Journal of Volcanology and Geothermal Research 113 (2002) 429^442 www.elsevier.com/locate/jvolgeores Stromboli: a natural laboratory of environmental science Massimo Chiappini a; , Giovanni P. Gregori b , Gabriele Paparo c , Carlo Bellecci d , Gino M. Crisci e , Giuseppe de Natale f , Paolo Favali a , Iginio Marson g , Antonio Meloni a , Bruno Zolesi a , Enzo Boschi a a Istituto Nazionale di Geo¢sica e Vulcanologia (INGV), Vigna Murata 605, Rome, Italy b IFA-CNR, via Fosso del Cavaliere 100, 00133 Rome, Italy c IDAC-CNR, Rome, Italy d II Universita' di Roma, Tor Vergata, Rome, Italy e UNICAL, Arcavacata di Rende, Cosenza, Italy f Osservatorio INGV Vesuviano, Naples, Italy g OGS, and Universita' di Trieste, Trieste, Italy Received 4 February 2000; accepted 13 November 2000 Abstract The science of environment is per se multi- and inter-disciplinary. It is not possible to separate the role of the physical, chemical, biological, and anthropic factors, respectively. Research must therefore rely on suitable natural laboratories, where all different effects can be simultaneously monitored and investigated. Stromboli is a volcanic island slightly North of Sicily, within a tectonic setting characterised by a Benioff zone, curved like a Greek theatre, opened towards the Tyrrhenian Sea, with deep earthquakes. Moreover, it is a unique volcano in the world in that since at least V3000 years ago, it has exploded very regularly, about every 15^20 min. Hence, it is possible to monitor statistically phenomena occurring prior, during, and after every explosion. The Istituto Nazionale di Geofisica e Vulcanologia (INGV) has recently established a permanent Laboratory and an extensive interdisciplinary programme is being planned. A few main classes of items are to be considered including: (1) matter exchange (solid, liquid, gas, chemistry); (2) thermal and/or radiative coupling; (3) electromagnetic coupling; (4) deformation; (5) biospheric implications; and (6) anthropic relations since either the times of the Neolithic Revolution. Such an entire multidisciplinary perspective is discussed, being much beyond a mere volcanological concern. We present here the great heuristic potential of such a unique facility, much like a natural laboratory devoted to the investigation of the environment and climate. < 2002 Elsevier Science B.V. All rights reserved. Keywords: Stromboli; natural laboratory; environmental science 1. Introduction * Corresponding author. Tel.: +39-6-5186-0313; Fax: +39-6-504-1181. E-mail address: chiappini@ingv.it (M. Chiappini). At present, the limits of science are biased by an excessive separation into disciplines. In particular, the study of the environment su¡ers from such a drawback, depending on the impossibility of separating the role of every speci¢c e¡ect inde- 0377-0273 / 02 / $ ^ see front matter < 2002 Elsevier Science B.V. All rights reserved. PII: S 0 3 7 7 - 0 2 7 3 ( 0 1 ) 0 0 2 7 6 - 1 VOLGEO 2385 6-5-02 430 M. Chiappini et al. / Journal of Volcanology and Geothermal Research 113 (2002) 429^442 pendent of the other e¡ects. The scope of the present paper is the discussion of the heuristic potential of a natural facility, a dedicated laboratory, such as the Stromboli island, a unique volcano in the world in that it has, for at least the last three millennia, exploded very regularly, about every 15^20 min. The present paper, however, is not concerned with volcanology, that is only part of such an entire investigation: within such a complex scenario, every discipline of Earth Science has its own limited and marginal, although relevant, role. It is a presentation of multidisciplinary and interdisciplinary ideas, based on extensive discussion and implicit agreement between several specialists from di¡erent disciplines, as a way of conceiving an integrated strategy for tackling fundamental unsolved general problems. Therefore, short mention must be made of the origin of such an articulated set of often unprecedented and/or sometimes even provocative concepts. Section 2: Climate, recalls some key aspects of climatology. Section 3: Coupling across the Earth’s surface, is concerned with the coupling across the Earth’s surface, with speci¢c reference to the surface layers at the boundaries between solid Earth, sea/ocean, and atmosphere, where several crucial phenomena controlling environment may occur. Three large sets of phenomena can be distinguished, i.e. (1) matter exchange (solid, liquid, gas, chemistry); (2) thermal and/or radiative coupling; and (3) electromagnetic coupling. The last item appears very intriguing, and seems to be crucial for the improvement of our ultimate comprehensive understanding. One related perspective deals with the origin of the magnetic ¢eld of the Earth and with some concepts formerly originated from consideration of the energy balance of the tidal interaction acting on a realistic Earth model. Section 4: Volcanism, the western Mediterranean, and Stromboli, recalls a few aspects of volcanology, with particular reference to the western Mediterranean basin, and to the speci¢c tectonic setting of Stromboli. Section 5: Natural laboratories and active experiments discusses some speci¢c concepts of active modi¢cation of the environment, and natural facilities used like dedicated multidisciplinary labo- ratories. Section 6: The Stromboli Laboratory: a list of observations, is a list of the several monitoring techniques that in principle are planned for application in the Stromboli Laboratory. Section 7: Conclusion, gives some conclusive remarks. 