Perspectives concerning Satellite EO and geohazard risk management: the way forward.
Draft community paper concerning volcanic hazards.
Coordinators of the International Forum:
Philippe Bally (ESA)
Francesco Gaetani (GEO Secretariat)
Lead Authors:
Fabrizio Ferrucci (IPGP)
Fred Prata (NILU)
Contributing Authors:
Gerald Bawden (USGS)
Pierre Briole (IPGP)
Ciro Del Negro (INGV)
Giuseppe Puglisi (INGV)
Fabio Rocca (POLIMI)
Steve Tait (IPGP)
Marialucia Tampellini (Carlo Gavazzi Space S.p.A.)
Nicolas Theys (Institut d'Aéronomie Spatiale de Belgique)
David Schneider (USGS)
David Norbury (EFG)
Acknowledgements:
This paper received editing input from Philippe Bally (ESA), Geraint Cooksley (Altamira), Andrew Eddy (Athena Global) and Marie-Josée Banwell (Altamira).
Contents
1Scope of this community paper
2Volcanic Hazards and Volcanic Risk exposure
3Users and their information needs
4The European case
5The global perspective
6Current state of satellite EO services & applications
6.1) Main EO capacities used or in development
6.2) Emerging Research
7The way forward
7.1) Technology & services
7.2) Science
7.3) Users and practitioners
8References
Page 1/38
Date 18/04/2012
This document is a draft of a series of papers with the scope of reflecting the community perspective on geohazards. It forms the basis for discussion to be held at the International Forum on Satellite Earth Observation for Geohazard Risk Management organised by the European Space Agency on May 21-23, 2012 (). The authors invite comments and further contributions from the community on this paper so they can be collated and discussed at the conference. The document will be updated on the basis of this discussion. Comments should be addressed to with the title of the draft paper.
1Scope of this community paper
This paper presents the perspectives on the contribution of satellite Earth Observation (EO) to volcanic and volcano related hazards, and the associated risk and disaster impact mitigation. Taking intoaccount the current state and expected evolution of applications and services, their realistic level of usage, and the achievable needs expressed by a compact set of qualified end users, the paper attempts tobroaden the view to provide a global perspective.
This paper outlines a 5 to 10-year vision for the volcanic hazard community, based on the assessment of state of the art research and composed of a set of possible outcomes including analysis of how to strengthen and consolidate the applications and to focus, orient and improve competitiveness of volcanic risk related EO services. This document builds on the 3rd International Geohazards workshop of the Group of Earth Observation (GEO), held in November 2007 in Frascati, Italy, which recommended “to stimulate an international effort to monitor and study selectedreference sites by establishing open access to relevant datasets according to GEO principles to foster thecollaboration between all various partners and end-users”.
Community papers are being elaborated for other types of geohazard for discussion at the International Forum on Satellite Earth Observation for Geohazard Risk Management (May 21-23, 2012) in Santorini, Greece. Other papers are being prepared on landslides, seismic hazards, coastal lowland subsidence and flood risk and inactive mines. The scope and theme of these community papers are described in Appendix 1.
2Volcanic Hazards and Volcanic Risk exposure
About 1,500 volcanoes are known to have erupted in the last 12,000 years, the Holocene geological Era; about 700 of these, mostly subaerial, have erupted at least once in historical times (Siebert et al., 2010).
Figure 1. Holocene active volcanoes (Global Volcanism Program of the Smithsonian Institution,
Worldwide, an average of about 100 volcanic unrests are observed yearly, and about a half of them turn into observable eruptions. As oceans cover two-thirds of Earth, the number of submarine active volcanoes – mostly associated with ocean ridges – is larger than subaerial ones but the precise number remains unknown. Figure 1 shows the spatial distribution of Holocene volcanic structures, emphasizing the association of almost all active volcanoes with the continental plate boundaries and the hot spots.
The most relevant spatial clusters of active volcanism are those associated with Far-East Asia between latitudes 60°N and 10°S, and to the western Americas, between 55° South and 60°North approximately. The overall framework is completed by the volcanic provinces associated with oceanic hot spots or emerged oceanic rifts, and by the areas of volcanism declining with weakening continental dynamics – over geological timescales – such as the central-eastern Mediterranean basin.
