Index | Diagnostic Characteristics | Geographic Occurrence | Investigation & Mitigation | Key Contacts & Expert Advice | Photo Gallery | Essential References & Further Reading | Definitions & Glossary |
Volcano Hazards Program
The objectives volcano hazards program are to design and implement a scientific investigation to advance the understanding of volcanic processes, to identify hazards associated with volcanoes and to investigate their consequences and potential risks. This may enable warning to be made for the emergency management and possible evacuation of people who live near potentially dangerous volcanoes. As such, volcano hazards programs may consist of the following activities:
Volcano hazards management
Volcano crisis response
Evacuation and volcano disaster management
Volcano hazards programs are often strategically developed by Government and managed and implemented by a national geological survey, with support form specialist experts based in academia and industry.
No volcanic catastrophes have occurred during the early phases of a volcanic eruption (Tazieff 1979). Therefore, if information can be provided on the types of volcanic hazards and their likely affects, a certain length of time may be available to take mitigative, precautionary, avoidance or evacuation measures. Such information may be obtained from a monitoring programme, which is usually designed, implemented and managed by a volcano observatory (and a national geological survey or a research organisation).
Volcano monitoring plays a crucial role in reducing the impact of volcanic hazards. This involves the investigation, recording and analysis of measured events that accompanies volcanic activity and takes place before, during and after eruptions have taken place. There is now available an array of instruments to monitor volcanoes all of which monitor a particular element of an active volcano.
The objectives of monitoring a volcano are to detect evidence of impending volcanic activity so that volcanic processes may be determined. The interpretation of this data and information may then be used for forecasting and predicting the likelihood of the volcanic hazards and their likely influence on people, people lives, land, villages, towns, and associated infrastructure.
One single technique is usually not sufficient to allow eruptions to be predicted. A suite of techniques are required and the interpretation of the acquired data must be undertaken in conjunction with an appreciation of the prehistorical eruptive record and the processes of the volcano.
A monitoring program is aimed to detecting changes to a volcano during the days, weeks, months and years before an eruption. For example, rising magma will trigger earthquake swarms and cause deformation of the volcanoes flanks and summit and the types of volume of volcanic gases emitted may change. Therefore, by monitoring these types of events it may be possible to give an indication of the likely timing of a volcanic hazard. Typical types of monitoring may include the following:
Geological field observations
Geological field observations
Geological field observations are a crucial part of volcano monitoring investigations. The importance of observations conducted by a trained professional geologist can not be over-emphasised and should be part of the monitoring programme. Observations allow geological and volcanic events and processes to become recorded and allow the interpretation of data provided by sophisticated instruments, which often are measuring only one parameter. Field observations enhance the value that sophisticated instrumentation can provide and presents the opportunity to integrate different types of data. Furthermore, field observations allow the instantaneous, ‘real-time’ integration and interpretation of data to provide information on the characteristics of the volcano.
In some cases, particularly in the early phases of a volcanic eruption field observation may be the only available means to assess the hazards associated with a volcano. Basic observations such as fissured developing on the flanks of a volcano, melting and fissuring ice caps, changes in lake levels, increases in fumarole activity, increased gas plume emissions, the sudden smell of hydrogen sulphide and sulphur dioxide, landslide generation and so on; all may provide the geologists with information upon which to base his decisions.
In some cases sophisticated equipment requires a period of time so that it can be installed, tested and validated. This often may require personnel and financial resources and assets. In general, sophisticated instruments are less adaptable, resourceful, inventive, flexible and ingenious than a trained geological observer. Although small changes may go unnoticed by a field geologist, large changes may go unnoticed by sophisticated monitoring instruments. Field observations may help to decide which instruments and parameters are likely to give the most reliable and useful information. The more sophisticated an instrument does no necessarily imply the most reliable parameters will be obtained. Often, simple observations may be sufficient. There can no substitute for an informed brain, a trained pair of eyes and a field geologist who is trained in making and recording observations. Keeping notes and taking photograph form are an important field task, as this information may have to be relied on, sometimes many years after the original observations were made.
Earthquake swarms frequently occur before the onset of a volcanic eruption. As magma rises into the magma chamber (or reservoir) this causes the deformation of solidified rock mass that surrounds the chamber. Networks of seismometers (an instrument which measures ground vibrations) are installed to monitor volcanoes to detect subtle, but significant variations in the type and intensity of seismic activity. In some cases however, there may be an absence of earthquake activity before an eruption, as was the case during the eruption of Heimaey, Iceland, in 1973. By contrast, the earthquake swarms which occurred on Guadeloupe in 1976 were not followed by an eruption.
A seismograph is an instrument that can detect, amplify and record ground movements which may be too small to be perceived by people. A seismometer is a mass mounted on a spring (a pendulum) and forms the internal part of a seismograph.
Ground motions that are detected by seismometers are converted into electronic signals which are transmitted and recorded on a seismogram (a graph showing the motion of the ground known also as seismic signatures), usually located in an observatory. The horizontal axis represents time and the vertical axis represents ground displacements (often in millimetres)
Seismograms may be subsequently analysed to provide information on the depth, location and time of the event (also known as a tremor). Mapping the location of each individual event allows the migration of the activity, and therefore magma movements, to be tracked.