2. Climate The conventional separation of Earth Science into disciplines basically relies on the distinction between solid Earth, oceans (including seas and eventually the entire hydrosphere), atmosphere, biosphere, space, etc. The most di⁄cult items, however, are concerned with their respective boundaries, although only a few of them have as yet been adequately investigated. The magnetosphere, as the boundary between atmosphere and space (Kamide, 1988; Jacobs, 1991; Brekke, 1997; Gregori, 1998, 1999), and the layer within the atmosphere being the boundary between atmosphere and Earth’s surface, are objects of a conspicuous literature. But, much less concern was devoted to either one of the ‘surface layers’ that are some thin (a few metres) layers either in the atmosphere, or within soil, or within sea/ ocean, where most critical phenomena occur that ultimately control the entire system. Also phenomena across the boundary between sea/ ocean and their respective £oor have, as yet, been inadequately investigated. For instance, the greenhouse e¡ect is generally conceived in terms of the gentle cover composed of atmospheric water (or of other greenhouse atmospheric constituents). One can also consider a greenhouse effect in the ionosphere (Kazimirovsky, 2000). The conventional separation of Earth Science into disciplines should always be considered as a temporary and partial way of approaching understanding. Every scientist should consider that his own investigation always has to be focused on its integration with other disciplines, with some kind of fusion of the knowledge exploited by di¡erent specialists. In the opposite case every understanding remains in the realm of academic speculation contributing no ultimate progress of comprehensive science. VOLGEO 2385 6-5-02 M. Chiappini et al. / Journal of Volcanology and Geothermal Research 113 (2002) 429^442 3. Coupling across the Earth’s surface Climate is a phenomenon whose surface layers within either the soil, or the sea/ocean, or the atmosphere are the keys for understanding its ultimate processes and mechanisms. Three classes of interaction can be considered : (1) gas, liquid, and solid matter exchange, and chemical coupling; (2) thermal and/or radiative coupling; and (3) electromagnetic coupling. 3.1. Gas, liquid, and solid matter exchange, and chemical coupling Atmospheric gases are exchanged between atmosphere and soil, the gas exhalation from ground re£ects either the air formerly penetrated underground, or the gas released by endogenous processes, or the exhalation from buried organic matter. Such phenomena can be investigated either by attempting to reproduce, within the laboratory, several di¡erent natural environments, or by monitoring the characteristics of air masses after their crossing over di¡erent kinds of the Earth’s surface (e.g. oceans, deserts, etc.). The multiannual variation, integrated all over the world, of gas exhaling from volcanic or geothermal areas as a key factor for the control of climate has never been monitored. The chemical, isotopic, and mass balance of matter exchanges across the di¡erent surface layers appears one of the main missing gaps in the present understanding of environmental science. 3.2. Thermal and/or radiative coupling Heat transport may occur through : (1) conduction, although in general soil has a very low thermal conductivity ; (2) advection by £uids, that include £ux of atmospheric gases underneath the Earth’s surface, water circulation or, whenever applicable, also lava; or (3) radiative processes. The prime energy feeding derives from: (1) endogenous energy; (2) solar radiation ; (3) heat transported by advection; or (4) radiative processes. Several such e¡ects maybe controlled by the biosphere. For instance, Otterman et al. (1993) observed in the Negev desert a lesser change of 431 colour of the desert, following the growth of some lichens or fungi or other, apparently capable of triggering rain precipitation with a statistically signi¢cant increase. They claim that such an e¡ect cannot be explained by evapo-transpiration by vegetation, rather in terms of a change of the thermal inertia of soil. Such an intriguing observation envisages that, maybe, a change by some very minor amount of the thermal regime very close to the surface could be capable of triggering some microvorticity. This is the ¢rst attempt at an interpretation. In any case, there is a need for monitoring gas and/or ion/electron exchange between the atmosphere and the soil surface, water £ow underground, underground temperature and electrical conductivity c vs. depth, water precipitation from air on the ground, microvorticity of air within some very thin surface layer immediately above soil. There is an endless variety of di¡erent morphological con¢gurations that ought to be speci¢cally studied in order to apply their respective inference for the study of climate and environment. 3.3. Electromagnetic coupling The electromagnetic interaction between the Earth and the solar wind appears to be one of the most di⁄cult problems of Earth Science. Even the same origin of the magnetic ¢eld of the Earth or of its endogenous energy budget appears extremely di⁄cult. It is customary to claim that the solar wind cannot a¡ect the Earth’s deep processes due to electromagnetic screening by the mantle. Such a hypothesis is worthy of further investigation. The underground electrical structure is, in general, quite heterogeneous. Hence, a geo^electromagnetic phenomenon of any kind, observed on the occasion of a given earthquake occurring within a given area, is hardly expected to be repeated anywhere else and/or at any other time, even at the same site. We should ¢rst investigate in much greater detail the heterogeneity of deep and shallow earth, prior to attempting or making any inference about any kind of electromagnetic precursor of earthquakes. In general, the system has a very large number of degrees of freedom, VOLGEO 2385 6-5-02 432 M. Chiappini et al. / Journal of Volcanology and Geothermal Research 113 (2002) 429^442 and we will never be capable of obtaining adequate information about any consistent faction. Every monitored phenomenon only re£ects some erratic and basically random information, which is physically signi¢cant and indicative only, as far as it is concerned, with the evidence of some change which is in progress. However, that can never be strictly expected to give a permanent, reliable, and quantitative monitoring of the state of the system. One should, therefore, attempt and carry out monitoring that should be widespread by using a large array of recording points along with a large set of di¡erent techniques. The aim is to get rid of the implicit and unavoidable limitation by which no recording can be considered alone as being reliable and adequately complete. In conclusion, earthquake precursors probably do exist, although no strict rule will ever be attained relating their respective nature and intensity, to their time advance with respect to the shock, and/or to the seismic magnitude. Rather, we should simply feel satis¢ed by detecting the changing trend of the system, and we should just try to model, as far as possible, the speci¢c system for the area of concern. Air^earth currents, spontaneous potentials, geo^electromagnetic behaviour of volcanic plumes, geomagnetic anomalies in volcanic areas, the long-period solar modulation of volcanic activity and of the deep structure of the Earth, the unexplained correlation between the geomagnetic secular variation and the anomalies of the spin rate of the Earth are observational items, all likely related to each other. We want to emphasise the fundamental lack of any adequate understanding of some basic phenomena that play a key role in the control of parameters controlling the evolution of natural and environmental phenomena. 