Space Shuttle photo of Martinique with Mt. Pelee on the north left end of the island. Pelee is famous for the May 8, eruption which killed 29, people and destroyed the city of St. This is the largest number of casualities for a volcanic eruption this century.
Photograph of Mt. Pelee by Heilprin, May 26, Photograph of the remains of St. Pierre by Heilprin, Ships in the harbor smoldered and sank. When rescuers eventually did enter the ruins, they pulled from a jail cels survivor of the disaster, Louis-Auguste Cyparis, who later toured with the Barnum and Bailey Circus. In a paper , he argued that this may explain why no one thought to evacuate St.
Pierre in the days before the eruption — the impending calamity was simply beyond comprehension. The catastrophe led geologists to invent a term for the blast that destroyed the city. In modern parlance, geologists would categorize this deadly mix of hot gas and rock as a type of pyroclastic flow, examples of which have since been observed during other volcanic eruptions, including Mount St.
Helens in Some said the eruption broke through the newly formed lava dome and spilled sideways out of the lowest point of the crater, which faced St.
Others said a massive column first rose high into the air, then collapsed under its own weight. With only eyewitness accounts and deposits of erupted material to go on, scientists have struggled to resolve the question. Overlooking the destruction of St.
Pierre and the bay. Credit: Angelo Heilprin, public domain. Pierre, and on Aug. In a letter to Science , Heilprin reports that the spine, just over a hundred meters wide at the base, grew at astonishing rates.
It rose 10 meters during one eight-day period and 6 meters in another span of four days, and at its peak, loomed meters above the rim of the crater.
Pictures of St Pierre after the eruption, from Lacroix C,B Were taken in May 11th, , 3 days after the passage of the pyroclastic current, from a boat a few hundred meters off the coast.
It shows the Mouillage and Center districts only partially destroyed. See Figure 9 for a closer view of St Pierre city map. Several studies have focused on the pyroclastic current deposits and their sedimentological characteristics, and are compiled in the deposit distribution map of the May 8th, eruption Figure 3 and also listed in Table 1.
Two types of pyroclastic current deposits have been identified:. Figure 3. Distribution map of the May 8th, eruption, with the ash-cloud surge and the block-and-ash flow inundated areas shown in gray and pink, respectively modified after Bourdier et al.
Main rivers are marked by blue dashed lines. The pre-eruptive extent of the city of St Pierre is represented by the blue area. The location of samples used in this study are represented by the white numbers from Bourdier et al. Table 1. Sample locations refer to the white numbers in Figure 1. Roobol and Smith estimated the devastated area at 58 km 2 , but using a geo-referenced digital elevation model we obtained a value of Fisher et al.
Total thickness of this stratified bed varies by more than a meter at Fond Canonville to a few millimeters south of St Pierre. The cumulated thickness of the 3 units, measured from twenty different locations by Bourdier et al. All studies attribute this deposit to a turbulent ash-cloud surge Fisher et al.
Figure 4. Units U1 to U8 are from Bourdier et al. Unit U1, corresponds to the May 8th deposit, separated in three layers, i. Only the middle and lower unit are visible at St Pierre. The picture on the right is a closer view of the U1 lower layer, sampled in the northern part of St Pierre yellow star in Figure 2 where the deposit is directly in contact with the pavement of the city. The white scale bar is 20 cm long. Using a variety of indicators, the local direction of propagation of the ash-cloud surge was reconstructed.
Lacroix estimated the direction of the ash-cloud surge to be parallel to Victor Hugo Street in St Pierre red line in Figure 3 using the N-S orientation of the remaining standing walls and the N-S alignment of the dead bodies in the streets. From measurements in cross-bedded deposits, Fisher et al. However, authors disagreed on the direction, either being from north to south or from the block-and-ash flow to the southwest.
The dynamic pressure of the ash-cloud surge can be estimated from its effects on buildings, especially in St Pierre, following the study of Jenkins et al. Comparing the damage at Merapi volcano Jenkins et al.
Based on the dynamic pressure calculation method of Jenkins et al. The rest of the city, only partially damaged, was exposed to a dynamic pressure less than 2 kPa, as deduced from the presence of standing walls in the Center and Mouillage districts, and then pressure drops under 1 kPa in the southernmost part of the city attested by the standing cathedral towers in Figures 3C,D.
