Lessons from Armero, Colombia

It should come as no surprise that efficient information flow is essential for effective hazards assessment and risk mitigation during a volcano crisis. That fact was made terribly clear on the night of 13 November 1985, when huge lahar triggered by a small eruption of ice-clad Nevado del Ruiz Volcano in Colombia killed more than 23,000 people, most of them asleep, and virtually erased the hapless town of Armero from the map.

The first warnings had come almost a year earlier. Climbers on the 5,200-m peak noticed increased fumarolic activity and felt earthquakes starting in November 1984 and, on 22 December, three widely felt earthquakes were accompanied by small explosions from Arenas Crater near the volcano’s summit. Felt earthquakes and abnormal fumarolic activity persisted into 1985. Seismometers installed at the volcano in July confirmed a high level of earthquake activity that intensified in early September before a strong phreatic eruption on 11 September. A preliminary volcano hazards map published in October warned that a magmatic eruption would almost surely trigger damaging lahars. The final version of the map, issued publicly on 8 November 1985, placed the town of Armero squarely in a zone of high lahar hazard. Five days later, and more than 8 hours after the beginning of the paroxysmal eruption, Armero and most of its residents disappeared under a river of mud (Herd and Comité de Estudios Vulcanologicos, 1986).

In hindsight, this terrible tragedy could have been avoided. The volcano provided ample warning, an accurate hazards zonation map was available, the volcano was being continuously monitored, an evacuation order was issued hours before the lahar reached Armero, and the safety of higher ground was within walking distance for most residents. What went so terribly wrong? In spite of hard work by many and heroic efforts by some, essential information about how to mitigate the impending disaster was not available where and when it was critically needed. The advance warning was not understood by many of those at risk, and they received no short-term instructions to move to higher ground as the giant lahar bore down on them. In the words of one scientist who experienced the Armero tragedy firsthand: ‘... the catastrophe was not caused by technological ineffectiveness or defectiveness, nor by an overwhelming eruption or an improbable run of bad luck, but rather by cumulative human error’ (Voight, 1996, p.719). Shortly after the disaster, the US Geological Survey (USGS) and the Office of Foreign Disaster Assistance of the US Agency for International Development (USAID OFDA) established the Volcano Disaster Assistance Program (VDAP) to mitigate losses from future eruptions worldwide by applying the lessons learned so tragically at Armero.

Source: Volcano Deformation by Dzurisin

Volcano Hazards Mitigation -- A Complicated Business

Faced with the daunting task of trying to unravel the tremendous variety and complexity of volcanic processes, especially during a rapidly evolving volcano crisis with lives and property at risk, today’s volcanologist can ill afford to rely on any one specialty to furnish the necessary tools and information. Instead, diverse techniques from geology, hydrology, seismology, geodesy, geochemistry, and related fields must be brought to bear and, if necessary, tailored to the situation at hand.

Experience worldwide has shown that no single technique can provide unambiguous information about the cause and likely outcome of volcanic unrest. For example, earthquakes near a volcano might indicate fracturing of brittle rock around a pressurizing magma body, intrusion of magma or other fluid into brittle host rock, or release of tectonic strain not directly related to magmatic processes. Even the occurrence of long-period (LP) earthquakes that are commonly associated with magmatic or hydrothermal processes is inherently ambiguous. Such earthquakes are generally thought to indicate the presence of a two-phase fluid in a resonating crack (Chouet, 1988, 1996a, 2003). However, this fluid could be gas exsolved from fresh magma that is rising and decompressing, hypersaline brine derived from stagnant magma that is cooling and crystallizing, or gas-charged magma itself. Without complementary data from other monitoring techniques to help resolve this ambiguity, the precise significance of LP earthquakes for volcano hazards is difficult to discern, especially during a crisis. Furthermore, some models for LP earthquakes require a relatively narrow range of conditions for resonance to occur, suggesting that some (most?) fluid transport occurs without generating LP earthquakes. Even when they do occur, swarms of shallow LP earthquakes could be caused by pressurization of a magma body as a result of intrusion, boiling within a hydrothermal system in response to magmatic or tectonic changes, or migration of fluids released from a stagnant, cooling magma body. The hazards implications vary widely and can be explored further only by combining seismic information with that from other disciplines.

