation of a fall (F1) is clearly visible where the river Anapo (the most important in that zone) meanders because of two rock spurs (S1 and S2). The town of Cassaro is in the top left corner (C). The silting up deposits filling the landslide dammed basin are clearly

seen in the foreground (SD). The second rock fall mentioned is missing, but it may have been a minor failure, concealed by subsequent processes. We suggest that it detached from S2 where indentation I is now visible. About mid 20th century the dam was constructed, and preliminary investigation revealed the presence of lacustrine clays (Dal Piaz 1948, ESE 1951).

visible, since the great size of many landslides and low landscape relief prevented land-based observers from perceiving landslides in their entirety (and still nowadays hamper such perception). By comparing what is explained in this section and previously (“Low Compatibility. . .”), it is evident that information on slope instability in SE Sicily has come through two distinct channels: geological and/or engineering investigations, mostly performed in the 20th century and indicating a very low propensity to landsliding; and seismic records with eyewitness accounts of earthquakes and describing serious earthquake-induced instability. Curiously enough, these two sets of knowledge seem to have had practically no point of contact, so that while in the first the possible triggering role of seismicity seems to have been overlooked, in the second, historical evidence seems to have never been field-checked. Some causes of this dichotomy are easy to understand. In fact, the last strong earthquake was in 1693 and since then the landslides may have been simply forgotten because later they did not move anymore and in many cases they are so large that their actual nature can be perceived only through air photographs. However, such a dichotomy has also had a relevant negative consequence: it gave rise to the widespread opinion that SE Sicily was overall landslide-free or nearly so,

thus causing one of the hazards affecting the area to be underestimated. Documentary Dating of a Landslide The 1693 earthquake destroyed several Hyblaean towns, some of which were rebuilt elsewhere. This was also the case for Avola, which, from the old site surrounded by canyons at the southern edge of the Hyblaean Plateau (Avola Antica in Figure 2), was moved to the coastal plain where it still lies (Gringeri Pantano 1996). Nothing of the old town survived, and over thousand victims were counted. West of the prequake town site, the River Miranda streamed in a canyon and provided energy to some mills. Down the coastal plain it also irrigated sugarcane plantations and fed the sugar factory. Therefore, it had a great value in local economy, and when, on 11 January 1693, it was dammed by a landslide, the repercussions were particularly heavy. This is probably the main reason why, unlike other cases, this specific landslide was well remembered and, more than half a century later, was mentioned in a book of local history (Di Maria 1745). The book’s author was born and always lived in Avola; he was a friar and had extensive knowledge of people and circumstance in the town. In his book, he wrote that due to the quake (pp. 129 –130, translation ours): “Mount Gisini split across and nearly

126


Figure 10. This rock slide, probably rotational (H/L ⫽ 1:2.3, corresponding to a gradient of ⬇23°), is located just south of the place where the town of Noto was located before its destruction in 1693 (Noto Antica, NA; cf. also Figure 2). The landslide crown and head are outlined. According to our

interpretation, this is the site of the road interruption mentioned by several contemporary sources (e.g., Tortora 1712). The road still exists (arrows). Note the typical Hyblaean landscape.

half the mountain, after detaching rampageously, plunged unto the valley bed . . . and three mills with all people inside remained beneath its portentous bulk.” Figure 11 shows the failed mass emplaced across the Miranda Valley. Notice the planar and steep slip surface, a feature really uncommon in this area but which clearly explains the sudden development described by Di Maria. Due to that damming, the River Miranda ceased to flow to the plain. However, in order to again get water from its source to feed the sugar factory and newly built mills, a channel system bypassing the blockage was soon built, as archival papers document (Gringeri Pantano 1996). Having lost its main source, the river became a torrent and was renamed Torrente Pisciarello, now used in official maps. Indeed, soon after the earthquake a plan was devised to take water from the Miranda source to the new town site, but the aqueduct was built only in 1912, based on plans from 1887. In the planned piping path (Figure 12), the blockage is clearly shown, accompanied by the note “Landslide occurred in the year 1693.”

