- Email: [email protected]
Field observations of the 2008 summit eruption at Piton de la Fournaise (Ile de La Réunion) and implications for the 2007 Dolomieu collapse
Journal of Volcanology and Geothermal Research 191 (2010) 60–68
Contents lists available at ScienceDirect
Journal of Volcanology and Geothermal Research j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j vo l g e o r e s
Field observations of the 2008 summit eruption at Piton de la Fournaise (Ile de La Réunion) and implications for the 2007 Dolomieu collapse Thomas Staudacher Observatoire Volcanologique du Piton de la Fournaise (OVPF), Institut de Physique du Globe de Paris, CNRS, UMR 7154 – Géologie des Systèmes Volcaniques, 14 RN3, le 27ème km, 97418, La Plaine des Cafres, La Réunion, France
a r t i c l e
i n f o
Article history: Received 22 July 2009 Accepted 8 January 2010 Available online 21 January 2010 Keywords: Piton de la Fournaise summit eruptions lava flow emplacement lava temperature dyke emplacement caldera ring fault
a b s t r a c t More than one year after the most important eruption ever observed at Piton de la Fournaise in April 2007, a new eruption started in 2008 at Piton de la Fournaise, with three different eruptive phases in September, November and December. They were located within the 340 m deep Dolomieu crater. Due to the steep (dip angle 45–80°) and unstable walls of the new crater formed in April 5, 2007, no measurements in situ of the 2008 eruption had been possible. Only observations from the Dolomieu crater rim and from helicopter were performed, using an infrared camera, a portable rangefinder and numerous photographs. These field observations allowed precise monitoring of the setting up of the lava flows with time. The total erupted volume of lava of the three phases was 2.2 × 106 m3 and the average flow rates ranged between 0.3 and 1 m3 s− 1. Lava temperatures of up to 1150 °C have been measured by an infrared camera. Overall, infrared images of the Dolomieu crater illustrate the control of the eruptive vents by the structure of the April 2007 Dolomieu collapse. © 2010 Elsevier B.V. All rights reserved.
1. Introduction Piton de la Fournaise is the active volcano in the South-East of Ile de la Réunion in the western Indian Ocean. It is located within the 8 km wide and 13 km long U-shaped Enclos Fouqué caldera, which is open to the sea (Fig. 1). Two craters crown the summit, the about 250 m wide Bory crater and the 1 km wide Dolomieu crater. Piton de la Fournaise produced over the last century about 150 eruptions and emitted more than 1000x 106m3 of lava (OVPF compilation). The last decade was particularly active with 27 eruptions producing 367 x 106m3, about 1/3 of the estimated volume emitted during the last century (Peltier et al., 2009; Staudacher and Peltier, 2010). Amongst this eruptive sequence, the April 2–May 1, 2007 eruption represents a major event. It was with >140 x 106m3 of lava flows the most voluminous and intense eruption for at least two centuries (Staudacher et al., 2009; Peltier et al., 2009). This voluminous eruption, which took place at a low elevation of 590 m in the Grand Brûlé area (Fig. 1), was the cause of the simultaneous Dolomieu caldera collapse. The rock column between the Dolomieu crater and the magma chamber collapsed into the latter on April 5 and 6. In 24 hours, a 1 km long, 800 m large and 340 m deep funnel-shaped caldera was formed (Urai et al., 2007; Michon et al., 2007; Staudacher et al., 2009). Even though the April 2007 eruption emptied between 30 and 40% of the shallow magma reservoir (Sigmarsson et al., 2005; Peltier et al., 2007, 2008) Piton de la Fournaise erupted again in 2008 within the
E-mail address: [email protected]. 0377-0273/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.jvolgeores.2010.01.012
Dolomieu crater. Three eruptive phases formed a 420 × 220 m large and about 75 m deep lava flow on the crater floor.
2. Overview of the 2008 summit eruption 2.1. September 21 eruptive phase Following the April 2007 eruption and the collapse of the central Dolomieu crater the whole Piton de la Fournaise massif was destabilized and needed to find a new state of structural stability. One of the consequences was the structural readjustment of the Dolomieu crater; its dimensions reduced about 3 m and 1.2 m in north–south in east–west directions respectively, and a subsidence of 40 to 99 cm of its border was recorded by the permanent GPS network (Staudacher et al., 2009). This subsidence lasted from April 2007 to July 2008. Between March and July 2008, a normal background seismicity of less than 10 volcano-tectonic seismic events per day was observed beneath the summit (OVPF internal report, 2008). On July 28 the general trend of subsidence reversed. The permanent GPS network recorded a new inflation of the summit cone (Fig. 2), with a crater elongation of about 5.6 cm on a north–south axis and 3.3 cm on a west–east axis. No significant tiltmeter or extensometer variations were recorded. The inflation was accompanied by an increase of seismicity beneath the summit and by repeated, 30 min to 3 h lasting seismic crises on August 4, 15 and 31 with 46, 471 and 241 events, and on September 8, 9, and 15 with 104, 514 and 296 events, respectively. Magnitudes of the seismic events ranged between −0.3
T. Staudacher / Journal of Volcanology and Geothermal Research 191 (2010) 60–68
61
Fig. 1. Map of Piton de la Fournaise volcano inside of the Enclos Fouqué caldera. Shown are the Bory and Dolomieu summit craters and the summit permanent GPS stations BONG (Bory north), SNEG (Soufrière north-east), DERG (Dolomieu east), DSRG (Dolomieu south) and BORG (Bory), as well as the Tunnel Cathérine tiltmeter station. Other tiltmeter stations are close to the corresponding GPS stations. The April to May 2007 lava flow is shown in dark grey in the Grand Brûlé and the 2008 lava flow is shown in black within the Dolomieu crater. Coordinates are Gauss Laborde in meters.