Accelerographs are used as instruments to record very strong ground motions. The output is proportional to ground acceleration, compared to a seismograph, whereby the output is proportional to ground velocity. Accelerographs may be used for example in civil engineering foundation design.
The numbers of seismometers needs and there distribution across a volcano will vary, depending upon the characteristics of each individual volcano and available resources. The denser the seismometer network the greater the likelihood of monitoring the smaller (earthquakes less than magnitude 2) events and the delineation of fracture zones and magma reservoirs. Improvements in computer technology and data processing, over the past few decades have improved our ability to monitor seismicity associated with volcanic processes (McNutt 1996).
Seismic precursory events do not always enable the timing of an eruption to be forecasted, but, on occasions (for example at the Cascades Volcano Observatory), by determining the cumulative seismic strain energy release of several different types of seismic events, within a time frame, some eruptions have been predicted from a couple of hours to several days in advance.
There are several types of seismic events that can be recorded by seismometers, these include for example:
Regional tectonic earthquakes, characterised by sharp arrivals.
Deep earthquakes located in the region beyond a volcano. These produce high-frequency signatures and sharp arrivals.
Shallow earthquakes located within a few kilometres to the volcano. These produce low to medium frequency seismic arrivals.
Surface events, including; rock falls and landslides, collapse of lava domes, snow and rock avalanches, gas and tephra emissions. These signals may be complex with no clear boundaries.
A long-duration, rhythmic signal (known as harmonic tremor), the source of which is not fully understood but may be associated with rising magma or volcanically super heated fluids.
Real-Time Seismic-Amplitude Measurements (RSAM) & Seismic Spectral-Amplitude Measurements (SSAM)
RSAM and SSAM are techniques for characterising changing seismicity on a volcano as it is occurring. For example, the seismic events may be caused by magma movements, explosions, rockfalls, volcanogenic earthquakes and regional tectonic earthquakes. This can be determined from the amplitudes and frequencies of the seismic signals instead of the locations and magnitudes of the earthquakes. During an eruption unless a conventional seismographs is disturbed, therefore preventing the continual measurements of the seismicity, it is difficult to distinguish individual seismic events. RSAM and SSAM was developed by the United States Geological Survey to enable real-time quantitative to be made available. RSAM calculates and stores the average amplitude of ground shaking caused the seismic events in 10 minute intervals. RSAM increases as the tremor amplitude or the rate of occurrence and size of the earthquakes increases. Instead of providing information on a single event, details are provided for all events in a 10 minute period. SSAM computes the average amplitude of the seismic signals in a specific frequency band, which may be related to magma movements.
Gravity & microgravity measurements
Changes in the Earth’s gravitational attraction occur in response to change in elevation, since gravity will be stronger the closer a measurement is taken to the centre of the Earth. Gravitational attraction also changes depending on changes to the mass of rock and debris below the ground. For instance, gravity may be expected to change if magma intrudes voids in a volcanic edifice.
Gravity metres (known also a gravimeters) may be established on bench marks, or survey stations, situated throughout the summit and flanks of a volcano to monitor slight changes in the gravitational attraction of the ground, caused by, for example magma intrusion and ground elevation. The results from microgravity monitoring are often analysed with data that has been obtained from ground deformation surveys (such as GPS and precise levelling).
Ground movements & ground deformation
Ground movements on volcanoes include; subsidence, upwards shift, tilting, fissuring, compression and bulging. These changes in shape to the volcano are caused by the movement of magma before the onset of an eruption. Some of these ground movements may be sudden, dramatic and extensive or subtle and only noticeable with the use of sensitive ground deformation monitoring equipment. For example, the summit of Kilauea in Hawaii has been recorded to increase in height by approximately 1 m in the months leading up to an eruption. The flanks of Mt. St. Helens, USA expanded by up to about 2 m per day up to a height of 100 m. However, such ground movements do not always occur before an eruption. Mount Asma in Japan underwent approximately 117 mm of uplift around the summit and 17 mm of subsidence on the middle and lower flanks of the volcano during the period 1934-1950 (Decker and Kinoshita 1971). Some of the main instruments used to monitor ground movement are as follows:
Basic, inexpensive ground monitoring methods include for example; (a) the use of tape measures to monitor the expansion of a fissure, (b) measuring the angle of a slope using a clinometer and (c) using spray paint to measure aseismic creep, (d) non-quantitative observations and recording of ground movements. Most ground deformation programs require more detailed and sophisticated instrumentation over periods of time that range from weeks to decades.
Global Positioning System (GPS)
Global Positioning Systems (GPS) make use of data transmitted from satellites which orbit the Earth to precisely locate fixed points on the surface of the Earth. Repeated measurements enable the horizontal and vertical positions of the survey points to be determined. Relative position is a particular techniques whereby one of the satellite receiver instruments is situated at a stable control site (bench mark) and a second is established above another point to measure any relative change in position.