Information on Malaria Recrudescence Some historical documents show that malaria recrudescence was among the effects of the 1693 earthquake. On the basis of analogy with a better-documented case that occurred in 1783 in the nearby region of Calabria (just beyond the Straits of Messina), we suggest that it may be indirectly correlated also with mass movement. The known facts are the following: Malaria had rooted in SE Sicily long before 1693: the first mention commonly interpreted as certain is provided by the Greek historian Thucydides and is related to a siege of Syracuse in the 5th century BC. In the summer of 1693—the earthquake had occurred in January—many victims seemed to have this disease [Boccone 1697, Franzo` in Agnello 1931, some contemporary official letters of the Viceroy of Sicily to the King of Spain, now preserved in the Spanish National Archives (Archivo General de Simancas) and quoted in Boschi and others 1995, 1997]. None of these sources mentions a relationship between malaria and landslides and/or landslide dams. However, such a relationship is known to have


127

Figure 11. This is, with certainty, the planar rock slide (H/ L ⫽ 1:1.9, corresponding to a gradient of ⬇28°) that detached from Mt. Gisini and dammed the River Miranda on 11 January 1693, destroying mills and killing people. Note the crest in the accumulation, which perfectly parallels the talus in the back-

ground. A closer examination of the failed mass reveals that the rock, although heavily jointed, still maintains the original subhorizontal stratification. The dammed basin, not visible, is only partly infilled. Note here also the typical Hyblaean landscape.

occurred following the strong 1783 Calabria earthquake: in that case 215 landslide lakes were formed and provided a remarkable impulse to the prevalence of malaria (Vivenzio 1788, Cotecchia and others 1969, 1986). According to our interpretation, the SE Sicily earthquakes also may have created a great many microenvironments favorable to mosquitoes—within ground deformations of any kind both related and unrelated to mass movement. Probably no other reasonable inference can be made at the moment, but in our opinion SE Sicily presents poorly explored links between earthquakes and slope instability on the one hand and malaria on the other.

sliding (Nicoletti and others 2000). In particular, those landslides represent a danger for, among others, 4 towns, 6 hamlets, 90 rural settlements, 9 segments of state roads, 19 segment of important roads, and 6 segments of railways. In many of these sites, damage could be expected in any case, due to shaking or shallow landsliding. However, there are sites, particularly along lines of communication, in which damage by shaking and shallow landsliding is unlikely, but which nevertheless can be badly damaged because of their location upon a deep-seated, preexisting landslide (Figure 13). This is of course a strongly insidious facet of the problem, since interrupting lines of communication and other lifelines hinders and delays the rescue operations and it deserves careful consideration. Overall, it can have serious socio-economic consequences (Schuster 1996).

Implications The elements illustrated above provided the background against which the observed, and initially puzzling, landslides could be set and explained. Because of the geometry of their slip surfaces, most landslides can be reactivated, and this adds considerably to other earthquake-induced hazards such as shaking and shallow land-

Summary and Conclusions The initially unexpected discovery of old, large, dormant landslides in SE Sicily led us to investigate the

128


Figure 12. A: The plan of the Avola aqueduct, drawn in 1887. Inset is enlarged in B and shows the site of Mt. Gisini landslide (cf. Figure 11), with the annotation “Frana avvenuta nell’anno 1693” (Landslide occurred in year 1693).

details of their existence. Results are as follows: (1) In nonseismic conditions, significant slope instability is scarcely compatible with the local geoclimatic environment and, actually, it is not described in the literature. (2) Landslides usually show low-angle basal shear surfaces, despite the fact that the properties of the forming material are overall good. (3) Landslides fulfill the known relationships between earthquake magnitude and epicenter–landslide distance. (4) Sources contemporary to high-energy historical earthquakes that occurred in 1169, 1542 and 1693 provide witness to the occurrence of earthquake-triggered landsliding. (5) Documents presented in this paper correlate with certainty a specific landslide to the 1693 earthquake. This geological and historical evidence, accompanied by the absence of contrasting elements, leads us to conclude that SE Sicily is affected by a previously undetected process of earthquake-triggered landsliding. The process has implications in terms of hazard,

probably particularly insidious as regards lines of communication and other lifelines. These implications would have not been discovered, and every subsequent preventive activity would have been impossible, had there not been the initial, fortuitous finding of something that seemed at odds with what had been taken for granted till then. It is worth stressing that a historical and geological approach can be a valuable tool in the first investigative stages and alternative to a much more expensive— both in time and money— geotechnical-engineering approach, particularly if a large territory is to be examined. The only really necessary prerequisite is that the considered region has a documented history. Can this experience be extrapolated? Are there other zones of known seismicity that are commonly thought of as landslide-free or nearly so? If such zones exist, it may be advisable not to trust what perhaps only


129

Figure 13. This huge, strongly listric rock slide (H/L ⫽ 1:11.5, corresponding to a gradient of ⬇5°) involves calcareous marlstones in the northwestern part of the area and is presently crossed by a highway (black line) for more than 2 km. Published by permission of the Italian Air Force (Concessione Aeronautica Militare-R.G.S. n. 0043 del 27/01/1999).

seems to be common knowledge, but has indeed never been fully substantiated, and to check the factual conditions. This is, in our opinion, the important lesson to be drawn from this case history.