and 2.6, but most were smaller than Md 1 (OVPF internal report, 2008). The first appearance of new activity was a short period of volcanic tremor on September 12, between 6:00 and 16:00 (GMT). No magma arrived at the surface, but simultaneous release of gas from the southwestern Dolomieu crater wall at about 2350 m elevation was observed. SO2 was detected by the 3 km distant NOVAC network on the Enclos Fouqué caldera (Garofalo et al. 2009). On September 21, at 11:05, a 30 min seismic crisis preceded a new eruption in the Dolomieu crater. Only insignificant GPS horizontal and vertical variations of <1 cm accompanied this seismic crisis. The eruptive vent was located on the western crater wall about midway between the floor and the top of the crater at 2340 to 2380 m elevation (Gauss Laborde coordinates [m]: 179080/37260–179110/ 37200). After a peak of intensity on the first day of the eruption, the tremor decreased strongly and reached a somewhat constant value (Fig. 3a). A higher and irregular eruption tremor was recorded on
September 25. The eruption ceased on October 2, with the disappearance of the eruption tremor at 01:30. The lava flow, emitted during the nine days of eruption accumulated on the Dolomieu crater floor (Fig. 4a) and stretches to about 300 × 160 m (Fig. 5a). The elevation of the surface was measured from the Dolomieu crater border with a TruPulse 200 laser rangefinder to 2205 m above sea level (Fig. 6b). These measurements had been made from different sites around the Dolomieu crater on bright rocks lying on the lava flow or on gravel adjacent to the lava flow border. Successive measurements allowed us to estimate an error of ±3 m. Based on the depth of 340 m of the Dolomieu crater measured after the April 2007 collapse (Staudacher et al., 2009), this technique allowed us to estimate the maximum thickness of the lava flow to about 50 m. The emitted volume was estimated between 0.8 and 1.0 × 106 m3 and the average flux was 0.9 m3 s− 1 (Section 4). As the lava flow was inaccessible for direct sampling, Pele's hairs have been collected for geochemical analysis (Di Muro et al., in
Fig. 2. Curves of four permanent summit GPS stations, showing pre, co- and post eruptive ground variations between January 2008 and April 2009. Data points are daily averaged values and show an uncertainty of about ±0.5 cm. Eruptions are shown in red straight lines and yellow background. Easting is represented in blue, northing in red. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
62
T. Staudacher / Journal of Volcanology and Geothermal Research 191 (2010) 60–68
Fig. 3. Eruption tremor recorded by the Bory seismic station of September 21 (a), November 27 (b) and December 14 (c) eruptive phases.
preparation) on a 1 m2 fabric placed previously on the Bory crater floor. Lava temperatures up to 1150 °C were measured by a FLIR ThermaCAM PM695 from the Dolomieu and the Bory crater rims (Section 3; Fig. 4b and Table 1). 2.2. November 27 eruptive phase During October and November, the seismicity remained high with 20 to 100 events per day, amongst which an event of Md 3.3 occurred on November 12, 21 h37. Several seismic crisis occurred on October 20 and 31 with 191 and 226 events, respectively (OVPF internal report, 2008). The latter was recorded on the tiltmeter network with significant variations at La Soufrière station (radial component + 22 μrad; tangential component +5 μrad), at Bory station (radial component − 5 μrad; tangential component −6 μrad) and Tunnel Catherine station (radial component − 9.5 μrad; tangential component −3.5 μrad). Station locations are shown in Fig. 1. Two further crises happened on November 6 and 21 (114 and 239 events respectively). The permanent GPS network recorded a Dolomieu crater opening between November 5 and 25 of 2.6 cm and 2.2 cm on a
north–south and west–east trending axis, respectively (Fig. 2). On November 27, a new seismic crisis started at 07:25, simultaneously with a north–south extension of the Dolomieu crater of 7 cm and a west–east extension of 3.5 cm, recorded by the permanent GPS stations (Fig. 2). At 07:49 an eruption tremor appeared (Fig. 3b). The eruptive vent was adjacent to the September vent (Fig. 5b). A Ushaped lava flow partly covered the lava flow emitted in September and October on the Dolomieu crater floor (Figs. 5b and 6c). The eruptive phase stopped next day, after only 26 hours of activity. Less than 0.1 × 106 m3 of lava were emitted at an average flux of <1 m3 s− 1. As for the September eruptive phase, Pele's hairs have been collected on a fabric exposed in the Bory crater. 2.3. December 14 eruptive phase Immediately after the November eruptive phase, seismicity increased again with up to 40 low energy events (most of which were smaller than Md < 1, (OVPF internal report, 2008)) per day. A slight summit inflation was also recorded by the radial component of the Soufrière tiltmeter (87 μrad). The permanent GPS network continued to
Fig. 4. Visible and infrared images of the lava flow in the Dolomieu crater, recorded on September 22 from the northern Dolomieu boarder, close to “La Soufrière”. Temperatures are given in [°C].