Measurements are taken from several orbiting satellites so that the receiving stations and computer processing can establish the relative position of the survey station on the surface of the Earth. The precision of GPS surveys is variable but is usually within approximately 100 mm. Repeated measurements provides information on the 3-dimesnion movement of the ground on, and around the volcano.
GPS measurements may be made where conventional ground based surveying is not possible due to, for example inaccessible terrain, no or poor lines of sight, dense vegetation, unfavourable weather conditions remote and rugged terrain. GPS measurements also provide information on both vertical and horizontal components of ground movements whereas conventional ground based surveying methods requires a combination of different techniques to provide information on the 3-dimensional movement of the ground. GPS equipment and receivers may be operated by one person, are portable and can operate for several months on solar panels or batteries.
Geodetic surveying networks (Electronic Distance Measurements)
Changes to the shape and volume of a volcano may occur in response to the build up of internal stresses caused by the movement of magma. Geodetic surveying networks may be established to record and monitor these ground movements. The most common types of techniques include a laser source electronic distance measurement instrument (EDM), tilt measurements (using an EDM and by conducting triangulation surveys), precise levelling (to obtain elevation changes).
EDM measurements between a network of bench marks that have been strategically placed across a volcano, tens to hundreds of metres apart, provides information on the 3-dimensional ground movements. EDM techniques use a laser beam to measure very small horizontal distance changes by making a series of precise levelling measures form stable bench marks (known also as survey stations). Alternatively, glass prisms (known sometimes as ‘targets’ or ‘reflectors’) may be secured (usually by grouting) onto parts of a volcano and monitored on a regular basis. The accuracy of EDM instruments may be +/- c1 to 5mmkm-1.
Precise levelling and total station theodolites
Geodetic levelling is a technique used to measure elevation (vertical) differences between stable bench marks by conducting repeated surveys at ‘regular’ time intervals. The interval between each successive survey will vary depending on several factors such as available resources and expected ground movements. Typical surveys may take place weekly, twice weekly, monthly, every six months, or in some cases on yearly basis. The levelling instrument (such as precise level) and a pair of graduated surveying (or levelling) rods are used to measure the different elevations between two fixed points (ie. vertical components of deformation). Each rod, measuring about 3m high, is placed on a stable bench mark. The instrument is placed in the middle of the two survey stations and forward measurement is taken first (known as the foresight) following by the backwards measurement (known as the backsight). Bench marks may be spaced 20 to 50 m apart and levelling traverses vary from about 500 m to 5 km. Shorter traverses often are designed to measure ground tilt, whereas longer traverses measure relative vertical ground movements.
Theodolites measure the horizontal and vertical angles from fixed survey stations (known as trilateration). Repeated measurements combined with EDM and precise levelling enable the 3-dimesionsal ground deformation to be measured and monitored associated with the inflation and deflation of the volcanoes flanks and summit. Theodolites can measure angle at distances up to 4 km to an accuracy of c10mm. The results form these types of observations may provide precursory indications for eruptions, volcanogenic landslides or magmatic intrusions. Total station theodolites enable both EDM and angles to be measured from a single instrument. The data is stored within the instruments and requires subsequent down-loading, processing and analysis.
Tiltmetres are one of the oldest methods for monitoring ground deformation on volcanoes and measure subtle changes in the amount and direction of tilt of the ground. These may be installed on the flanks, or around the carter of a volcano and the data is telemeetred back to the observatory to provide ‘real time’ data.
Extensometers and displacement metres (fissure measurements)
An extensometer, telemetered electronic tape measurements (and displacement metres) consists of a wire and anchored securely into rock either side of a fissure. The extension of the fissure and extension of the wire provides information on the rate and amount of dilation. This may be installed across a fissure to provide automated information that can be transmitted by radio back to the observatory for analysis. Measurements of the change in width of fissures may also be made using a steel or plastic tape between stable bench marks.
Widening of fissures may provide an indication that magma was moving or rising leading to an eruption. These fissures may be radial and concentric, located around the rim of a carter, or linear running longitudinally along the flanks of a volcano. The fissures may vary in length from less than a metre to tens of metres, and from a few hundred millimetres to tens of metres long. Incandescent rock and hot gases may often be associated with these types of fissures.
Modelling and Digital Terrain Models (DTM)
Computer simulated models and digital terrain models of a volcano may be developed to enable lava, lahars and pyroclastic flow paths to be predicted. Such techniques were used during the 1991 to 1993 eruption of Mount Etna, in Sicily, to forecast the direction and of lava flows that threatened the town of Zafferena (Barberi et al. 1993). There was in general a good correlation between the observed and predicted flows. In other cases lava flow lengths have been predicted based on knowledge of those factors and mechanisms which control lava flows. This was undertaken for Unzen Volcano in Japan (Chen et al 1993).
Tidal gauges may used to monitor changes in the levels of lake water in carters (for example at Rabaul caldera in Papua).
Terrestrial Light Detection and Ranging (LIDAR)
Terrestrial laser scanner (LIDAR) combined with GPS enables the accurate location of fixed survey points, which, when surveyed on a repeated basis, allows a detailed 3D terrain model to be produced. Ground deformation studies using LIDAR have been deployed to detect ground movements associated with instability on the lava dome and crater walls of the Soufrière Hills Volcano, Montserrat, West Indies (Jones 2006).