As far as the conversion of hazard to risk is concerned, this depends on the location of assets at risk, and their dependence on time. This leads to two risk terms, one spatially stationary - related to geographically permanent exposures, such as cities and mega-cities at the foot of active volcanoes (in Italy, Japan, Iceland, the whole of western central and southern Americas, northern-western USA and Alaska, Kamchatka, Indonesia, Philippines, Hawaii, Lesser Antillas, Azores, Canarias, Congo, etc.) - and one transboundary related to the emissions of dense plumes of volcanic ash and gases.
Eruptive styles generally correlate with viscosity and temperature of magmas. In essence, high-viscosity magmas display high-Silica content (typically >60%) and relatively low-temperature (typically <1000°C). They are associated with explosive volcanism as highly viscous lavas tend to retain high-temperature volcanic gases, such as water vapour, Carbon dioxide and Sulphur dioxide, which form vesicles within the entrapping matrix. The driving force for explosive eruptions, the products of which are ash and SO2, is provided by dissolved gas in viscous magmas: these clouds can disperse in the troposphere and stratosphere travelling very large distances from the eruption source.
Conversely, magmas with relatively low silica content (in the order of 50%), low-viscosity and temperatures typically over 1000°C, give rise to effusive eruptions with limited or no explosive activity as gases migrate through and escape them, without major build-up of internal pressures.
Lava flows, the non-turbulent fluid product of effusive eruptions, may travel long distances on land at velocities modulated by instantaneous effusion rates, terrain slopes and viscosity. Velocities are typically less than, or much less than 1 km per hour, and flow lengths are in the order of a few kilometres: noteworthy cases include flow lengths of a few tens of kilometres (Nabro 2011, e.g.) - and an exceptional velocity of about 10 km/h observed once at Mauna Loa (Hawaii) in 1950.
In a few cases worldwide, persistent molten lava lakes may form if a dynamic equilibrium is reached between the magma load, its volatile content and the pressure in the underlying shallow plumbing system. Among the more or less long-lived lava lakes worldwide, that at Nyiragongo is peculiar as it is less than 20 km away from, and ~2,000 m above the crowded city of Goma (Congo), which underwent a major volcano emergency in parallel withan extreme humanitarian crisis in 2002.
Still on land, pyroclastic flows – a comprehensive term that includes “nuées ardentes”, pyroclastic “dry” surges and water/steam rich “wet” (colder) surges – are fast, horizontally and vertically moving streams of fragmented rocks and superheated gases.
Originating from the gravitational collapse of ‘Plinian’ columns (e.g., as at Pinatubo, 1991 or in the famed Pompeii disaster of 79 A.D.), or from the collapse of spines of very viscous lavas at ‘dome forming’ volcanoes as soon as they have grown enough to become unstable (e.g. Mt. St-Helens 1980, Montserrat 1995-2010 or Puyehue-Cordón Caulle, 2011), they present observed speeds in the order of hundreds km/h and top temperatures even in excess of 900°C for dry surges and glowing clouds. In consideration of the huge quantity of motion, these are the most destructive features associated with volcanic eruptions in general.
In highly explosive eruptions, the turbulent jet composed of rock fragments and super-heated gas, heats the troposphere and rises dramatically fast and high by convective thrust. The ceiling of such explosive eruptive columns depends on the difference in temperature between the jet and the surrounding atmosphere, and the fourth root of the actual mass eruption rate. The observed altitudes where explosive eruptions ‘umbrellas’ can expand because they reach neutral buoyancy, are below 50 km, which is already deep into the stratosphere. Volcanic aerosols at these altitudes can be transported and spread to large distances without precipitation or chemical transformation.