Acknowledgments We express our warm gratitude to our friends and colleagues Emilio Catalano and Giulio lovine, for useful discussions and advice.

Literature Cited Achilli, V., M. Anzidei, P. Baldi, R. Velardita, and F. Broccio. 1993. La rete geodetica GPS della Sicilia sud orientale, Pages 151–162 in E. Boschi and A. Basili (eds.), Contributi allo studio del terremoto della Sicilia orientale del 13 dicembre 1990. Publ. 537. Istituto Nazionale di Geofisica, Rome.

Adorni, G., and A. Aureli. 1989. La climatologie des Monts Hybleens (Sicilie sud-orientale) analyse´ e a des fins hydroge´ ologiques et ge´ omorphologiques. Pages 195–204 in Publications de l’ Association Internationale de Climatologie, Vol. 2. Afan De Rivera, C. 1832/42. Considerazioni sui mezzi da restituire il valore proprio a` doni che ha la natura largamente conceduto al regno delle Due Sicilie. Stamperia e Cartiera del Fibreno, Naples, 3 vol., 1295 pp. Agnello, G. 1931. Memorie inedite varie sul terremoto siciliano del 1693. Archivio Storico per la Sicilia Orientale, series II, year 7, pp. 390 – 402. Agnesi, V., and T. Lucchesi. 1987. Bibliografia geologica ragionata delle frane in Sicilia (dal 1886 al 1987). Quaderni del Museo Geologico “G. G. Gemellaro” 3. Dipartimento di Geologia e Geodesia, Universita` di Palermo, 123 pp. Anonymous. 1693. Sincera, ed esatta relazione dell’orribile terremoto seguito nell’Isola di Sicilia il di 11 di Gennaio 1693. Colla nota delle Citta`, e Terre sprofondate, de’ morti,

130


e luoghi, che hanno patito, e con tutte le particolarita` piu` degne da essere registrate. Rome.

the International Symposium on Engineering Geology Problems in Seismic Areas, 13–19 April 1986, Vol. 6. Bari, Italy.

Azzaro, R., and M. S. Barbano. 2000. Analysis of seismicity of Southeastern Sicily: Proposal of a tectonic interpretation. Annali di Geofisica 43:1–18.

Crino`, S. 1921. Distribuzione geografica delle frane in Sicilia e periodi di maggiore frequenza dei franamenti. L’Universo 6:421– 466.

Baldacci, L. 1886. Descrizione geologica dell’Isola di Sicilia. Memorie Descrittive per la Carta Geologica d’Italia 1, 403 pp.

Cruden, D. M., and D. J. Varnes. 1996. Landslide types and processes. Pages 36 –75 in A. K. Turner and R. L. Schuster (eds.), Landslides: Investigation and mitigation. Transportation Research Board special report 247. National Research Council, Washington, DC.