Monitoring volcanic gases
Gas monitoring is aimed at detecting changes to the composition and volumes of gases that are emitted from a volcano. The release of gas dissolved in magma is one of the fundamental forces that influence volcanic eruptions. The gases themselves may be a primary hazard as well as potential early warning indicator of a possible eruption. Changes in fumarole gas compositions and the emission rates of SO2 are possibly linked to changes to the type of magma, magma supply rates and magma movements.
Gas sampling may take place directly, by inserting probes into fumaroles and active events, or indirectly by the use of remote instrumentation. Although in some cases the automated on-site gas monitoring is possible. Some gases, such as SO2 easily dissolve in water, and this may prevent its accurate measurements. CO2 however, is not influenced as much by the presence of water and therefore variations in the CO2 volumes of emitted may enable assessment of whether magma is undergoing degasification (Sutton et al. 1992, Doukas et al. 1995). Environmental and logistical factors such as wind speed and direction, accessibility, poor weather conditions and hazardous locations, all will have an influence on the ability to detect and monitor volcanic gases. Volcanic gases may be monitored by several methods, such as the following:
The concentrations of volcanic gases may be may be automatically measured in the air, soil or fumaroles and the data transmitted by radio or telephone to the volcano observatory. This provides continual and real time information on the variability of gas concentrations, ranging from minutes to months.
Soil-efflux measurements & gas geochemistry
Some volcanic gases, such as CO2 disseminate into the upper layers of soil or recently accumulate pyroclastic debris, on the summit and flanks of volcanoes. Soil-efflux measurements can measure the gas concentration in the soil. If sufficient number so measurements are taken this enables the contouring of the data to provide an indication of the locations of highest gas emissions.
Direct gas sampling and laboratory analysis
The sampling of gases from fumaroles enables the origin and compositions of gases to be determined, but this method can not provide information on rates of gas emissions. Solution filled bottles are conventionally used to collect the gas samples, where they are subsequently transported to a laboratory for analysis.
Measurement of volcanic plumes
Gas plumes are usually always present above active volcanoes, although not always visible. Measuring changes in the rates of emissions of gases such as CO2 and SO2 it may be possible calculate the daily volume of gases emitted, which is conventionally expressed as tonnes per day. Significant increases in gas concentrations may be observed as a pre-cursory indication of a pending eruption, although the exact timing of an eruption, based on this information alone, is difficult to determine. During large eruptions SO2 is injected into the stratosphere where the plume can be detected by an instrument on board a satellite. However, in the events leading up to an eruption such gas concentrations and volumes may be measured using airborne or ground-based techniques such as Fourier Transform Infra Red (FTIR) and COSPEC.
Correlation Spectrometer (COSPEC)
COSPEC, originally developed to monitor atmospheric pollution, may be used to measure the emission of SO2 in volcano emission plumes. The amount of solar ultraviolet light absorbed by SO2 in the plume is measured and compared to an international standard. COPSEC compares the spectra of light with the known spectra of sulphur dioxide to calculate the amount of SO2 in the plume.
Numerous traverses are made from the ground or with the instrumented mounted on a vehicle, helicopter or boat. Measurements are made beneath the path of the plume, at both right angles and parallel to the direction which the plume is travelling. Wind speed and direction are also measured using a ground based, hand-held anemometer. This information is significant as it is used during the calculations to determine the quantity of SO2 in the plume. Measurements taken in the air often use wind data from local airports. This process is repeated at regular intervals, usually several times each week. Both SO2 and CO2 are measured and their relative amounts are reported as fluxes in tonnes per day.
Hydrological and Hydrogeological monitoring
Groundwater characteristics such as composition, temperature, pH and rates of flow may change before the onset of a volcanic eruption. Hydrological monitoring provides information on the understanding of hydrological processes, such as the generation of lahars, stability of river channels, flooding and fluid dynamics that transports huge volumes of sediments.
Volcanic activity can deposit loose debris that can be eroded and transported into river valleys and destroy surface vegetation that can increase rates of surface runoff during rainstorms. Hydrogeological monitoring seeks to provide information on the following
Lahar detection systems may be installed to monitor the arrival and passage of debris flows, floods and lahars in valleys that drain the upper reaches of volcano slopes. This enables timely warning systems to be emplaced to help save lives of people who live downstream.
Sediment types and volumes may also be monitored as these potentially sterilise huge areas of agricultural land, mainly during of prolonged rainstorms. Stream gagging instruments may be used to measure the volume of sediment and water transported by rivers (Major et al. 2000). Hydrological hazards monitoring may involve; the determination of sediment erosion, transportation and deposition, erosion of gullies and channels, water and sediment discharge measurements, river channel surveys, lake monitoring, assessing the likelihood for dams to become breached, failure of debris dams, liquefaction caused by earthquake shaking.