For ‘ordinary’ mass eruption rates below 8-10 m3 /s, ash can be injected in the upper troposphere and propagate even at distances of several hundred kilometres before being diluted to non-dangerous concentration levels, thus transforming the volcanic hazard into a global threat to air navigation. In Europe, Mt. Etna has been the main volcano intermittently causing temporary problems for aviation. However, on April 14, 2010, the moderate eruption of the Icelandic volcano Eyjafjallajökull which began one month earlier suddenly turned into phreato-magmatic explosive activity. What ensued became the largest ever economic impact from a volcano-related event. From 14 to 20 April, full closure of North and Central Europe airspaces led to the cancellation of ~100,000 flights, with some 10 million passengers stranded (about half of the world's air traffic). Oxford Economics (2010) estimated a total global economic impact of ~5bn€ and IATA [1] stated that the total loss for airline industry could be close to 1.5bn€, whereas another 0.2bn€ was the cost claimed by AOA [2] as the major hubs of London, Amsterdam, Paris and Frankfurt were virtually shut-off by the effects of the ash clouds. Additional costs not included in these figures, are those borne by airlines for travellers being unable to return home and those, much larger but still unconstrained, caused by interruption of supplies of all types.
The response of a jet engine when exposed to volcanic ash depends on a number of variables, including the ambient ash concentration and composition (which influences the melting point), time of exposure, and obviously engine type and thrust settings. It is understood, therefore, that both flying briefly across a high ash concentration, and flying a long time along an unnoticeable low-concentration ash plume, may result into severe engine damage up to failure of engines or severe sandblasting of propellers.
Three final elements worth being addressed when completing the volcanic hazard assessment are the expected length of flows, the duration of unrests, and the impact of both on territorial management.
Forecasting the extent of flows vs. time is of paramount importance to prepare the operational response, and to constrain time, location, type and extent of prevention works to be fielded by the civil protection services, or their equivalent. Among the many parameters that have been proposed as the controlling factors of the length of lava flows (terrain’s slope, viscosity, yield strength, total volume, mass effusion rate) it was concluded that the effusion rate is the leading one.
In the case of explosive eruptions, where ultrafast development of climaxes does not allow a timely response on land, all prevention measures –mostly involving the retreat of exposed persons and movable activities to safer and distant places– should be taken well in advance, before the crisis’ onset or when the unrest is building-up. In general, however, the ability to take enlightened decisions is affected by the lack of knowledge concerning two crucial questions: when the first breakout will happen, and when the eruption –or the eruptive cycle– will come to an end. Recent major crises worldwide (among others: Campi Flegrei 1982-84, Pinatubo 1991, Etna 1991-93, Montserrat 1995-2010, Eyjafjallajökull 2010, among others) illustrate that we can usually only answer the first question, with variable uncertainty.
3Users and their information needs
Conceptually, the monitoring of volcano dynamics is dealt with by volcano observatories which run monitoring arrays of instruments, and carry out multi-parameter networked measurement for constraining elastic, mass, geometric, magnetic, chemical and gas parameters, in time and space. As volcano assessment and forecasting are still supervised, a dominant part of monitoring relies on visual observation and terrain inspection.
To date, however, less than 10% of active volcanoes worldwide are monitored by less than a hundred observatories cooperating in the World Organization of Volcano Observatories (WOVO). This indicates that about 90% of sources of volcanic hazards do not have a dedicated observatory and are not monitored, or only occasionally, in situ, typically using temporary mobile systems. The last decade has seen increased use of satellite remote sensing for monitoring volcanic unrest for those volcanoes not having a well-established, dedicated observatory.
By nature, a volcanic eruption is a locally relevant event that may turn into a transboundary event. Consequently, there are two categories of potential users of spaceborne information on volcanic activity (monitoring) and volcanic hazards in general (risk exposure assessment and mapping):
- the first category is national, and should be selected case by case from those responsible for disaster & risk management, or of giving scientific advice to those who makedecisions to protect lives and property. Typically, the former is a Ministry or a mandated National Agency, whereas the latter is a volcano observatory, a geological survey or their equivalent.
- the second category is transnational and, as such, has no ruling nor advising powers on the territory hosting the volcano. Typically represented by the VAACs, it is an intermediate link between WMO, ICAO and individual airlines, requiring timely warned by volcano observatories –where they do exist– on major ash and gas emissions.