Baratta, M. 1901. I terremoti d’Italia. Saggio di storia, geografia e bibliografia sismica italiana. Torino. Anastatic reprint, Arnaldo Forni, Sala Bolognese (BO), 1979. Barbano, M. S. 1985a. The Val di Noto earthquake of December 10, 1542. Pages 28 –29 in D. Postpischl (ed.), Atlas of isoseismal maps of Italian earthquakes. Quaderni di “La Ricerca Scientifica” 114, 2A. CNR-Progetto Finalizzato Geodinamica, Rome. Barbano, M. S. 1985b. The Val di Noto earthquake of January 11, 1693. Pages 48 – 49 in D. Postpischl (ed.), Atlas of isoseismal maps of Italian earthquakes. Quaderni di “La Ricerca Scientifica” 114, 2A. CNR-Progetto Finalizzato Geodinamica, Rome. Barbano, M. S., and M. Cosentino. 1981. Il terremoto siciliano dell’11 gennaio 1693. Rendiconti della Societa` Geologica Italiana 4:517–522. Beccaluva, L., A. Di Grande, A. Lo Giudice, G. Macciotta, and F. Siena. 1993. Carta geopetrografica dell’area centro-settentrionale iblea, Sicilia sud-orientale, scale 1:50 000. Universita` di Catania, 2nd ed. S. EL. CA., Florence. Boccone, P. 1697. Intorno il Terremoto della Sicilia, seguito l’anno 1693. Pages 1–31 in Museo di fisica e di esperienze variato, e decorato di osservazioni naturali, note medicinali, e ragionamenti secondo i principij de’ moderni. Tipografia G. B. Zuccato, Venice. Bonaiuti, V. 1793. Continuazione dello stesso soggetto (Particolarita` intorno al tremuoto che ruino` la Sicilia nel 1693). Pages 34 – 43 in Compendio delle Transazioni Filosofiche della Societa` Reale di Londra, parte 1 (Storia Naturale), tomo 1 (Vulcani e Tremuoti). Boschi, E., G. Ferrari, P. Gasperini, E. Guidoboni, G. Smriglio, and G. Valensise. 1995. Catalogo dei forti terremoti in Italia dal 461 a.C. al 1980. Istituto Nazionale di Geofisica, Storia Geofisica Ambiente, Bologna, 973 pp. with CD-ROM. Boschi, E., E. Guidoboni, G. Ferrari, G. Valensise, and P. Gasperini. 1997. Catalogo dei forti terremoti in Italia dal 461 a.C. al 1990, 2nd ed. Istituto Nazionale di Geofisica, Storia Geofisica Ambiente, Bologna, 644 pp. with CD-ROM. Campisi, B. 1961. Note geologiche sulla regione di Cassibile e San Michele (Monti Iblei, Sicilia). Bollettino del Servizio Geologico d’Italia 81 (fasc. 2–3):287–312. Coltro, R., V. Ferrara, and F. Musarra. 1978. Studi di conservazione del suolo in alcuni bacini siciliani. Sviluppo Agricolo, Supplemento Speciale 1:38 pp. Cotecchia, V., G. Travaglini, and G. Melidoro. 1969. I movimenti franosi e gli sconvolgimenti della rete idrografica prodotti in Calabria dal terremoto del 1783. Geologia Applicata e Idrogeologia 4:1–24. Cotecchia, V., A. Guerricchio, and G. Melidoro. 1986. The geomorphogenetic crisis triggered by the 1783 earthquake in Calabria (Southern Italy). Pages 245–304 in Proceedings of

D’Addezio, G., and G. Valensise. 1993. Metodologie per I’individuazione della struttura sismogenetica responsabile del terremoto del 13 dicembre 1990. Pages 151–162 in E. Boschi and A. Basili (eds.), Contributi allo studio del terremoto della Sicilia orientale del 13 dicembre 1990. Pubblicazione 537. Istituto Nazionale di Geofisica, Rome. Dal Piaz, G. 1948. Relazione sulla presa di S. Nicola sull’alto Anapo. Unpublished report. ENEL Archive, Palermo, 5 pp. Di Grande, A., and M. Grasso. 1977. Lineamenti geologici del margine occidentale ibleo nei dintorni di Comiso-Ragusa (Sicilia). Bollettino della Societa` Geologica Italiana 96:209 –224. Di Maria, F. 1745. Ibla rediviva. 1989 Reprint. Pro Loco, Avola. ESE 1951. Impianto idroelettrico sull’Anapo. Bollettino d’Informazione 3. Ente Siciliano per I’Elettricita`, Catania, 33 pp. Ferrara, F. 1810. I Campi Flegrei della Sicilia e delle isole che le sono intorno e descrizione fisica e mineralogica di queste isole. Stamperia dell’Armata Britannica, Messina, 424 pp. Finetti, I., F. Lentini, S. Carbone, S. Catalano, and A. Del Ben. 1996. II sistema Appennino Meridionale-Arco Calabro-Sicilia nel Mediterraneo Centrale: Studio geologico-geofisico. Bollettino della Societa` Geologica Italiana 115:529 –559. Finley M. I. 1968. A history of Sicily. Ancient Sicily to the Arab conquest. Chatto & Windus, London. Gaetani-Specchi. 1695. Scritto sulle gabelle e sui diritti baronali. Archivio di Stato di Siracusa, Archivi Privati, b. 60, fasc. 1, cc. 11 and 18. Galli, P., and F. Meloni. 1993. Nuovo catalogo nazionale dei processi di liquefazione avvenuti in occasione dei terremoti storici in Italia. II Quaternario 6:271–292. Gasperini, P., and G. Ferrari. 1997. Stima dei parametri sintetici: nuove elaborazioni, Pages 56 – 64 in Boschi and others, Catalogo dei forti terremoti in Italia dal 461 a.C. al 1990, 2nd ed. Istituto Nazionale di Geofisica, Storia Geofisica Ambiente, Bologna. GCO. 1988. Guide to rock and soil descriptions. Geoguide 3. Geotechnical Control Office, Civil Engineering Services Department, Hong Kong. Ghisetti, F., and L. Vezzani. 1980. The structural features of the Iblean Plateau and of the Mount Judica area (southeastern Sicily): A microtectonic contribution to the deformational history of the Calabrian Arc. Bollettino della Societa` Geologica Italiana 99:57–102. Grasso, M. 1997. Carta geologica del settore centro-meridionale dell’Altopiano Ibleo (provincia di Ragusa, Sicilia sudorientale), scale 1:50.000. Universita` di Catania, Istituto di Geologia e Geofisica. SELCA, Florence.