Earth Observation (Remote Sensing)
Satellites imagery and advances in remote sensing techniques enable volcanoes to be monitored (Rothery 1992) to:
Satellite thermal monitoring may be undertaken using weather satellites that can give an indication that eruptions may be days or weeks away. However, the weather conditions may prevent, or hinder the ability of the satellite to continually monitor a volcano which may be hours or days away from an eruption.
Satellite radar interferometry utilises radar beams that detect the subsidence or inflation of a volcano by taking measurements months to years and comparing the data.
Satellite optical cameras are mounted on some satellites and can provide images of volcanoes. These images however, tend to be expensive and can not operate in poor weather conditions or uninterrupted winter nights.
Interferometric Synthetic Aperture Radar (InSAR)
InSAR is a remote sensing technique that acquires and compares radar satellite images, which have been recorded months and years apart, but from the same location in space, to provide information on large areas of deformation across a volcano. This deformation may be observed as swelling or uplift of several kilometres of the ground surface. The accuracy of this technique varies but in general can be obtained to within tolerance limits of a few hundred millimetres. InSAR measures regional, rather than localised ground movements of discrete points.
Light Detection and Ranging (LIDAR)
LIDAR is similar to RADAR technology, which can generate high resolution digital elevation models (DEM) with a vertical accuracy to approximately 100 mm. The LIDAR equipment includes a laser scanner, Inertial Navigation System (INS) and Global Positioning System, (GPS). The equipment is mounted on a small fixed-wing aircraft and a series of traverses are flown to obtain the imagery. The laser scanner transmits laser pulses to the ground surface, whereby they subsequently become reflected or scattered back to the laser scanner. The distance between the laser scanner and the ground can be calculated based on the time taken for the laser to travel from the instrument to the ground and then back to the instrument. The INS determined the direction of the laser and the GPS records the position of the aircraft. Following the acquisition of the data processing is required because multiple surfaces may reflect the laser beam (such as tree canopies and buildings), so that the ground surface position can be determined. The DEM is constructed from the data provided by the laser scanner, INS and GPS.
The three main types of LIDAR are; (a) Range finders, which are used to measure the distance from the instrument to the target, solid object (b) Differential Absorption Lidar (DIAL), which measures chemical concentration in the atmosphere, including ozone, pollutants and water vapour, and (c) Doppler lidar, which is used to measure the velocity of a target, based on the change in the wavelength of the lidar reflected from a moving target, such as atmospheric dust and aerosol particles.
Thermal imagery may provide a useful technique to locate ‘hot spot’ areas developing on the flanks of lava domes. A range of thermal imaging devises are available that may be deployed from the ground or remotely (usually from a helicopter). Intensifiers can magnify abnormal thermal hot spot zones which may intensify in the weeks or days before an eruption (Francis 1979, Tazieff, 1979, Baker 1979).
Even when there is a comprehensive monitoring network the present state of knowledge does not enable the forecasting of the exact timing and location of volcanic hazards or an impending eruption, but these may sometimes be estimated. However, the majority of active volcanoes throughout the world are not monitored and some are potentially dangerous to populated regions and life threatening. The levels of volcanic risk may be reduced by the implementation of disaster prevention, preparedness and emergency response measures. These may be undertaken as part of a volcano hazards assessment program. As part of the United Nations (UN) designated International Decade for Natural Disaster Reduction (IDNDR) this supported the implementation of monitoring and mitigation measures on a number potentially active volcanoes (see for example IAVCEI IDNDR Task Group 1990).
Is a volcano active or extinct?
Predicting whether a volcano is active or extinct may be one of the first steps to be undertaken before a volcanic hazards assessment can be undertaken. ‘Extinct’ is usually applied to volcanoes that have not erupted during historical times, however, catastrophic eruptions have occurred on volcanoes that were assumed to be extinct, for example; the eruptions of Mount Lamington, Paupa New Guinea in 1951 and Bezymianny, USSR in 1956. It has been estimated that the active life span of a volcano is probably between 1 and 2 years (Baker 1979).
The prediction, or forecasting of volcanic eruptions, relies on the ability to monitor and observe precursory symptoms of an eruption, which may be geophysical, geochemical, hydrogeological or related to changes in ground deformation, plume activity and so on (Gorshkov 1971; Mori et al. 1989, Anon 1971). Full scale monitoring of volcanoes if usually time consuming (data is required over many years) and expensive. Only few volcanoes are subjected to full scale continuous (24 hours a day, 7 days a week) monitoring. Even in these cases it is not currently possible to consistently predict the timing, magnitude and scale of a volcanic eruption. One of the principal objectives of predicting volcanic eruptions is to ensure there is sufficient time for an evacuation to take place so that lives can be saved.
