The Warning Period
How did so many people get caught by surprise by Ruiz’s catastrophic lahars, in spite of accurate risk assessments and intensive efforts at public education? How is it that so many people lost their lives in Armero alone? The purpose of this chapter is to describe the sequence of events that took place following the initial volcanic activity on November 13, 1985, in order to examine what types of decisions were made in response to cues and messages about the impending disaster.
The preceding chapter describes what was done to create a public awareness of possible volcanic eruptions and mudflows. The creation of awareness does not itself, however, assure that people will take appropriate actions when emergencies occur. One goal of the study team was to investigate what actions individuals actually take when they receive information about an imminent threat. Social scientists have found (Perry, et al., 1981) that individual responses depend on the following factors:
what type of message they get—formal or informal, environmental or from humans, specific or vague;
what type of confirmation of the message they get;
whether or not they view the threat as real;
whether or not they think the threatening event is likely to affect them personally;
whether or not they visualize the potential impact of the threatening event as serious enough to warrant a particular personal protective action;
whether or not they can devise a “plan” for how to protect themselves and can execute that plan of action for themselves and their families.
Warning messages can come in several forms:
unofficial estimations of the danger and recommendations for prudent response;
official statements of the danger and recommendations for prudent response.
Typically, to activate the official warning system, information is relayed from scientific observers to emergency response officials. The latter then decide who is in danger, who they need to mobilize to make an appropriate response, and what types of protective actions should be recommended to those at risk. All stages in the process are likely to involve considerable uncertainty about what the scientific information means and what the appropriate response would be (Mileti et al., 1985).
It is of interest to consider information on both what messages were sent from regional level emergency system decision-makers and what responses were made to them by local level emergency system decision-makers. To facilitate this, the warning period for the Nevado del Ruiz November 13 eruption and lahars is described here in terms of
the sequence of events related to the decision by responsible officials that an evacuation warning should be given, and
the sequence of events in Armero that preceded the arrival of the mudflow.
A brief sketch is provided here of the warning communication system and major decision-making points on November 13, 1985. This is based on information obtained during the team visit to Colombia, as well as some written accounts (Herd, 1986; Lima, 1986). Not all accounts of the sequence of events on November 13 agree with respect to the details. Also, an important part of the story was lost with those Armerans who perished. The evacuation decision process of the emergency officials included several other communities in Tolima and Caldas, and deaths occurred in other towns as well. 1 The brief description here focuses on the events related to the disaster in Armero.
The major groups involved in the warning system on November 13 for the Department of Tolima included the following:
scientists in the city of Manizales, where the volcano data were compiled and analyzed, and those at observation posts on Ruiz;
emergency committee members in Manizales, linking the scientists, national level emergency response decision-makers, and regional emergency committees;
regional (located in Ibague, capital of Tolima) and local (e.g., in Armero) emergency response officials;
regional and local private emergency response agencies (e.g., Red Cross);
traditional information sources (e.g., radio, priests);
3:00-5:00 P.M., NOVEMBER 13
Around 3:00 p.m., Ruiz emitted a black column of ash, the result of a phreatic explosion (mainly steam). Although Ruiz often is shrouded in clouds and not visible, some people north of the volcano observed this ash column. About two hours later, ash began falling in Armero, which is east-northeast of the volcano.
Within about an hour of the eruption, the regional director of Civil Defense (CD) in Ibague received information about its occurrence over the CD radio. 2 He made contact with the regional director of INGEOMINAS in Ibague, who considered the event to represent a serious threat to the vulnerable populations and recommended that preparations for evacuation be made. The regional CD director also contacted the national CD director in Bogotá, who instructed him to alert CD stations in northern Tolima.
5:00-7:00 P.M., NOVEMBER 13
A regularly scheduled session of the regional emergency committee was convened at 5:00 p.m. Members were briefed by the scientific representative for INGEOMINAS. Using the volcanic hazards map that had been prepared by INGEOMINAS, the towns in Tolima—including Armero—that warranted evacuation preparations were specified. It was suggested that police stations in Armero and neighboring towns be alerted immediately.
There are various reports that traditional information sources (e.g., radio announcements, or the Catholic priest who provided reassurances at a 6:00 p.m. mass) urged people to stay calm and to stay indoors. Some say that after a while the ash stopped falling in Armero, suggesting that the eruption was over. About 7:00 p.m., a severe storm arose in the vicinity.
7:00-9:00 P.M., NOVEMBER 13
When the emergency committee meeting ended about 7:00 p.m., some of the members, including the INGEOMINAS representative, proceeded to the Red Cross office to request that the towns of Armero, Mariquita, and Honda be prepared for evacuation. 3 It had been reported that the Ibague Red Cross,
while communicating by radio to its Armero office at about 7:30 p.m., ordered an evacuation. This apparently was to be executed by local officials.
In Armero there was heavy rainfall and frequent thunder, which may have masked sounds and movements associated with the volcano. The storm also caused intermittent electrical outages throughout the evening. If all the alerts reported to have been given in Armero were actually given— we were unable to confirm them all—at least the Red Cross, the Civil Defense, and the police would have known that the volcano was now viewed as a serious threat by the members of the regional emergency committee. However, there were no reports that systematic efforts to warn large numbers of Armero residents had actually taken place.
Although most families evidently used the lack of information and general uncertainty as justification for continuing routine activities, some families did leave Armero during this time. At the same time, the storm and the darkness made conditions under which to convince the public to leave the town difficult at best.