Gringeri Pantano, F. 1996. La citta` esagonale. Avola: I’antico sito, lo spazio urbano ricostruito. Sellerio, Palermo, 269 pp. Gruppo di Lavoro CPTI. 1999. Catalogo parametrico dei terremoti italiani. Istituto Nazionale di Geofisica, Gruppo Nazionale per la Difesa dai Terremoti, Storia Geofisica Ambiente, Servizio Sismico Nazionale, Bologna, 92 pp. http:// emidius.itim.mi.cnr.it/CPTI Guidoboni, E., A. Comastri, and G. Traina. 1994. Catalogue of ancient earthquakes in the Mediterranean area up to the 10th century. Istituto Nazionale di Geofisica, Storia Geofisica Ambiente, Bologna, 504 pp. ha-Kohen, Y. 16th century. Letter of Yosef ha-Kohen to Yishaq ha-Kohen. Manuscript, letter collection KM 55, The Jewish National and University Library, Jerusalem.

131

Ministero dei Lavori Pubblici. 1964. I movimenti franosi in Italia. Consiglio Superiore dei Lavori Pubblici, Direzione Generale dell’A.N.A.S., Technical Service, Rome, 174 pp. Mongitore, A. 1743. Istoria cronologica de’ terremoti di Sicilia. Pages 345– 445 in Della Sicilia ricercata nelle cose piu` memorabili, 2. Montanari, G. 1941. Notizie sui movimenti franosi in Sicilia con speciale riferimento alle zone del latifondo sottoposte a colonizzazione. Annali dei Lavori Pubblici 2:94 –112, 184 –194. Nicoletti, P. G., and E. Catalano. 2000. Frane e sbarramenti fluviali di antica origine sismica in Sicilia sud-orientale—III parte: ricognizione sul Foglio 274 —Siracusa. Bollettino della Societa` Geologica Italiana 119:97–104.

Hansen, W. R. 1965. Effects of the earthquake of March 27, 1964, at Anchorage, Alaska. US Geological Survey Prof. Paper 542-A. US Government Printing Office, Washington, DC, 68 pp.

Nicoletti, P. G., G. Adorni, and A. Scalzo. 1999a. Frane e sbarramenti fluviali di antica origine sismica in Sicilia sud-orientale—I parte: Inquadramento generale e ricognizione sul Foglio 277 Noto. Bollettino della Societa` Geologica Italiana 118:461– 477.

Jappelli, R., and A. Musso. 1981. Slope response to reservoir water level fluctuations. Pages 437– 442 in Proceedings of the X International Conference on Soil Mechanics and Foundation Engineering, Vol. 3. Stockolm.

Nicoletti, P. G., G. Adorni, and A. Scalzo. 1999b. Frane e sbarramenti fluviali di antica origine sismica in Sicilia sudorientale—II parte: ricognizione sul Foglio 276 —Ragusa. Bollettino della Societa` Geologica Italiana 118:513–522.

Jibson, R. W. 1996. Use of landslides for paleoseismic analysis. Engineering Geology 43:291–323.

Nicoletti, P. G., G. Iovine, and E. Catalano. 2000. Frane e sbarramenti fluviali di antica origine sismica in Sicilia sud-orientale—IV parte: ricognizione sul Foglio 273—Caltagirone e conclusioni generali. Bollettino della Societa` Geologica Italiana 119:451–472.