Volcanic hazard mapping, vulnerability risk assessment and zonation
Volcanic hazard assessments are based on the general assumption that the future volcanic activity will be similar to the past activity in terms of the magnitude of the eruptions and eruption type. The expected recurrence interval of a particular type of eruptions, the magnitude of events, location of deposits, patterns of volcanic activity also must be assessed. The geological data obtained from monitoring is used in conjunction with historical records, if and when available and the inspection, examination and interpretation of historical and ancient deposits. All of this collective data and information is used to help predict future events and construct hazards maps. A hazard involves a degree of risk, the elements at risk being life, property, possessions and the environment. Risk involves quantification of the probability that a hazard will be harmful and the tolerable degree of risk depends upon what is being risked, life being much more important than property. The frequency of a particular hazard event can be regarded as the number of events of a given magnitude in a particular period of time at a certain location. The risk to society can be regarded as the magnitude of a hazard multiplied by the probability of its occurrence. Risk, as mentioned, should take account of the magnitude of the hazard and the probability of its occurrence. Risk arises out of uncertainty due to insufficient data and information being available about a hazard and to incomplete understanding of the mechanisms involved. The uncertainties prevent accurate predictions of hazard occurrence. Risk analysis involves identifying the degree of risk, then estimating and evaluating the risks. To quantify risk it is necessary to carryout vulnerability assessments. Vulnerability is the degree of loss, such as people, property, land, economy as a direct consequence of the occurrence of the volcanic hazards. Vulnerability is often expressed on a scale ranging from 0 (no loss or damage) to 1 (total loss or damage) and may be expressed empirically or analytically. The empirical approach assesses vulnerability by investigating the adverse impacts of a particular volcanic hazard during a previous event. The analytical approach considers the materials, construction types and designs and estimates the failure in response to a given physical situation caused by volcanic hazards. Volcanic risk for a particular hazards zone and a specified period of time may therefore be estimated as follows:
Volcanic risk = vulnerability x hazard probability x value
Value is expressed in financial (monetary) or numbers of people killed.
Volcanic hazards maps usually involve the mapping of previous eruptions to identify possible hazards zones. This may include; for example, the path taken lava flows, the possible route of pyroclastic flows, the direction of flow of lahars, lateral distances reach by lethal and non-lethal ejecta, area of possible ash coverage, ash plume developments and so on. Any spacial aspect of a particular hazard can be mapped providing there is sufficient information on its distribution. Hence, when hazard assessment is made of an area, the results can be expressed in the form of hazard maps. An ideal hazard map should provide information relating to the spacial and temporal probabilities of the hazard mapped. However, hazard maps, like other maps, do have disadvantages. For instance, they are highly generalized and represent a static view of reality. They therefore need to be updated periodically as new data becomes available. Volcanic hazard maps should represent a source of clear and useful information for disaster management, evacuation, preparedness and post disaster action. In this respect, they should realistically present the degree of risk (i.e. the probability of the occurrence of a hazard event). Descriptive terms such as high, medium and low risk must be defined, and there are social and economic dangers in overstating the degree of risk in the terms used to describe it. Ideally, numerical values should be assigned to the degree of risk. However, this is not an easy task as numerical values can only be derived from a comprehensive record of events, which unfortunately in the case of volcanic eruptions is available only infrequently. Furthermore, many events in the past may not have been recorded, which throws into question the reliability of any statistical analysis of data. Hazard zonation maps may be used by authorities to assist with disaster management, evacuation of inhabitants and for the development of contingency plans for future eruptions. They must however, not be replied on for a precise guide for the future activity and hazards associated with a volcanic eruption. Together with a comprehensive monitoring programme they provide a guide and a bench mark, indicating the possible best means to reduce injuries and loss of lives, mainly by offering as much time as possible for the evacuation of the areas of highest risk. Nonetheless, if those managing disasters can be provided with some form of numerical assessment of the degree of risk, then they are better able to make sensible financial decisions and to provide more effective solutions to problems arising.
Mitigating the effects of lava flows
Since lava flows tend to occur in one or more relatively narrow channels, fed by a tube or lava system, there have been a number of successful methods used to intervene, divert or stop lava flows, with varying success rates. These have included the following:
As lava flows are gravity driven, their paths may be predicted by an assessment of the topography and by the production of hazard zonation maps (Guest and Murray 1979, Chester et al. 1985). Remedial techniques were implementing during the 1983 and 1991 to 1993 eruption of Mount Etna, in Sicily, to reduce the affects of ‘aa’ lava flows that threatened the town of Zefferana and other parts of the volcano (Abersten 1984, Colombrita 1984, Barberi et al. 1993, GNV 1993). The following techniques were used to change the direction of flow:
Mitigating the effects of pyroclastic flows
Due to the total destructive power of pyroclastic flows the only methods that may be used to mitigate their effects is to evacuate people who live in their paths.
Mitigating the effects of lahars
As lahars are mainly controlled by topography and mainly controlled by incised river valleys, their effects may be mitigated. The mapping of older lahar deposist may give some indication on possible flow paths, run-out distances and vulnerable communities, enabling the production of hazards zonation maps. However, as was demonstrated at Nevado del Ruiz, in Colombia, the availability of such information is little use unless it is communicated properly to the authorities. In the upper reaches of valley which drain volcanoes trip-wires and seismometers may be installed to provide sufficient warning for people who live further down-stream, in time to evacuate before the arrival of the deadly lahars. Sediment traps, artificial earth dams and baffles can reduce the mass of debris in the flow and hinder flow rates. Dredging and the construction of levees may be used to control the sedimentation of river channels, estuaries and lakes, although the cost of these activities may be significant and not sustainable for developing countries.