4. Volcanism, the western Mediterranean, and Stromboli 4.1. Items in volcanism (1) The origin of the energy supply to volcanoes is controversial. Several authoritative volcanolo- gists claim that the hypothesis of endogenous radioactivity and/or slow phase transformation appears hardly believable. Historically (Gregori and Dong, 1996), the most ancient idea considered wind friction inside underground caverns. During the 18th century, they hypothesised underground combustion of oil. This was later substituted by the hypothesis of radioactivity and of the slowphase transformation of fossil energy, since the time of Earth’s formation. Both on Stromboli’s island and elsewhere, soil deformation, ground porosity, and gas exhalation either from ‘normal’ ground or from fumaroles or both, ought to be related to the water content within ground, hence to its c, and also to the vegetation cover and to its vegetative stage. This is a delicate and as yet inadequately understood problem of the interaction across soil surface. A comparison of the same kind of recordings collected in tectonically and morphologically di¡erent areas ought to give some relevant indication for a more concrete understanding of the ultimate processes that control climate, seismic and volcanic phenomena. From such a viewpoint, Stromboli needs to be monitored for several years prior to exploiting its entire heuristic potential. (2) Every volcano is a case history on its own, and no general rule can be given. Volcanism is originated by an excess endogenous heating (whatever its origin may be) that causes an increase in pressure that must be released. Such a process occurs ¢rst of all through water, gas, etc. that have great mobility and can transport heat by advection. Heat must accumulate and pressure increase. Whenever the lithostatic pressure is adequately low, a new £uid, i.e. lava, is eventually formed acting much like all other £uids. Such a lava liquefaction process is similar to what happens in a blast furnace. Under other circumstances, when the heat and pressure increase occurs at some comparatively greater depth, no lava can be formed due to the constraints by equation-ofstate, and the volcano warms up much like either a pressure cooker or the boiler of a steam engine. Di¡erently stated, in such a case, pressure increases faster than temperature, while the equation-of-state forbids liquefaction. Finally, the volcano will eventually explode. Therefore, every VOLGEO 2385 6-5-02 M. Chiappini et al. / Journal of Volcanology and Geothermal Research 113 (2002) 429^442 speculation about the ‘magma chamber’ of a volcano can sometimes be misleading. It is often speculated that a magma chamber is a reservoir that always exists, and that either originates as a lava emplacement, or in any case it is the prime heat source for a phreatic explosion. In contrast, a heat source always exists, but there appears to be no sound reason by which it should be always associated with a real magma chamber. For instance, a few years ago, the crisis of the Phlegrean Fields displayed an upheaval of a large area by almost 2 m in 1 yr. The most convincing explanation of such in£ation and subsequent de£ation was in terms of a hydrothermal, rather than a magmatic, process. A related item deals with the speculated magma chamber of mid-ocean ridges (¢rst discussed by (Gregori and Lanzerotti, 1982), that a few years later could not be found by standard ocean bottom geo^electromagnetic sounding, magnetotellurics (MT) and other (Constable, 1990 and references therein), either by searching for an electromagnetic coupling with a transoceanic communication cable (Meloni et al., 1984), or by means of the polarisation of the perturbations recorded within the network of geomagnetic observatories (Gregori et al., 1987). (3) An important drawback of present volcanology is the lack of monitoring of submarine volcanism (Simkin and Siebert, 1994), notwithstanding the geologic time scale, in terms of outpoured volume, it appears that it is responsible for V3/4 of global volcanism. Such routine monitoring could be achieved either by recording the associated minor seismic activity by means of some unrealistic and very dense network of sea-bottom seismometers. Otherwise, the currents induced into the network of the commercial submarine communication cables could provide (Meloni et al., 1983; Lanzerotti and Gregori, 1986), after suitable calibration, e¡ective real-time and hightime resolution monitoring, although restricted to several spatial bands all over the Earth’s surface, and with a limited spatial resolution. (4) Chemical geodynamics (e.g. Wilson, 1989) was started and de¢ned only a few decades ago. A successful classi¢cation of ocean £oor basalt was attained by considering the isotope concentration ratios 206 Pb/204 Pb, 207 Pb/204 Pb, 208 Pb/ 433 204 Pb, 87 Sr/86 Sr, and 143 Nd/144 Nd, plus a few others for which, however, some comparatively more limited and as yet inadequate data set is available. This was interpreted by speculating on di¡erent reservoirs that ought to exist, by di¡erent extension and possibly at di¡erent depths, in different regions. All observations ought thus to be the consequence of the output of primary components originated from such reservoirs, and by subsequent basalt contamination, while they cross the crust. An alternative interpretation may consider rather the space-varying prime heat intensity, by which the melting of the basalt occurs at a di¡erent depth depending on the intensity of such a prime heat supply. Hence, the outpouring basalt samples the isotopic chemistry of a layer at a di¡erent depth in di¡erent areas. In this way, the prime heat intensity, chemical geodynamics, magma £uidity, and hence the island topographic height (or equivalently the bathymetry of a seamount) do, in fact, depend on each other according to some apparently regular and simple law. (5) Volcanism may control climate and environment. A great concern deals is the impact of catastrophic explosions (e.g. Pinatubo, Unzen, etc.) by analogy to nuclear winters, etc. (Rampino et al., 1988; Pittock et al., 1988; Harwell and Hutchinson, 1988). The problem, however, is very complex. 