Jibson, R. W., and D. K. Keefer. 1988. Landslides triggered by earthquakes in the central Mississippi Valley, Tennessee and Kentucky. US Geological Survey Professional Paper 1336-C. US Government Printing Office, Washington, DC, 24 pp. Jibson, R. W., and D. K. Keefer. 1993. Analysis of the seismic origin of landslides: Examples from the New Madrid seismic zone. Geological Society of America Bulletin 105:521–536.

Passerini, G. 1922. Le “valanghe” di Monte San Giorgio presso Caltagirone nel processo di formazione e di transformazione e nella tecnica della sistemazione. Bollettino dell’Istituto Agrario di Scandicci 8:225–246.

Keefer, D. K. 1984. Landslides caused by earthquakes. Geological Society of America Bulletin 95:406 – 421.

Perrey, A. 1848. Me´ moires sur le tremblement de terre de la pe´ ninsule italique. Me´ moires Couronne´ s et Me´ moires des Savants Etrangeres de I’Acade´ mie Royale de Belgique 22. Bruxelles.

Lentini, F. 1984. Carta geologica della Sicilia sud-orientale, scale 1:100 000. Universita` di Catania, Istituto di Scienze della Terra. S.EL.CA., Florence.

Pertz, K. 1866. Monumenta Germaniae Historica. SS. tomo 19, Hannover, Germany, pp. 236 –266.

Lentini, F. 1986. Carta geologica del settore nord-orientale ibleo, scale 1:50 000. Universita` di Catania, Istituto di Scienze della Terra. S.EL.CA., Florence. Lentini, F., S. Carbone, S. Catalano, and M. Grasso. 1996. Elementi per la ricostruzione del quadro strutturale della Sicilia orientale. Memorie della Societa` Geologica Italiana 51:179 –195. Liguori, V. 1977. La difesa del suolo in Sicilia: aspetti geologici. II Mediterraneo 7/8:3–24. Liguori, V. 1988. Stato di dissesto del territorio siciliano. Quaderni della Lega per I’Ambiente 2. Universita` Verde di Siracusa, Syracuse, 38 pp. Lombardo, G. 1985. The Catania earthquake of February 4, 1169. Pages 12–13 in D. Postpischl (ed.), Atlas of isoseismal maps of Italian earthquakes. Quaderni di “La Ricerca Scientifica” 114, 2A. CNR-Progetto Finalizzato Geodinamica, Rome.

Schuster, R. L. 1996. Socioeconomic significance of landslides. Pages 12–35 in A. K. Turner and R. L. Schuster (eds.), Landslides: Investigation and mitigation. Transportation Research Board special report 247. National Research Council, Washington, DC. Sciuto Patti, C. 1896. Contribuzione allo studio dei terremoti in Sicilia. Atti Accademia Gioenia Scienze Naturali in Catania, s. IV 9:1–34. Siragusa, G. B. 1897. Fonti per la storia d’Italia. SS., sec. XII, 22, Rome. Tortora, F. 1712. Breve notitia della citta` di Noto prima e dopo il terremoto del 1693. Pages 11– 60 in F. Balsamo (ed.), Noto nelle cronache settecentesche di Filippo Tortora e Ottavio Nicolaci. Istituto per la Storia e la Valorizzazione di Noto e delle sue Antichita`, Noto (SR), 1972.

Mercalli, G. 1883. Vulcani e fenomeni vulcanici in Italia. Milano. Anastatic reprint, Arnaldo Forni, Sala Bolognese (BO) 1981.

Vivenzio, G. 1788. Istoria de´ tremuoti avvenuti nella Provincia della Calabria Ulteriore, e nella Citta` di Messina nell’anno 1783. E di quanto nella Calabria fu fatto per lo suo risorgimento fino al 1787. Preceduta da una teoria ed istoria generale dei tremuoti, 2nd ed, vol. 2. Stamperia Reale, Neaples.

Ministero dell’Agricoltura e delle Foreste. 1976. Carta della Montagna. Monografie Regionali, 19, Sicilia. Geotecneco, Rome, 301 pp.

Wieczorek, G. F. 1984. Preparing a detailed landslide-inventory map for hazard evaluation and reduction. Bulletin of the Association of the Engineering Geologists 21:337–342.

Mack-Smith, D. 1968. A history of Sicily. Medieval Sicily: 800 – 1713; Modern Sicily: after 1713. Chatto & Windus, London.