Mitigating the effects of volanogenic landslides (debris avalanches)
Since debris avalanches and volcanogenic landslides travel at velocities in excess of about 360 kh-1 (100ms-1) there are no known measures that may be put in place to prevent their destructive effects. The only means of mitigation, toe reduce loss of lives, is evacuation. Houses, buildings and associated infrastructure located in the path of volcanogenic landslides will almost certainly become buried beneath huge volumes of debris, or severely destroyed when located on the outer fringes of the landslide. Geological mapping may enable slopes to be identified which are prone to failure, however, it is not currently possible to forecast the timing of failure, the volume of potentially displaced material, rate of deformation, the run-out distances and whether the event will trigger an eruption.
Mitigating the effects of tephra, ash and ballistic projectiles
Tephra (the term for all volcanic ejecta) can be the most disruptive, although less dramatic geological hazard. Ash in particular can accumulate of roofs of poorly constructed building and this increased load may induce their collapse. For example, when wet a layer of ash about 100 mm thick may cause the collapse of a corrugated iron roof. This problems may be alleviated by the building of sloping roofs or by local people become educated do their roofs are regularly swept to reduce ash accumulations (although during the Pinatubo eruption, children often fell from the roofs of houses and were killed whilst trying to clear the ash). Contingency plans may prevent the associated problems caused by ash, such as downing communication and power lines, making roads impassable and hindering travel. These may include for example, the ready availability of earth moving equipment, repair and maintenance services, filters on key power providing facilities, supplies of uncontaminated drinking water for livestock and people, face masks to prevent ash inhalation to reduce respiratory problems such as asthma and silicosis. Satellite based warning systems that can detect and track eruptions clouds, standardised protocols for aircrew and on-board ash detecting radar for air-line routes that cross parts of the Earth that are susceptible to volcanic eruptions are some of the measures that have been taken to reduce the problems caused by atmospheric ash on aircraft. Ballistic projectiles only affect the area within a few kilometres around a volcano. Buildings and house in general can not withstand the damage caused by large projectiles and therefore the only mitigative measure is the pre-evacuation of the hazardous zone.
Mitigating the effects of noxious gases
Some volcanoes continually emit sulphur rich gas that generates a persistent hazard. The gas may be carried by the prevailing wind to local communities who live on, or near the volcano, causing health problems, damage to agricultural land, livestock and crops. This may also be associated with the deposition of fluorine and other trace elements that can become fixed in animal feeds and water supplies. Prolonged exposure to this may cause teeth and bone problems in the local population. The effects form gas can only be mitigated by the evacuation and resettlement of the population effected by the gas. In other situations volcanic gases may note be continually present but can be emitted occasionally on both active and dormant volcanoes. For example, CO2, being less dense than air, may sink along the flanks of volcanoes and concentrate in topographic lows, basement and cellars. Monitoring for CO2 and CO and the deployment of automated gas alarms, which respond when gas concentrations approach critical concentrations. These must be implemented however, in conjunction with an education program, to help local communities better understand the potential problems associated with gas hazards and their mitigation.
Mitigating the effects of volcanogenic earthquakes and faults
The establishment of a seismograph network across a volcano is the only way to mitigate against volcanogenic earthquakes. Geological mapping and hazard zonation, paying particular attention to the delineation of fault outcrops and rifts zones may enable future structures to be situated well away form the zone of influence of such features that can be susceptible to aseismic creep. This has been a notable process on Mount Etna, in Sicily, causing damage to roads, retaining walls and buildings some which have collapsed (Lo Guidice and Rasa 1992; Rasa et al. 1996).
Mitigating the effects of volcanogenic tsunami
Ocean warning systems may potentially be used to provide a tsunami warning system, which if combined with effective disaster management plans could potentially educe injuries and loss of lives. There are no known mitigative measures that may be enforced to prevent structural damage and the protection of coastal towns.
Communication and awareness
Geologists’ are frequently required to convey the results, advice and recommendations from geological investigations to a variety of end users (e.g., policy makers, the public, the media). Often, it is the communication of the information that is the most challenging and can be more difficult than the investigation itself. This is because of the many potential end users that require the geological information. Many volcanic hazards investigations use highly sophisticated scientific techniques and geological terminology and when combined with cultural and language barriers, social, political, religious or economic constraints that frequently exist, this makes it difficult to convey the correct message, and for the recipient to understand the implications of the geological information. The failure to effective and accurately communicate this message, may reduce the usefulness of the information being provided. Communication must be considered part of a geological investigation because if the correct message. Good communication is important during the monitoring and prediction of volcanic eruptions.