4.2. The west Mediterranean volcanism Tectonics of the Mediterranean basin is very complex (Mantovani, 1997; Mantovani and Albarello, 1997; Mantovani et al., 1997a,b,c; Viti et al., 1997, and references therein). The portion of peninsular Italy west of the Apennines displays relevant geothermal and volcanic features. The Tyrrhenian Sea £oor has the character of an ocean (Fig. 1), with a great number of seamounts (and corresponding magnetic anomalies), volcanoes active during the last few million years and elevating from an abyssal plain down to a depth of V4000 m. Sicily is part of the African plate and it pushes against peninsular Italy, being the likely cause of the Italian seismicity. The Sicily channel (or Pantelleria graben) is shallow (a few hundred metres deep), and includes the volcanic VOLGEO 2385 6-5-02 434 M. Chiappini et al. / Journal of Volcanology and Geothermal Research 113 (2002) 429^442 Fig. 1. Magnetic anomalies of the Italian region. Notice the conspicuous anomalies on the bottom of the Tyrrhenian Sea, the most of which are associated with volcanic seamounts. After Chiappini et al. (2000). VOLGEO 2385 6-5-02 M. Chiappini et al. / Journal of Volcanology and Geothermal Research 113 (2002) 429^442 islands of Pantelleria (EsperancPa and Crisci, 1995), and Linosa. Another island, named Ferdinandea (by the Sicilian government of that time), or Graham (by the British £eet), or Julie (by the French geologist Constant Pre¤vost), was born on 13 July 1831, but it soon eroded, until it disappeared on 28 December 1831. At that time, it was the object of considerable international controversy, due to its strategic location. A V2000-m-deep escarpment is located on the prolongation towards Africa of the eastern Sicilian coast. Vesuvius and Etna are comparatively close to each other, and are probably the best historically documented volcanoes of the world, having, however, a completely di¡erent character. Vesuvius has often been explosive, with varying cycles of the order of almost V7 centuries’ duration. Every such long cycle begins by three periods of apparent repose, followed by a ‘spectacular’ period characterised by e¡usive eruptions alone. At the end of every ¢rst and third such ‘repose’ period, it experienced a sub-Plinian explosive eruption (i.e. producing more than 0.1 km3 of ejecta). The origin of the energy supply for either Vesuvius or Etna is still unknown. The Messina Strait was carefully monitored and shown to have a fairly constant width. Upon consideration of the aforementioned pressure by Sicily against the Italian peninsula, some highly e⁄cient pivot or gear should be located within such an area, originating from some conspicuous local friction that must be released, thus possibly explaining Etna’s, although not Vesuvius’ character. Another most relevant volcanic feature is the Phlegrean Fields slightly NW of Naples (Rosi and Sbrana, 1987). They are a nested caldera resulting from the collapses of the so-called Campanian Ignimbrite (36 kyr BP) and Neapolitan Yellow Tu¡ (12 kyr BP). They have been the site of both widespread volcanism and block resurgence. The younger volcanism occurred in two periods between 10 and 8 and 4.5 and 3.7 kyr BP. One famous episode occurred in 1538 AD from which a new mountain originated, the Monte Nuovo. Moreover, between 1969 and 1972, great concern was caused by an uplift of V1.7 m, with a peak rate of 0.02 mm day31 (de Natale, 1995). Since 435 December 1984, subsidence occurred, and between the end of 1988 and April 1989, a minor episode of uplift (maximum 7.5 cm) was observed (Civetta et al., 1995). The Aeolian archipelago slightly North of Sicily is composed of several islands renowned since classic antiquity. This area is particularly intriguing from the archaeological point of view. It was, in fact, crucial for investigating the former occupation of the Mediterranean region by mankind, either at the time of the Neolithic Revolution or after the deserti¢cation of the Sahara. This event compelled its former inhabitants to emigrate, giving rise to the Basques, the Etruscans, the Minoans (Arna'iz Villena and Garc|'a, 1998). From such a perspective, it appears to be a still unexploited important archaeological resource. The ancient Greeks knew about active volcanoes only occurring in western Mediterranean, and knew about the existence of huge caverns underground. Hence, it appeared reasonable to correlate caverns, winds, air friction underground, and volcanism (Gregori and Dong, 1996). The entire morphology of the Aeolian archipelago recalls an island arc, which, in fact, it is named in this way. But, its basalt isotopic signature recalls ocean £oors, rather than island arcs. Crisci et al. (1991) and EsperancPa et al. (1992) proved the time variation, on the geologic time scale, of its basalt isotopic chemistry. This aspect seems to have never been focused in any other volcanic island. A few more recent isotope ratio measurements for the Vulcano island are reported by Pinarelli et al. (1995). The island of Vulcano had an exceptionally great activity documented by geologic evidence during V1250^550 BC, and historically reported eruptions during the 5th century BC (Gregori, 1997). In the late 1980s, and early 1990s, some serious concern arose about such an island (Baubron et al., 1990, 1991; Italiano et al., 1998) due to a signi¢cant large increase of gas exhalation observed from ground, being a possible precursor of a restarting activity. The phenomenon later faded o¡. Frazzetta and La Volpe (1991) studied its history, morphology and cycles, while concerning a 6th century eruption, see Gioncada et al. (1995) and references therein. VOLGEO 2385 6-5-02 436 M. Chiappini et al. / Journal of Volcanology and Geothermal Research 113 (2002) 429^442 4.3. Stromboli The Stromboli Island is a unique volcano of the world in that about every 15^20 min it has an explosion (typically called ‘Strombolian type’), and it is possible to monitor its entire system and to investigate the processes that