By itself, any single volcano-monitoring technique provides information that is ambiguous or subject to alternative explanations. For example, an increase in the flux of sulfur dioxide gas might indicate the recent arrival at shallow depth of fresh, gas-charged magma, but the absence of such an increase does not preclude the same interpretation. This is because SO₂ is very soluble in water and can be effectively ‘scrubbed’ by groundwater (Doukas and Gerlach, 1995). Carbon dioxide is less soluble in water than SO₂ , so an increase in CO₂ followed by an increase in SO₂ might be attributed to newly emplaced magma that released both CO₂ and SO₂ and eventually dried out a conduit to the surface, allowing SO₂ to escape. For confirmation, though, the volcanologist is likely to reach for additional tools from seismology and geodesy.

No amount of monitoring is likely to provide definitive information about the explosivity of newly emplaced magma, and thus about the extent or degree of hazards posed by an impending eruption. For this, the volcanologist relies on the tools of field geology and stratigraphy, under the assumption that the volcano will likely behave in a manner similar to the ways it has behaved in the past. Past performances are chronicled in the stratigraphic record of past eruptive products, but the record is always incomplete and impossible to read unequivocally. In some cases, past eruptive behavior can be highly variable and therefore is a poor guide to the next event. Moreover, there is always the possibility of an unusual or unprecedented event, such as the catastrophic landslide and lateral blast at Mount St. Helens in 1980. In hindsight, similar events had occurred before but the evidence lay buried until the landslide and subsequent erosion cut deeply into the volcano’s north flank (Hausback and Swanson, 1990). Only by combining information from a variety of sources can scientists hope to decipher the cause of volcanic unrest, and thus provide useful information to public officials charged with the difficult task of reducing risk while minimizing social and economic disruption.

source: Volcano Deformation by D. Dzurisin

Volcano in motion - when deformation gets extreme

For me, the idea that volcanoes stretch, bulge, shrink, sink, or rise wholesale out of the sea is emotionally stirring and intellectually captivating. Knowing that the ground we live on is constantly being shoved around by subterranean processes, including buoyant ascent of molten rock from deep within our planet's interior, is wondrous beyond words. The next four samples attempt to capture a sense of the visceral excitement that fast-moving volcanoes elicit in those who study them. From bradyseisms  at a Roman marketplace to the famous bulge at Mount St. Helens -- such are the 'true legends' of volcano geodesy.

1. The Ups and Down of a Roman market - Phlegraean Fields Caldera, Italy

Consider the case of the coastal town of Pozzuoli, Italy, Which sits near the center of a partly submerged caldera called Campi Flegrei (Phlegraean Fields). Serapeo, an ancient Pozzuoli marketplace constracted by Roman artisans in the first century BC, has two generation of floors and columns: the second was built after the first sank below sea level. Borings in the columns by Lithodomus lithopagus, a marine mollusk that burrows into rock for shelter, indicate as much as 11 m of subsidence by about 1000 AD. The area then rose about 2 m from 1000 to 1198, when an eruption occurred at nearby Solfatara Crater. Uplift continued unabated after eruption and accelerated in about 1500. By 1503, the amount of uplift was so great that a portion of newly emerged shoreline was deeded to the local university! Net uplift during the period of 1000-1538 was about 12 m.