There are several examples of successes and failures. For example, Nevado del Ruiz volcano is located in the Andean Cordillera of Colombia, approximately 100 km south west of Colombia’s capital city, Sante Fe De Bogota. On 13 November 1985, a plinian eruption generated series of pyroclastic flows which interacted with snow and ice, forming the summit ice cap. The rapid transfer of heat from the eruption, combined with the seismic shaking generated lahars (mud flows) and avalanches of saturated snow, ice, felled trees and rock debris. These flowed along drainage channels and within four hours had travelled over 105 km, descending 5100 m, leaving a wake of catastrophic destruction and obliterating everything in their path. The town of Armero was buried beneath the blanket of mud. Approximately 24,740 people were killed or missing, 4420 injured and 5092 made homeless. Geohazards investigations were undertaken at Nevado del Ruiz, prior to the 1985 eruptions. Previous pyroclastic flow deposits and lahars were mapped and their extent known, accurate reports of historical events exists, following a period of monitoring the volcano advise was available from Colombian and international scientists. In the months prior to the eruption, communications were established between geologists’ and Government. Geologists attempted to explain the significance of the observed precursory activity which included low intensity earthquake swarms, a steam (phreatic) eruption, explosions, ash-falls deposits and small lahars within 30 km of the summit. The Colombian Red Cross issued alerts to prepare for mudflows, but unfortunately these reports were not properly disseminated. Pyroclastic flows and surges were generated, but it was not announced that these events were significant and this was met with scepticism by the local authorities and the population. An evacuation of Armero was considered to be unnecessary by government officials (this may also have been influenced by the fact it was night with heavy rainfall). The violent lahars came in two surges, the first cold, the second hot and these engulfed Armero for at least 2 hours. The Colombian people were made aware of the consequences of an eruption, geological hazards maps were produced over a month before the fatal event. The catastrophe at Nevado del Ruiz and Armero was caused by a failures in communications, cumulative human error, misjudgement, indecision and bureaucracy (Williams 1990a,b).
Montserrat is a British dependant island located in the West Indies. The Soufrière Hills volcano, situated in the southern part of this island, has been in a state of almost continuous volcanic activity for the past 14 years, since 1995, after being dormant for about 400 years (Druitt and Kokelaar 2002; Donnelly 2007). The Montserrat Volcano Observatory (MVO) was established soon after the occurrence of phreatic eruptions on 12th July 1995. The eruption of the Soufriere Hills was an event for which the local population was completely unprepared. Pyroclastic surges and lahars have radiated from the volcano, travelling along river gullies towards the sea engulfing numerous villages. This has resulted in the loss of use of a large part of the island, including the airport, main jetty and capital town, Plymouth (a new airport and jetty has now been built, Plymouth has been evacuated of all its residents and is currently buried beneath volcanic deposits) there were some fatalities. During the early stages of the eruption some of the islanders and scientists were conscious of historical volcanic eruptions on neighbouring Caribbean islands. For instance, in 1909 the eruption of Mount Pelée on the island of Martinique, generated pyroclastic flows that killed at least 29,000 people. More recently, in 1976-1977, approximately 70,000 people on the island of Guadeloupe, were evacuated following a relatively small steam eruption on La Soufriere volcano which lasted about 9 months. No major eruption followed and the evacuation was considered to have been not necessary by many of the local people. There was a breakdown in communications between the geologists, Government and people, no lives were lost but there was significant negative economic impact on businesses and farms (Robertson 1995) The move towards the evacuation of much of the population of Montserrat resulted in a situation where the communication of geohazards to the Government and public was very important. In the early stage of the eruption on Montserrat different types of communications were established with the public. These included the daily issuing of statements via the media (TV and radio), regular meetings with community representatives and the issuing of newsletters. During the early stages of the eruption the author experienced the benefits of personal engagement with the local community. This supported more formal volcanic hazards announcements provided by the observatory, sometimes via the media and Government of Montserrat. An appreciation of interpersonal and social skills was necessary for creating an environment of trust within which a dialogue can be established to convey the necessary messages of the nature of volcanic eruptions and their implications for those threatened by them. This approach demonstrated the need for social and inter-personal skills as well as technical and scientific expertise, for the effective monitoring and communication of volcanic hazards.
At Rabaul Volcano, Paupa New Guinea, during the 1980s, the deployment of a ground deformation and seismic monitoring programme contributed significantly in ensured the public participation during contingency planning for an eruption. No more eruption occurred in the 1980s, but a major explosion in September 1994, enabled the now well-educated population and public officials to respond appropriately to the threat. As a result a well organised evacuation occurred with minimal deaths. The eruptions of Pinatubo in 1990 and Rabaul in 1994 have demonstrated how good communications between scientists and the authorities can result in a more positive response from the people affected by volcanic eruptions.
At Pinatubo, public lectures and meetings were held which included the viewing video footage (produced by Maurice and Katja Kraft for IAVCEI who were killed while filming the 1991 eruption on Unzen Volcano in Japan) showing the destructive affects of volcanic eruptions. This facilitated communications and help in persuading over 250,000 people to evacuate the region. It has been estimated this saved at least 10,000 lives.
Further information on communication in Geoscience and during the monitoring and mitigation of hazards associated with volcanic eruptions can be found in Donnelly 2008 (Montserrat), Petterson et al. 2008 (Solomon Islands) and Barclay et al. 2008; all in Liverman et al. 2008.
Index | Diagnostic Characteristics | Geographic Occurrence | Investigation & Mitigation | Key Contacts & Expert Advice | Photo Gallery | Essential References & Further Reading | Definitions & Glossary |