precede, prepare, develop, and ¢nally exploit every such explosion. Every explosion is the start of the stopwatch for a new experiment. Di¡erent explosions are never exactly the same. There is a unique and great heuristic potential related to the possibility of investigating phenomena with large statistics. The radiochronological dating of its ejecta during the last V100 000 yr was given by Gillot and Keller (1993). The energy supply to Stromboli is of some concern. Upon a formal estimate (Giberti et al., 1992) it appears that, if the origin is some continuous heat-re¢ll by a steady lava £ow (operated much like a cooling circuit that warms water and triggers volcanic explosions), one should expect a £ow of V200 kg s31 . It seems hard to believe that such an impressive regularity persisted for several millennia. On the other hand, suppose that the explanation is in terms of a subduction mechanism, occurring within such an unusual and peculiar ‘island arc’ (provided that it can be likened to a typical island arc of the Paci¢c and/or Atlantic oceans). In such a case, such a mechanism ought to be capable of explaining its explosive V15^20 min regularity, compared to the several centuries activity of its close islands in the same archipelago. Moreover, whatever the origin is, is the supply of Stromboli via either friction, or Joule heating, or something else? Chemical geodynamics can provide, perhaps, some hunches concerning the analogies and/or di¡erences between the Aeolian archipelago and a standard island arc. 5. Natural laboratories and active experiments The concept of a natural laboratory, opposed to the concept of a standard indoor laboratory, is concerned with every case history in which either a given system is ‘almost’ isolated with respect to the other systems, or some phenomena can be observed under di¡erent conditions For instance, Stromboli is a system that is well de¢ned and substantially ‘isolated’ with respect to other emerged lands, and it allows the use of large statistics in that its regular and frequent activity enable the same experiment to be repeated as many times as is necessary under slightly di¡erent conditions. Hence, it is a unique facility for carrying out a feasible observational programme for investigating several mechanisms. The same argument can be applied to every island, sea/ocean basin, lake, valley, or catchment area of every given river system, etc. The second aforementioned occurrence of a ‘natural laboratory’ can be better explained in terms of the following two examples, where the ‘natural laboratory’ is the ensemble of several sites altogether as follows. The investigation, e.g. by means of optical techniques of the quality of air masses after their crossing over large extensions of a speci¢c surface (e.g. steppe, prairie, ocean, desert, permafrost, ice, etc.), can be accomplished by means of routine observations from a network of high-altitude atmospheric observatories. This is one example where the ensemble of all such observatories is the natural laboratory, while any kind of even greatly sophisticated recordings collected at one observatory alone could hardly provide any information comparable with the multi-observatory comparison. The other example is concerned with the recordings of acoustic emissions (AE) monitored, e.g. at 200 kHz, in di¡erent environments. Fig. 2 shows recordings on Gran Sasso, the highest mountain of the Apennines in central Italy that is mainly composed of dolomite, marl, and limestone. The two curves show the output of transducers for AE at 200 kHz inside rock and air, respectively. The transducer is located on a steel rod entering into ground for V12 m. The signal originates within ground, as a consequence of the daily thermal and/or tidal soil deformation. The thermal e¡ect occurs only within some shallow layer (several tens of centimetres), while the tidal deformation a¡ects the entire mountain. The difference between curves A and B shows that the VOLGEO 2385 6-5-02 M. Chiappini et al. / Journal of Volcanology and Geothermal Research 113 (2002) 429^442 437 Fig. 2. Recordings in the area of the Gran Sasso in central Italy, during May 29 through June 6, 1996. It shows the output of AE transducers at 200 kHz both inside rock and in air, and the natural electric ¢eld recorded by using a metal rod inserted into ground for V12 m and used as a receiving antenna tuned on the same frequency. The AE appear to be much larger during the daily cooling of soil occurring during local nighttime. e¡ect has no perturbation of external origin. For additional con¢rmation, the natural electric ¢eld was measured by using the same metal rod inserted into soil, and operated as a receiving antenna tuned on the same frequency 200 kHz. Such a natural electric ¢eld is well known to originate by natural sources (ionosphere and its corresponding induced telluric currents underground), and it shows the expected large diurnal variation being maximum during day time. The likely interpretation is that the AE signal shows a maximum during night-time, when soil contracts and cools o¡. AE recordings were collected on the Stromboli island. A several month (discontinuous) data set is available, and extensive analysis is in progress. In general, an invariant law is that an AE burst is observed a few minutes prior to every explosion (Diodati, 1994). AE recordings at low (25 kHz) and high (160 kHz) frequency are being collected on Vesuvius at a site located a few hundred metres from the rim of the crater. The transducer is put on top of a glass rod inserted for V50 cm into a lava dike. Glass ensures the absence of electromagnetic perturbations. The noise is V0.2 mV, there is a di- urnal modulation of amplitude V1 mV, and AE bursts are observed with a varying observed amplitude ranging between V10 mV and V2 V. One example is shown in Fig. 3. The present preliminary inference shows ¢rst a burst at 160 kHz, that seems to be later followed by a burst at 25 kHz. Summarising, every such system can be considered as a part of a true natural laboratory, in the fact that we know that nature shows us some phenomena that appear likely to originate by some speci¢c mechanism. Moreover, we can repeat the same experiment in di¡erent environmental settings, much like in the case we operate inside a standard laboratory in order to assess the relevant natural laws governing the phenomenon. It ought to be