The uplift rate accelerated dramatically during several years before an eruption that began on 29 September 1538. The most rapid uplift occurred on 26-27 September, when the ground level at Pozzuoli rose at least 4-5 m in 48 hours, causing the shoreline to recede 40 m. On 28 September, hot and cold water poured from fissures in the uplifted block, and on the morning of 29 September the uplifted ground subsided by as much as 4 m. Land near what was soon to become the eruptive vent stopped subsiding about noon on 29 September and began to rise again, so that "... by eight o'clock it was as high as Monte Ruosi -- i.e., it was as high as that hill where the little tower stands".

2. Remarkable uplifts in the Galapagos Islands--Fernandina and Alcedo Volcanoes

There are other notable examples of volcanoes deforming at a breakneck pace, including two from Galapagos Islands, Ecuador. In 1927, three fishing boats anchored for the night at Punta Espinoza near northeast point of Fernandina, a frequently active shield volcano that forms one of the islands. During the night, the crews were awakened by violent bubbling of the water. Two of the boats put to sea immediately, but the third was unable to leave because, by the time the crew recognized the danger, the seafloor had risen and their keel was grounded! In the morning, their boat was stranded by an uplift of 'several feet', presumably caused by a shallow magmatic intrusion. The rusted remains of the hapless sailors' boat can still be seen on the beach.

At nearby Alcedo volcano, a 6-km length of shoreline including Urvina Bay was uplifted by as much as 4.6 m, probably in early 1954. More than 1 km² of coral reef became exposed above sea level. The uplifts was so rapid that fish were stranded in pools before they could escape to the sea.

3. Rabaul Caldera, Papua New Guinea, 1994

Another striking, more recent example of volcano rising noticeably out the sea occurred in September 1994 at Rabaul Caldera, Papua New Guinea, prior to simultaneous eruptions from two vents on opposite sides of the caldera rim. The following composite account is excerpted from Lauer. 'There was a massive jerking of the house as if it had been slammed by a truck. I leaped up, wide awake. It was the biggest guria (earthquake) we'd had for months and months. Nick (husband) promptly went back to sleep. I looked at the clock  -- 2.50 a.m. (Sunday, 18 September 1994) ... The gurias continued all day Sunday ... Around 5. p.m. .. the ground was in almost constant motion ... The seismographs (at Rabaul Volcano Observatory, RVO) were a mass of almost unreadable black scribble and the recording pens were having difficulty holding up with the constant movement ... About 7.45 p.m. I got into my car to drive around to a friend's house for a game of cards. The whole car rocked. I noticed a number of people walking and carrying bed-rolls and bags. Interesting -- they obviously knew something I didn't ... Next thing there was a loud thump-thumping noise. It was 5.50 a.m. (19 September) and a helicopter was landing next to bedroom ... They were going to check around the harbour. Even from where we were standing the uplift was visible ...' Incredibly, a tide gauge on the eastern side of Vulcan had been lifted nearly out of the water overnight, indicating uplift of the seafloor by about 5.5 m.

' ... After the helicopter took off I went to ring a couple of friends who were still in their homes to warn them. Seconds later Lynden (daughter) came running in, yelling that there was black smoke coming out of Tavurvur ... It had been just 27 hours since the first guria to the eruption of Tavurvur. Not long after (71 minutes) there followed a mighty BOOM! Vulcan, on the opposite side of the harbour to Tavurvur, had decided to join the eruption. Explosion followed explosion. Rocks and ash were hurled with amazing force skywards. Day seems to become night as the heavy black cloud oozed its way across town. The eeriness was broken by fork lightning and claps of thunder.'

A massive eruption column quickly rose into stratosphere. During the peak of the eruption, the column towered 20-30 km above the residents of Rabaul town and nearby communities, who self-evacuated in the face of obvious danger. About 90,000 people were displaced but fewer than 10 died as a result of the eruption or its immediate effect. The number would surely have been much higher had it not been for the vigilance of RVO and the Rabaul Disaster Plan, completed in 1983, which spelled out four alert stages, established evacuation routes, and called for practice evacuations -- the last of which was held in April 1994, just five months before the eruption.