Following these extensive field studies, various interpretations of the May 8th, pyroclastic current source conditions and its internal dynamics have been inferred. We regroup them in two main theories:. This concentrated flow is thought to have been generated from the collapse of a short column formed by an intra-crater vertical explosion Fisher et al.
This idea came from Hill who initially located the source to be approximately at Morne Lenard 2. Later, Fisher et al.
Because of the two differing interpretations outlined above, the exact nature of the eruption source conditions i. Moreover, despite insightful descriptions of the pyroclastic current deposits, the total deposit volume is still missing. No study has estimated the volume of each separate current May 8th and 20th, June 6th, or August 30th as the field studies conducted following the eruption compiled the effects of the individual currents.
Nevertheless, the total volume of the May 8th pyroclastic current, as well as the volume portions of the block-and-ash flow and ash-cloud surge components, are still unknown. The numerical model used in this study is the newer two-layer version of VolcFlow, which was developed to more accurately simulate the dynamics and extent of pyroclastic currents Kelfoun, This version was used to simulate: i block-and-ash flows and ash-cloud surges at Merapi Volcano Kelfoun et al.
The code is based on two coupled, depth-averaged currents: one for the basal concentrated flow also called block-and-ash flow in this study and one for the overriding ash-cloud surge. The dynamics of each current are modeled using depth-averaged equations of mass and momentum balance in the x and y directions.
The ash-cloud surge requires an additional equation, as density varies in time and space due to loss of mass through sedimentation. The two layers are then coupled and exchange mass and momentum following two exchanges laws arrows in Figure 5. The complete description of the physical model, the equations, and all the parameters used in VolcFlow are summarized in Supplementary Material.
The reader can also refer to Kelfoun and Gueugneau et al. Figure 5. Sketch of the general model of the two-phase version of VolcFlow Kelfoun, To simulate stresses applied to the concentrated flow during transport using a depth-averaged approach, the plastic rheological law is used, involving a constant retarding stress T see Supplementary Material. Despite the lack of physical explanation for applying this rheology to pyroclastic currents, several studies have demonstrated the ability of the constant retarding stress to reproduce various features of such currents and their deposits Kelfoun et al.
The ash-cloud surge is simulated as a turbulent continuum that loses momentum due to turbulent drag stresses. To perform the numerical simulations, the Observatoire Volcanologique et Sismologique de Martinique provided a 5 m resolution LiDAR DEM of Martinique Island, constructed in , that was down-sampled to 10 m to save computational time.
Despite the current debate regarding the source conditions that generated the May 8th pyroclastic current i. The crater shape seems to have reoriented the expansion of the fragmented material into this V-shaped outlet. This caldera is approximated in the DEM by a bowl-shape of m wide for m deep, and centered roughly on the lava dome, as illustrated in Figure 6.
With this modification, the outlet on the southern part of the crater rim is reconstructed at the same location as it was prior to May Figure 6. This simplified source can thus model the collapse of either a short column or a lava dome. The resulting overflow is self-regulated and dependent on the supply rate in the crater, as illustrated in Figure 6C blue curve. Because of the synthetic crater, a large part of the deposit volume remains stuck inside the depression and does not feed the simulated pyroclastic current.
Therefore, we distinguished the volume of material supplied in the crater V ini to the total volume of deposit V that escaped through it to constitute the pyroclastic current.
Figure 6. The red line in B highlights the position of the newly formed V-shape crater outlet. Table 2. Input parameters for the best-fit VolcFlow simulation presented in Figure 7. Only the scenario of Fisher et al. Consequently, the higher the block-and-ash flow velocity, the higher the surge production. To model such complex dynamics of the pyroclastic current, the code requires 11 input parameters for each simulation see Supplementary Material. Each of these parameters has a clear influence on the morphology of the simulated flows and on the resulting deposit footprint, which makes it relatively easy to estimate the best fit.
To better see the influence of each parameter on the model dynamics, the reader can refer to Kelfoun et al. To quantitatively evaluate the simulation results and to identify a best fit simulation, the differences between simulated and observed flows are calculated using validation metrics, which compare areas inundated by the simulation Asim to the real deposit Aobs.