emphasised that operating the same experiment in di¡erent natural settings, is much di¡erent compared to carrying out a multidisciplinary investigation, always on the same system. Another item related to Stromboli deals with the concept of active experiment. When dealing with a standard laboratory experiment, the researcher usually changes the boundary conditions of every experiment in order to assess the laws VOLGEO 2385 6-5-02 438 M. Chiappini et al. / Journal of Volcanology and Geothermal Research 113 (2002) 429^442 Fig. 3. Recordings in the area of Vesuvius. Several AE at 25 and 160 kHz are shown. Every row datum is the sum of all peaks of all bursts occurring during a total time lag of V5 ms. that govern it. Research in the environment rather implies, in general, only passive observation with no possibility of modifying the system. Such a property, however, has some exceptions, as reported in the following. Some planetary-scale experiments of active modi¢cation of the Earth’s ionosphere and/or magnetosphere were concerned with high-altitude nuclear tests (McNoe, 1996). The weekend e¡ect associated with the operation of the BART train in the San Francisco area (Fraser-Smith and Coates, 1978; Fraser-Smith, 1981) is another example, as well as the electric ¢eld measurements by Haerendel’s classical Ba releases (see e.g. Kamide, 1988; Jacobs, 1991; Brekke, 1997). A few years ago, an entire set of active experiments dealing with plasma in space was concerned with the use of the Tethered Satellite System (TSS) operated with a conducting tether. E¡ects associated with the anthropic action can produce an active modi¢cation of the environment. Typical examples are e.g. the eutrophisation of a lake or of a river, with consequences on the greenhouse e¡ect, the increasing injection into the environment of warm water (Ellsaesser et al., 1986), or even the impact of CFC on the ozone layer or of atmospheric CO2 on global warming. In this respect, Stromboli appears particularly stimulating, in the fact that it is possible to start the stopwatch at the time of every explosion, and to monitor all that happens either before, or during, or after it. Every explosion is thus one active modi¢cation, by which a given measured amount of matter and of energy is injected into the environment. We can monitor, investigate, study its behaviour, di¡usion, decay, recovery, etc. under ever-changing conditions, with the possibility of exploiting large statistics. Considering the great amount of uncertainty dealing with the fundamental laws, processes, and mechanisms that control and determine climate, and the great present environmental hazard, threaten, and concern, such a possibility de¢nitely appears to be of paramount relevance. 6. The Stromboli Laboratory : a list of observations A permanent laboratory has recently been es- VOLGEO 2385 6-5-02 M. Chiappini et al. / Journal of Volcanology and Geothermal Research 113 (2002) 429^442 tablished on Stromboli by INGV, and an extensive interdisciplinary programme is being planned. Let us recall the possibility of carrying out some analogous measurement also on the occasion of an eventual eruption of Etna, which could occur sometimes while operating the Stromboli experiment. This could allow for some interesting comparison, when considering the possible di¡erent prime energy supply of the two volcanoes. Concerning Stromboli, several classes of problems are envisaged: (1) matter exchanges (chemical and isotopic aspects, including chemical geodynamics); (2) thermal and/or radiative coupling; (3) electromagnetic coupling, including magnetic and electric properties of the volcano; (4) deformation ; (5) biospheric interactions; and (6) anthropic relations (i.e. archaeology since Palaeolithic times). The last item appears interesting considering the particular geographical location of the entire Aeolian archipelago for a strategy for the investigation of the former occupation of the Mediterranean area either at the time of the Neolithic Revolution or such a process was, perhaps, also correlated with the volcanic activity vs. time of the di¡erent Aeolian Islands. Also in this respect, the Stromboli Laboratory appears to be a stimulating opportunity for promoting an operative interaction between specialists of exact sciences and of humanistic disciplines, with the purpose of studying the anthropology of the relations between man and environment (Gregori and Gregori, 1998). A few aspects appear worthy of special mention, as such experiments can be carried out only at Stromboli. Measuring the average speed, the total mass, and the total electric charge of the ejecta of every explosion; studying the lightning phenomena in volcanic plumes ; searching for possible lesser transient disturbances in the ionosphere associated with every explosion are all activities which can be carried out in the laboratory. Other aspects are self-explanatory in the following list. Their interpretation has to be carried out by considering that the prime agents are: (1) endogenous thermal heat; (2) tidal deformation of soil; (3) meteorological factors ; and (4) biological control. 439 The monitored phenomena being presently envisaged are: b The chemical and isotopic composition, and relative temporal evolution, of the Aeolian volcanism during geologic past b Mineralogical, geochemical, and isotopic data (soil, water, air) b Aerogeophysics (high resolution, multilevel repeated measurements) b Sea £oor monitoring (GEOSTAR) (Beranzoli et al., 1998) b Gravimetry b Tilt, stress, strain (at surface, and within a V50-m bore-hole) b Temporal variation of bathymetry (high sensitivity) b Soil deformation b Seismicity and tomography b Microseismicity and acoustic emission b A ring of buoys for monitoring acoustic emissions and temperature b Geothermics (shallow and deep) b Thermal pro¢le within shallow soil ( 6 3.5 m depth), i.e. ‘geotherms’ b Chemistry and gases associated with water wells b Fumarolic activity (both on land and under sea) b Gas exhaled from soil b Total balance of gas exhalation and exchanges (thermal, radiative, chemical, solid, liquid) between soil, sea, and atmosphere, including submarine fumaroles b Tracking the underground bubble formation and its propagation, prior to its explosion into the atmosphere b Local meteorological and environmental impact (including standard meteorological recordings, radiative balance, tethered