4. The Bulge at Mount St. Helens, 1980

Among the better known examples of runaway volcano deformation is the famous bulge on the north flank of Mount St. Helens, which developed during the two months preceding the catastrophic debris avalanche and eruption of 18 May 1980. Electronic distance meter (EDM) and theodolite measurements tracked the surface of the bulge marching and nearly horizontally northward at rates of 1.5-2.5 m/day from 23 april through the early of 18 May. By 12 May, the high point of the bulge stood 150 m above pre-existing topography, and by 18 May the net volume increase was a staggering 0.11 km³. Working on the volcano during the build-up to the 18 May eruption was an exhilarating experience. Titanic forces were at work, preparing to disassemble the volcano, while the struggled to understand what was happening beneath our feet.

If these stories don't quicken your pulses, there's no reason to read further. Honestly, it doesn't get any better than this.

source: Volcano Deformation by D. Dzurisin

The Beginning

He [Pliny the Elder] was at Misenum in his capacity as commander of the fleet on the 24^th of August, when between 2 and 3 in the afternoon my mother drew his attention to a cloud of unusual size and appearance. He had had a sunbath, then a cold bath, and was reclining after dinner with his books. He called for his shoes and climbed up to where he could get the best view of the phenomenon. The cloud was rising from a mountain – at such a distance we couldn't tell which, but afterwards learned that it was Vesuvius. I can best describe its shape by likening it to a pine tree. It rose into the sky on a very long ‘trunk’ from which spread some ‘branches’. I imagine it had been raised by a sudden blast, which then weakened, leaving the cloud unsupported so that its own weight caused it to spread sideways. Some of the cloud was white, in other parts there were dark patches of dirt and ash. The sight of it made the scientist in my uncle determined to see it from closer at hand.

From Pliny the Younger’s eyewitness account of the 79 CE (Common Era; substitute for AD, Anno Domini) Vesuvius’s eruption, translated from his Letter 6.16 to the historian Tacitus by Prof. Cynthia Damon (2000), Amherst College, Massachusetts.

Almost two millennium after Pliny the Younger’s insightful description of the Vesuvius eruption that resulted in the death of his uncle and entombed the residents of Pompeii and Herculaneum, modern volcanology is still in its infancy. Like a child with boundless enthusiasm and seemingly endless opportunities, volcano science has yet to develop fully an identity of its own. Instead, it freely draws upon such diverse disciplines as geology, seismology, hydrology, geochemistry, and geophysics. Fewer than half of the scientists who study volcanoes would call themselves volcanologists, and therein lies a great strength. Volcanology is inherently a multidisciplinary pursuit with ample room for specialists and generalists alike – physicists, chemists, biologists, hydrologists, limnologists, dendrochronologists, ecologists, remote-sensing specialists, and many others. Scientists who study volcanoes often find themselves looking beyond their own expertise for answers, surrounded by others doing the same. For novices and experts alike, volcanoes are humbling but very exciting subjects to explore.

In spite of the best efforts of countless volcano watchers dating back at least to Pliny the Younger, no unifying theory yet exists to explain the diversity and complexity of magmatic and volcanic systems. This may be due in part to the tremendous breadth of volcanic phenomena (e.g., bubbling mudpots and lava lakes; steaming hot springs and jetting lava fountains; lavas that flow freely, others seemingly too stiff to move, and others still that explode into towering ash columns and swift pyroclastic flows; geysers that entertain onlookers by erupting on schedule and rare, caldera-forming cataclysms that threaten entire civilizations). To make matters worse, all of this diversity is served up with a heavy dose of complexity. Earth’s deep interior is not only hidden from direct view, but it is also heterogeneous in virtually every way imaginable. Before it explodes or oozes onto the surface, magma forms deep within the Earth, rises through extreme conditions of pressure and temperature, interacts physically and chemically with diverse crustal rocks, cools, partially crystallizes and degases, sometimes mixes with other magmas of similar complexity, responds to regional tectonic influences, and interacts to varying degrees with groundwater. Each of these processes leaves an imprint on the eruptive or intrusive end products, so the complete sequence of events can be exceedingly difficult to decipher.