The matching area between simulated and observed flows is called true positive TP , the over-simulated area is called false positive FP and the missing simulated area is called false negative FN. Three coefficients were used:. The reader can refer to Charbonnier et al. Values obtained for the best-fit simulation are included in Table 3. Table 3. Values of validation metrics used for evaluating the best-fit simulation. For more than a hundred simulations of 10—20 h each, seven unconstrained input parameters Table 2 were adjusted to obtain a combination of highest possible values for the three metrics, obtained for the 90th simulation.
The volume of material in the crater and the supply duration are first adjusted by matching the general aerial distribution and thicknesses of the simulated deposits to the real one. Then, the surge characteristics production and sedimentation parameters as well as the constant retarding stress of the concentrated flow are adjusted to find the combination of highest values for the three metrics for the area covered by the ash-cloud surge only.
The choice to focus primarily on the ash-cloud surge was motivated by the fact that: 1 its extent and limits, as extracted from the field, are based on robust evidence and therefore only contain small uncertainties, and 2 it covered a much larger area than the block-and-ash flow, restricted to valleys.
However, the maximum runout in the south of St Pierre is underestimated where the simulated ash-cloud surge traveled m less that the real flow. Also, a large part of the area inundated by the ash-cloud surge around the northern part of the crater is not reproduced by the simulations.
Since the location of the initial mass flux in the simulations was set to be in the southern crater outlet, the simulated ash-cloud surge derived from the block-and-ash flow in the proximal area was unable to spread northward and inundate that part of the crater. Figure 7. Results of the best fit simulation obtained using the input parameters presented in Table 2.
A Final distribution of the simulated deposits from the best-fit simulation, which include the extent of the simulated ash-cloud surge green color scale for the thickness and the simulated block-and-ash flow pink color scale.
For ease of comparison, outlines of the observed ash-cloud surge and block-and-ash flow, as extracted from the field, have been added with a white and black outline, respectively. B—E Sequence of four snap shots of the best-fit simulation at 30, , , and s after the mass starts to overflow through the crater outlet, showing the propagation of the flows overlain on the DEM.
Comparisons between the simulated surge deposit thicknesses with those measured at 20 locations in the field Bourdier et al. This sleepy little village shows little of the grandeur of turn-of-the-century St. Pierre, which was a vibrant colonial city, known to European tourists as the "Paris of the West Indies.
In the official census, the population of St. Pierre was around 20, Although most were native Martiniquans, the wealth and political power were controlled largely by Creoles and a few French colonial officials and civil servants. No one at the time could have predicted the horror that was to descend on this tropical paradise with the reawakening of Mt. Although in January Mt. This changed, however, on April 23 when minor explosions began at the summit of the volcano.
Over the next few days, St. Pierre was rocked by earth tremors, showered in ash, and enveloped in a thick cloud of choking sulfurous gas. These nightmarish conditions deteriorated further when the city and outlying villages were invaded by ground-dwelling insects and snakes driven from the slopes of Mt. Horses, pigs, and dogs screamed as red ants and foot-long centipedes crawled up their legs and bit them.
Thousands of poisonous snakes joined the fray. An estimated 50 humans, mostly children, died by the snake bites, along with some animals. As the summit eruptions intensified, water in the Etang Sec crater lake was heated to near boiling.
On May 5, the crater rim gave way, sending a torrent of scalding water cascading down the River Blanche. The hot water mixed with loose pyroclastic debris to generate a massive lahar with a downslope speed of nearly kilometers per hour. This large volcanic mudflow buried everything in its path.
Near the mouth of the river, north of St. Pierre, it overran a rum distillery, killing 23 workmen. The lahar continued into the sea, where it generated a three-meter-high tsunami which flooded the low-lying areas along the waterfront of St.
Living near the volcano became increasingly stressful, leading many to consider leaving St. Pierre for Martinique's second city, Fort-de-France. On the day of the lahar, however, Governor Louis Mouttet received a report from a committee of civic leaders who climbed the volcano to assess the danger.
The only scientist in the group was a local high school teacher. The report stated that " there is nothing in the activity of Mt.
0コメント