balloon, and numerical micro-meteorological modelling) b Boundary layer meteorology either with or without volcanic injection b Gas entering from the atmosphere into soil b Atmospheric electrical conductivity b Atmospheric electrical potential gradient b Atmospheric microvorticity very close to ground surface (e.g. by monitoring the twinkling of light and the atmospheric optical depth) VOLGEO 2385 6-5-02 440 M. Chiappini et al. / Journal of Volcanology and Geothermal Research 113 (2002) 429^442 b The total amount, the average speed, and the chemistry, of the ejecta of every explosion and of its exhaled gases and aerosols, by means of either SODAR, or LIDAR, or aureolametry b Non-parametric characterisation of biota both underground and on the soil surface, and of their vegetative stage b Water content within soil vs. both depth and time b Electrical conductivity c of soil vs. both depth and time b Time variation of soil porosity by monitoring microdeformation in terms of: (1) acoustic emission; and (2) high-sensitivity tiltmetry b Spontaneous potentials and geoelectrics b Geomagnetic ¢eld (variation, and absolute), and telluric currents b Geo^electromagnetic tomography b Lightning discharges (strikes and strokes, both within the plume and independent of it) b Ionospheric transients (recorded from the Gibilmanna ionospheric observatory, permanently operated by INGV on the northern coast of Sicily). 7. Conclusion The science of the environment cannot be reconciled with the standard separation into disciplines and their further splitting into specialities. There is a strict need for substantial and wide teamwork, a perspective that is di⁄cult considering the instinctive tendency of scientists to isolate themselves within their own cultural concern. Moreover, it appears very di⁄cult to envisage concrete experimental con¢gurations where specialists from di¡erent branches of knowledge do e¡ectively and practically co-operate with a common target in mind. Stromboli, with its incredible millennial regularity, with its isolation with respect to all other emerged lands, with the comparatively simple feasibility of an integrated multidisciplinary monitoring project, o¡ers some opportunities that could hardly be met in every other site all over the world. With this perspective in mind, the Stromboli Laboratory installed and operated by ING is likely to be a great opportu- nity for the international scienti¢c community for exploiting some unprecedented and substantial progress in the understanding of the mechanisms that challenge and threaten our environment. Acknowledgements It is not possible to do justice to all the inputs by friends, colleagues, and authoritative scientists over the last few years of talks, meetings, workshops, private discussions, etc. We want to stress our sincere appreciation and gratitude, in geographical order, to Prof. M. Riuscetti and co-workers from the University of Udine; to Prof. A. Vettore from the University of Padova ; to Prof. J. Chahoud from the University of Bologna; to Dr. G. Martinelli of the Servizio Geografico e Geologico della Regione Emilia-Romagna (Bologna); to Prof. S. Lombardi and co-workers from the University La Sapienza (Rome), to Dr. V. Iafolla from IFSI (CNR, Rome); and to several colleagues of INGV, and of CNR (IDAC, IFA, IAS, IFSI, IRTR) in Rome and IMAAA in Potenza. We are indebted with Prof. Paolo Diodati from the University of Perugia and Dr. Paolo Palangio from INGV for providing Fig. 2, and with Prof. E. Mantovani from the University of Siena for the references on the geodynamics of the Mediterranean area. Concerning the archaeological items, we want to thank Arch. G. Campo and Arch. A. Rotella of the Soprintendenza ai Beni Culturali ed Ambientali (Messina), and Dr. Neda Parmegiani and Dr. Lucia Vagnetti from the Istituto di Studi Micenei ed Egeo-Anatolici (CNR, Rome). Finally, we are glad to acknowledge the several invited and contributed talks and the discussions during the two national Italian workshops devoted to Stromboli, held in Rome on 9 March, 1998, and on 29^ 30 October, 1998, respectively. The international workshop’s observational data base and mechanisms of climate held at Erice (Trapani) on November 21^27, 1998 and the Bridge between the Big Bang and Biology (Stars, Planetary systems, Atmospheres, Volcanoes : Their Link to Life), held at Stromboli on September 13^17, 1999, provided important inputs for the present paper. VOLGEO 2385 6-5-02 M. Chiappini et al. / Journal of Volcanology and Geothermal Research 113 (2002) 429^442 References Arna'iz Villena, A., Garc|'a, J.A., 1998. El Origen de los Vascos y Otros Pueblos Mediterra'neos. Estudios Complutenses, Editorial Complutense, Madrid, 139 pp. Baubron, J.C., Allard, P., Toutain, J.P., 1990. Di¡use volcanic emissions of carbon dioxide from Volcano Island, Italy. Nature 344, 51^53. Baubron, J.C., Allard, P., Toutain, J.P., 1991. Gas hazard on Vulcano Island. Nature 350, 26^27. Beranzoli, L., Favali, P., Etiope, G., Frugoni, F., Smriglio, G., Gasparoni, F., Marigo, A., 1998. GEOSTAR geophysical and oceanographic station for abyssal research. Phys. Earth Planet. Inter. 108, 175^183. Brekke, A., 1997. Physics of the Upper Atmosphere. John Wiley and Sons and Praxis, Chichester, 491 pp. Chiappini, M., Meloni, A., Boschi, E., Faggioni, O., Beverini, N., Carmisciano, C., Marson, I., 2000. Shaded relief magnetic anomaly map of Italy and surrounding marine areas. Ann. Geo¢s., 43, 983^989. Civetta, L., del Gaudio, C., de Vita, S., di Vito, M., Orsi, G., Petrazzuoli, S., Ricciardi, G., Ricco, C., 1995. Volcanism and resurgence in the densely populated Campi Flegrei nested caldera. Period. Mineral. 64, 135^136. Constable, S.C., 1990. Marine electromagnetic induction studies. Surv. Geophys. 11, 303^327. Crisci, G.M., de Rosa, R., EsperancPa, S., Mazzuoli, R., Sonnino, M., 1991. Temporal evolution of a three component system: the island of Lipari (Aeolian arc, southern Italy). Bull. Volcanol. 53, 207^221. de Natale, G., 1995. Mechanisms of unrest episodes al Campi Flegrei Caldera. Period. Mineral. 64, 163^163. Diodati, P., 1994. Ultrasonic signals before, during and after seismic activity revealed on Stromboli. Acoust. Lett. 17, 233^237. Ellsaesser, H.W., MacCracken, M.C., Walton, J.J., Grotch, S.L., 1986. Global climatic trends as revealed by the recorded data. Rev. Geophys. Space Phys. 24, 745^792. EsperancPa, S., Crisci, G.M., de Rosa, R., Mazzuoli, R., 1992. The role of the crust in the magmatic evolution of the island of Lipari (Aeolian islands Italy). Contrib. Mineral. Petrol. 