All of this says nothing about the additional complexity that arises when magma finally reaches the surface, where it interacts with the atmosphere, hydrosphere, and biosphere. During the past two centuries, more people have died as a result of tsunamis, lahars (volcanic mudflows), and post-eruption starvation and disease associated with eruptions than as a direct result of eruptive processes. On the island of Hawai 0 i, many residents’ complaints of unusual respiratory problems have been attributed to ‘vog’ (volcanic fog) or ‘laze’ (lava haze), a product of previously undocumented interactions between seawater and basaltic lava flows as they pour into the Pacific Ocean from a long-lived eruption of Kilauea Volcano (Monastersky, 1995). Atmospheric scientists still debate the relative contributions of anthropogenic and volcanic sources of several gases that play an important role in sustaining life on Earth. Because they reflect our planet’s inherent complexity from mantle to stratosphere, volcanoes present an immense challenge to those who strive to understand them.

Responding to crises that threaten the lives and livelihoods of thousands of people, volcano scientists at the beginning of the 21st century still rely to a great extent on collective experience and intuition to anticipate the outcome of volcanic unrest. In the words of two scientists who played key roles in the mitigation effort surrounding the historic 1991 eruption of Mount Pinatubo, Philippines, ‘Successful mitigation of volcanic risks requires correct forecasts, effective warnings, and a willingness of public officials and citizens to take necessary precautions. Volcanologists should be prepared for obstacles and delays at each step and should be aware that, even when progress is being made toward such forecasts, warnings, and precautionary actions, the margin of safety can be alarmingly narrow.’ (Newhall and Punongbayan, 1996, p. 807). Happily, this situation is improving rapidly. The past 25 years have witnessed several important advances in our understanding of volcanic processes, products, and attendant hazards. Progress has been spurred both by technological improvements and by inexorable encroachment of cities onto the flanks of dangerous volcanoes. Tragically, a major impetus for this change was the highest death toll from a single eruption since the beginning of the 20th century. In 1985, more than 23,000 people died as a result of lahars triggered by an eruption of Nevado del Ruiz Volcano in Colombia – the deadliest eruption since 1902, when pyroclastic flows from Mont Pelée claimed nearly 30,000 lives in the town of St. Pierre on the island of Martinique, West Indies (Fisher et al., 1980; Fisher and Heiken, 1982).

In response, volcanologists focused their efforts on eruption prediction and volcano-hazards mitigation, with encouraging results. Volcano seismology has evolved from a descriptive science based on cataloging events toward a deterministic theory of the mechanisms of several distinctive types of earthquakes that commonly occur beneath volcanoes. Volcano geochemistry has taken its place with seismology and geodesy as a mainstay of modern volcanology, in part by providing some of the earliest and most definitive indicators that fresh magma was involved in several recent episodes of volcanic unrest that culminated in eruptions. Remote-sensing techniques have been applied to volcanoes with increasing frequency and success, promising a future of global volcano surveillance by spaceborne sensors.

The global telecommunications revolution, especially the Internet and World Wide Web, has profoundly changed the way in which scientists monitor volcanoes and share information. Although the presence of trained scientists at volcano crises will continue to be essential for effective hazards mitigation, in most places it is now possible to establish a virtual volcano observatory to share information globally in near-real time. Scientists monitoring a remote volcano can draw upon the expertise of colleagues around the world, even as hazardous events are unfolding. At the same time, advances in volcano geodesy have brought powerful new tools to bear on problems of volcano monitoring and hazards assessment. At the start of a new millennium, volcanology seems to be passing from infancy to adolescence, anxious to establish a unique identity among its more mature siblings in the Earth sciences.