9:00-11:00 P.M., NOVEMBER 13
A few minutes after 9:00 p.m. a paroxysmal eruption and some explosions commenced on Ruiz. The ice melt accompanying the eruption led to a large accumulation of glacial meltwater around the crater that was then channeled into the rivers draining the flanks of the volcano. This water picked up soil and debris in the river valleys, thereby creating the lahars that damaged the communities along the rivers.
Specific evidence about the lengthy tremor that accompanied the eruption was not available until seismograph records were collected the following day. Immediate information about rumblings and explosions apparently came mainly from scientific observers on the mountain. An airline pilot reported the thick column of smoke and ash. River observers on the mountain made visual sightings of the mudflows rising in the rivers and radioed this information to others.
At about 9:45 p.m., CD officials in Murillo radioed the regional CD director in Ibague that Ruiz had erupted. He attempted to radio Armero to order an evacuation, but could not make contact. There is another account of a river observer above Armero also being unable to make radio contact with the CD office in Armero to report the mudflow.
Around 10:30 p.m., approximately the time a lahar destroyed part of Chinchina on the west side of the mountain, the regional CD director overheard various other radio transmissions directed at Armero. These included messages from the towns of Ambalema and Murillo that an “avalanche” (the word often used in Spanish) was approaching and Armero should be
evacuated, and a conversation between a CD official in Líbano and one in Armero in which the latter was advised to leave for safety.
11:00 P.M.-1:00 A.M., NOVEMBER 13-14
Sometime after 11:00 p.m., Armero was inundated by a swift flow of watery mud, rocks, and other debris. Thus, there may have been as much as 1-3/4 hours in which local officials could have sounded a general alert and directed residents to seek higher ground. Many residents could have reached higher ground on foot in less than half an hour, and thousands of lives might have been saved.
Some interviewers reported that the mayor of Armero was talking with someone over his ham radio and saying that he did not think there was much danger when he was swept away. Also, a Red Cross staff person in Ibague reported that he was discussing the situation by radio with the Red Cross in Armero only to have the conversation abruptly cut off when the mudflow overran the office.
Some citizens of Armero reported in interviews with our research team that they received unofficial warnings by telephone from relatives and friends. These calls most probably came from Líbano, which is about 12 miles upstream from Armero and close enough to the river that the mudflow was clearly heard, if not seen, in the dark. Families who received this information and acted quickly had time to escape.
Officials from the impacted areas and towns informed us that, for the most part, the nature of the impending event simply was not understood. They reported that some thought that the major consequence would only be ashfall, or a flood, not unlike a recent one that had done some damage in Armero and seemed to pose no real threat. When the event was described as mud, the mayor of Armero was described to have envisioned a slow-moving substance affording plenty of time to take action later. A national television station had broadcast news of the 9:00 p.m. eruption, but most remember the message as having indicated that there was no cause for alarm. It is likely that many people in the community were totally unaware of any of the events that had taken place after the midafternoon ashfall.
With the electricity out and the weather stormy, efforts to escape a little-understood natural phenomenon with a very rapid onset were unsuccessful for most. Many survivors report that they were asleep by 11:00 p.m. and took flight only after hearing others running and shouting in the streets after the first of the surges of mud had reached the community. There was much confusion in the streets, with no electricity, explosions from gas storage facilities, and many people trying to leave in their cars. Most families did not possess cars and went on foot. Those near enough to some of the higher
elevations in town and having the presence of mind to head in that direction were able to escape on foot. But for an estimated 22,000 Armerans, there was no escape from death in the swirling mud. Thousands of others in the area were injured. Most of those who made it to high ground ahead of the mud, along with those who escaped from it, spent the remainder of the night in the rain on the hills along the mudflow’s path.
News of the disaster was quickly spread by radio operators. Red Cross paramedics from Ibague informed other emergency response personnel and left for Armero after losing radio contact during a conversation with the Armero Red Cross office. They had no other information about what was happening at this time. They were among the first responders on the scene, arriving at Armero around 1:00 a.m. Nonetheless, rescue and medical efforts were somewhat limited until daylight.
SUMMARY AND CONCLUSIONS
A warning system was in the process of being developed at the time of the November 13 eruption. A meeting was scheduled for November 15 to address, in part, how local officials were to transfer the warning message to the public in Armero and instruct the populace in exactly what to do.
Prior to the sighting of mudflows, the decision to call for community evacuation was based on a fairly high level of uncertainty about its need. A recommendation that people leave their homes amidst the darkness and rain for a fairly unspecified reason would not have been effective.
Advice given to townspeople earlier in the day relating to the ashfall ran counter to what would have been an appropriate message about preparing to leave the area. Even with eyewitness information about the mudflow, there apparently was not sufficient understanding of what the eruption meant to the town, with the consequence that responsible officials discounted the extreme nature and imminence of the threat and took insufficient action.
Although some public information materials had been distributed, it is not certain that a large proportion of the population had received information about the hazards associated with the volcano. Further, the available materials had not provided adequate specificity about the mudflow hazard in particular, and what to do if a warning was given. Although some families are said to have left the area throughout the day and evening, for most families the first effective message came from either seeing the advancing mud or hearing or seeing others fleeing from it.
Since the night of November 13, the Colombian preparedness agencies have made some efforts to correct the obvious deficiencies in the warning system and evacuation process. These actions are discussed further in Chapter 8 .