112, 450^462. EsperancPa, S., Crisci, G.M., 1995. The Island of Pantelleria: a case for the development of DMM-HIMU isotopic composition in a long-lived extensional setting. Earth Planet. Sci. Lett. 136, 167^182. Fraser-Smith, A.C., 1981. E¡ects of man on geomagnetic activity and pulsations. Adv. Space Res. 1, 455^466. Fraser-Smith, A.C., Coates, D.B., 1978. Large-amplitude ULF electromagnetic ¢elds from BART. Radio Sci. 13, 661^668. Frazzetta, G., La Volpe, L., 1991. Volcanic history and maximum expected eruption at ‘La Fossa di Vulcano’ (Aeolian Islands, Italy). Acta Vulcanol. 1, 107^113. Giberti, G., Jaupart, C., Sartoris, G., 1992. Bull. Volcanol. 54, 535^541. Gillot, P.-Y., Keller, J., 1993. Radiochronological dating of Stromboli. Acta Vulcanol. 3, 69^77. 441 Gioncada, A., Gurioli, L., Sbrana, A., 1995. The hydrothermal magmatic eruption of ‘Breccia di Commenda’. La Fossa di Vulcano, Eolian Islands, Italy. Period. Mineral. 64, 191^192. Gregori, G.P., 1998. The magnetosphere of the Earth. A theory of magnetospheric substorms and of geomagnetic storms. In: Schro«der, W. (Ed.), From Newton to Einstein. Festschrift in honour of the 70th birthday of Hans-Juergen Treder. Science Edition, Bremen, pp. 68^106. Gregori, G.P., 1999. The external magnetic sources over the polar caps. Feasible modelling vs. unrealistic expectations. Ann. Geo¢s. 42, 171^189. Gregori, G.P., Lanzerotti, L.J., 1982. Electrical conductivity structure in the lower crust. Geophys. Surv. 4, 467^499. Gregori, G.P., Dong, W., 1996. The prime mover of volcanoes. History of a concept (in the ‘western’ science). In: Schro«der, W., Colacino, M. (Eds.), Global Change and History of Geophysics. IDCH of IAGA, Bremen, pp. 12^75. Gregori, G.P., Gregori, L.G., 1998. Solar^terrestrial relations. A reminder. Acta Geod. Geophys. Hung. 33, 391^459. Gregori, G.P., Lanzerotti, L.J., Alessandrini, B., de Franceschi, G., Cipollone, R., 1987. The planetary scale distribution of telluric currents and the e¡ect of the equatorial electrojet: an investigation by Canonical GDS. Pure Appl. Geophys. 125, 369^392. Gregori, L.G., 1997. The myth of Hephaestus and the activity of the island of Vulcano. In: Schro«der, W., Geomagnetism and Aeronomy with Special Case Studies. Science Edition, Bremen, pp. 295^314. Harwell, M.A., Hutchinson, T.C., 1988. Environmental Consequences of Nuclear War. 2nd ed., Vol. II, Ecological and Agricultural E¡ects. SCOPE 28, ICSU, and John Wiley and Sons, Chichester, 523 pp. Italiano, F., Nuccio, M., Pecoraino, G., 1998. Steam output from fumaroles of an active volcano: tectonic and magmatic hydrothermal controls on the degassing system at Vulcano (Aeolian arc). J. Geophys. Res. 103, 29829^29841. Jacobs, J.A. (Ed.), 1991. Geomagnetism, Vol. 4. Academic Press, Harcourt Brace Jovanovich, London, 806 pp. Kamide, Y., 1988. Electrodynamic Processes in the Earth’s Ionosphere and Magnetosphere. Kyoto Sangyo University Press, Kyoto, 756 pp. Kazimirovsky, E.S., 2002. Coupling between troposphere, stratosphere, thermosphere and lower ionosphere. In: Gregori, Piervitali, in press. Lanzerotti, L.J., Gregori, G.P., 1986. Telluric Currents: The Natural Environment and Interactions with Man-Made Systems. In: Kride, E.P., Roble, R.G. (Eds.), The Earth’s Electrical Environment. National Academic Press, Washington D.C., pp. 232^257. Mantovani, E. (Ed.), 1997. Geodynamics of the Mediterranean region and its implications for seismic and volcanic risk. Ann. Geo¢s. 40, 569^781. Mantovani, E., Albarello, D., 1997. Middle term precursors of strong earthquakes in southern Italy. Phys. Earth Planet. Inter. 101, 49^60. Mantovani, E., Albarello, D., Tamburelli, C., Babbucci, D., 1997. Recognizing the Italian zones most prone to next large VOLGEO 2385 6-5-02 442 M. Chiappini et al. / Journal of Volcanology and Geothermal Research 113 (2002) 429^442 earthquakes: possible approaches and present chances. Ann. Geo¢s. 40, 1329^1344. Mantovani, E., Albarello, D., Tamburelli, C., Babbucci, D., Viti, M., 1997. Plate convergence, crustal delamination, extrusion tectonics and minimization of shortening work as main controlling factors of the recent Mediterranean deformation pattern. Ann. Geo¢s. 40, 611^643. Mantovani, E., Albarello, D., Babbucci, D., Tamburelli, C., 1997. Recent/present tectonic processes in the Italian region and their relation with seismic and volcanic activity. Ann. Tecton. 11, 27^57. McNoe, P.A., 1996. Auroral and magnetic e¡ects of high altitude nuclear explosions in the magnetically conjugate region near western Samoa. In: Schro«der, W., Colacino, M. (Eds.), Global Change and History of Geophysics, IDCH of IAGA, Bremen, pp. 161^172. Meloni, A., Lanzerotti, L.J., Gregori, G.P., 1983. Induction of currents in long submarine cables by natural phenomena. Rev. Geophys. Space Phys. 21, 795^803. Meloni, A., Gregori, G.P., Lanzerotti, L.J., Medford, L.V., 1984. Search for a possible electromagnetic coupling between a transatlantic communication cable and the magma chamber in the mid-Atlantic ridge. J. Geophys. 55, 185^ 190. Otterman, J., Novak, M.D., Starr, D.O’C., 1993. Turbulent heat transfer from a sparsely vegetated surface: two-component representation. Bound. Layer Meteorol. 64, 409^ 420. Pinarelli, L., del Moro, A., Gioncada, A., Sbrana, A., 1995. Sr, Nd, and Pb isotope characterization of Vulcano Island (Aeolian Arc Italy). Period. Mineral. 64, 249^250. Pittock, A.B., Ackermann, T.P., Crutzen, P.J., MacCraken, M.C., Shapiro, C.S., Turco, R.P. 1988. Environmental Consequences of Nuclear War. 2nd ed., Vol. I, Physical and Atmospheric E¡ects. SCOPE 28, ICSU, and John Wiley and Sons, Chichester, 359 pp. Rampino, M.R., Self, S., Stothers, R.B., 1988. Volcanic winters. Annu. Rev. Earth Planet. Sci. 16, 73^99. Rosi, M., Sbrana, A. (Eds.), 1987. Phlegrean Fields. Quad. Ric. Sci., (114), (Progetto Finalizzato ‘Geodinamica’, Monogra¢e ¢nali Vol. 9), CNR, Rome, 168 pp. Simkin, T., Siebert, L., 1994. Volcanoes of the World, 2nd edn. Geoscience Press, Tucson, AZ, 349 pp. Viti, M., Albarello, D., Mantovani, E., 1997. Rheological pro¢les in the central^eastern Mediterranean. Ann. Geo¢s. 40, 849^864. Wilson, M., 1989. Igneous Petrogenesis. A Global Tectonic Approach. Unwin Hyman, London, 466 pp. VOLGEO 2385 6-5-02