This blog focuses on one aspect of volcanology’s recent advances that I have been fortunate to experience firsthand – a revolution in volcano geodesy. During my career, not one but two breakthrough techniques for measuring surface deformation from space have burst on the scene: the Global Positioning System (GPS) and interferometric synthetic aperture radar (InSAR). Earth scientists have coined the phrase ‘space geodesy’ and put the concept to good use studying geodynamic processes, including volcanism. Other technological advances have made it possible to measure crustal strain in situ with amazing precision over a very wide range in frequency, blurring the distinction between seismology and geodesy. Although traditional techniques are still the best choice for many tasks, a growing collection of new tools has greatly extended the reach and capability of classical geodesy.

In one sense, volcano geodesy is a very specialized field with relatively few practitioners. On the other hand, the subject is sufficiently intuitive to have widespread appeal among many non-specialists who are fascinated by volcanoes and curious about how they work. This blog is by no means a comprehensive treatment of modern geodetic techniques applied to volcanoes, nor is it a historical account of the development of volcano geodesy. For a more systematic treatment of geodetic principles and practices, the reader might want to consult a reference book on geodesy or surveying such as Bomford (1980), Davis et al. (1981), or Krumm et al. (2002). For an overview of geodetic techniques applied to volcanoes, the article on ground deformation methods and results by Van der Laat (1996) is a good choice. Readers interested in a more general treatment of volcanology or volcano hazards might want to consult excellent texts on those subjects by Williams and McBirney (1979), Latter (1989), Francis (1993), Scarpa and Tilling (1996), Fisher et al. (1997), or Decker and Decker (1998).

Introduction

Volcanoes and eruptions are dramatic manifestations of dynamic processes within the Earth, mostly but not exclusively localized along the boundaries of Earth's relentlessly shifting tectonic plates. Anyone who has witnessed volcanic activity has to be impressed by the variety and complexity of visible eruptive phenomena. Equally complex, however, if not even more so, are the geophisical, geochemical, and hydrothermal processes that occur underground -- commonly undetectable by human sense -- before, during, and after eruptions. Experience at volcanoes worldwide has shown that, at volcanoes with adequate instrumental monitoring, nearly all eruptions are preceded and accompanied by measurable change in the physical and (or) chemical state of the volcanic system. While geochemical methodologies of volcano monitoring have shown increasing sophistication and promise in recent decades, seismic and geodetic (ground-deformations) techniques remain the most widely used tools in volcanic surveillance. These two geophysical workhorses have proven to be robust and, generally, the most diagnostic and reliable techniques for volcanologists.

In the early 20th century, systematic measurements of ground deformation were initiated at few active volcanoes in Japan and the USA. Since then, VOLCANO GEODESY -- a specialized field of the still-young science of volcanology --  has come along way, not only in term of greatly increased diversity and precision of measurements, but also in the number of volcanic systems being monitored worldwide. There has been a virtual explosion of volcano-geodesy studies and in the modeling and interpretation of ground deformation data over the past three decades. Nonetheless, other than selective, brief summaries in journal articles and general works on volcano-monitoring and hazard mitigation (e.g., UNESCO, 1972; Agnew, 1986; Scarpa and Tilling, 1996), a modern, comprehensive treatment of volcano geodesy and int applications was non-existent, until now.

Prologue

On 28th Septembre 2012, For the first time I go to a volcano. It is Krakatau volcano which is recorded has massif eruption on 26th and 27th August 1883. That massif eruption remained three main islands (Panjang, Rakata and Sertung island). Since 1927, new island rose in the middle of those islands See figure 1.

Figure 1. Maps of Krakatau in 2008

My team and I went there with to conduct GPS survey for monitoring volcano activities because it erupts on 2nd Septembre 2012. We actually have seven GPS stations, one in each  surrounding islands and four in Anak Krakatau but two of them are gone because of previously eruption.So, what we do is only conducting survey on the remainning GPS stations.