A realistic review of objective user needs –as they emerged in major crises worldwide and in current European projects in volcano remote sensing– leads to discriminate volcanic ground features from atmospheric ones, and crisis management needs from strategic activities for hazard assessment and prevention.
a)Volcanic ground features
To date, the prevalent demand in EO from volcano observatories is centred on pre-/syn-eruptive EO, that is: monitoring and mapping of lava flows and lava discharge rates as a clue on the dynamics of the shallow plumbing system, tracking of ash dispersal; ground deformation as a clue on the location and severity of shallow magmatic intrusions, velocity deformation maps to gain insight into the evolution of magmatic system; and remote control –thermal and geometrical– of critically bulging domes.
Requested refresh rates of information vary as a function of parameters monitored. Volcano observatories request observation rates compatible with the proper monitoring of fast developing phenomena such as explosive events of all type (lava fountains, pyroclastic flows), where the need for high-rate measurements prevails on the spatial resolution of imaging, and complementary to the ground based systems.
Conversely, circa-daily assessments at medium-high spatial resolutions are an appropriate target for controlling slower, syn-eruptive phenomena such as the evolution of lava flows or the dynamics of ephemeral lava lakes.
An intermediate level of temporal and spatial resolution (days/weeks vs. tens/hundreds of meters) is needed for monitoring the growth of viscous lava domes, the dynamics of permanent, near-stationary lava lakes or the cooling of pyroclastic and lava flows.
Finally, volcano observatory requests in ground deformation spread on shorter temporal resolution (weekly) to imagine magmatic intrusions and longer timespans, to focus on the reconstruction of multi-year series of uplifts for assessment of slow, strain field evolution associated with long-term inflation/deflation cycles. During volcanic unrest periods shorter revisiting times will be mandatory.
b)Volcanic features in the atmosphere
The example of the hazard to jet aircraft from dispersing clouds of ash is very relevant to EO as volcanic ash is transported and dispersed by atmospheric winds and hence its geographic reach is enormous.
Prior to April 2010 the advice to aviation when encountering volcanic ash clouds was to avoid them. This advice works well in daylight and with favourable view angles to the sun, and for isolated ash clouds in regions of the world where air traffic is sparse; it becomes impractical where air traffic is dense, as in the case of continental Europe, and when ash clouds persist, as was the case in the Eyjafjallajökull eruption.
In response to the latter, a broad community of end users composed of aviation regulators, policy makers, engine manufacturers and representatives of commercial airlines, agreed upon three levels of ash concentration thresholds, ranging from «safe to fly» (below 200 µg m-3) to «special safety procedures» (between 200 and 2000 µg m-3): beyond 2000 µg m-3, the airspace becomes «no-fly zone».
As these ash concentration levels are «Forecast» and not «Observed», this puts a large burden on the achievement of an all-weather quantitative observation capacity, on one side, and the ability of atmospheric dispersion models to make accurate forecasts reliably fitting the actual concentration and location of ash clouds, on the other.
The current needs consist in the timely (refresh rates are also here in the order of minutes) provision of the following quantitative information: (i) detection, location and quantitative characterization of the active volcanic source on ground; (ii) detection, accurate 3-D location and concentration imaging of the volcanic ash cloud, and (iii) forecast of the cloud dynamics in concentration, space and time, since timeliness and complete temporal coverage (day and night) are needed.
The end user for these data are essentially the airline industry, but the data pass through several filters, including third party EO data product providers, advisory and warning centres (e.g. VAACs), official channels (e.g. Meteorological Watch Offices) and aviation stakeholders, such as Airport Authorities, airlines, air-freight companies, private and commercial business jet operators and Defence Agencies. The value chain is therefore quite long, and the increase in worth of the EO data considerable.
This overview of user needs can take advantage of experience gathered in over 10 years of the International Charter on Space and Major Disasters, an operational system unifying a virtual constellation of different EO missions that works on a reactive basis in the immediate disaster response phase i.e. where users’ requests usually come with little to no advance notice. Evidence proves that in such cases the effectiveness of EO response may remain questionable or unsatisfactory, unless it relies upon systems that are already in operation, or encompass areas wide enough to allow dealing with multiple crises at once, or address EO at refresh rates which can be readily re-programmed to fit the need.