T. Staudacher / Journal of Volcanology and Geothermal Research 191 (2010) 60–68
63
Fig. 5. Mapping of the three different eruptive phases. a) First phase: September 21 to October 2 lava flow in red. The dashed line indicates the limit of the lava flow on the crater floor and the crater wall; b) second phase: November 27 and 28 lava flow in yellow; c) and d) third phase: December 14 to 19 and December 24 to February 2 in different shades of blue, the darkest one being the most recent lava flow extension. Black lines display the eruptive vents. The sites where the infrared images were taken are shown in Fig. 5a; B = Bory; S = La Soufrière. The dotted oval curve denotes the proposed ring fault of the caldera. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
record summit inflation (Fig. 2). Between November 28 and December 14, a Dolomieu crater elongation of 5.1 and 4.8 cm on the north–south and west–east axis, respectively, was recorded. Compared to August– November, the inflation center shifted from the west to the east of the Dolomieu (Peltier et al., submitted for publication). On December 14, at 04:10, a 19 h-long lasting seismic crisis started, with a total of 1006 earthquakes. Such a long seismic crisis preceding a summit eruption is unusual for the Piton de la Fournaise volcano. It was accompanied by insignificant surface deformations (<1 cm) recorded by the permanent GPS network. At 23:00, an eruption tremor appeared beneath the summit and announced a third eruptive phase. Two vents opened in the Dolomieu crater at about 2340 m elevation (Figs. 5c and 7b). The main vent was located at 179550/37610 [m] (Gauss Laborde coordinates) on the northern crater wall; a secondary vent was located at 179750/37430 [m] on the north-eastern crater wall. This second vent was active for a few days and produced only an insignificant amount of lava. It is noticeable that the elevation of the eruptive vents is identical to the elevation of the vents of the September and November eruptive phases. The intensity of eruption tremor was quite irregular and increased until December 24. It stayed at a high level until January 1, 2009 (Fig. 3c). At 10:00 on the same day the tremor started to decrease and disappeared completely at 17:00. It reappeared on January 2 at 06:10 and continued at a low and very stable intensity until February 2, before disappearing on February 4. September and November lava flows on the Dolomieu crater floor were first flooded from the east by aa-type lava. In addition, from December 17 gas poor lava around the former lava flow started to appear, forming pahoehoe-type lava (Fig. 5c). Two weeks after the beginning of the eruptive phase, most of the lava flows emitted from the northern vent flowed down through channels beneath the surface of the Dolomieu crater floor. Resurgences of gas poor lava from below formed large pahoehoe plateaus on the boarder of the lava flow (hot area in Fig. 7a and
grey area in Fig. 7b). The lava flow and its elevation grew up continuously (Fig. 6d–g); the final elevation of the surface was determined after the end of eruption by use of the TruPulse 200 rangefinder to 2230 m and by the end of the December to February eruptive event, the surface has raised up for a further 25 m (Fig. 6g). Frequent pictures taken from the Dolomieu crater rim revealed that older lava flow surface emitted during the September eruptive phase was still visible in the western part (area inside of the ellipses in Fig. 6b–g). The total volume of the lava flows emitted during the September 2008 to February 2009 was estimated to 2.2 × 106 m3 (see Section 4). The December eruptive phase produced therefore 1.3 × 106 m3, with an average flow rate of 0.3 m3 s− 1. Again we determined the temperature of the lava flows with a FLIR ThermaCAM PM695 camera from Bory crater and from the Dolomieu border just above the active vent. Only very few Pele's hairs for geochemical analysis were found on the surface close to the former “La Soufrière” site just above the eruptive vent and on a new fabric in the Bory crater. 3. Lava temperature measurements During the September and December 2008 eruptive phases, infrared images had been taken using a FLIR ThermaCAM™ PM695. The temperatures were determined by the ThermaCAM Researcher Pro 2.9 program. The infrared images of the September eruptive phase were taken always in two ranges (0–500 °C and 350–1500 °C) from the Bory crater just above the September eruptive vent (B in Fig. 5a) and from the Dolomieu crater rim, close to La Soufrière (S in Fig. 5a) at slope distances from the vent of 385 m and 680 m respectively. For the lava flow itself (Figs. 4b and 7a) slope distances were estimated for each individual point. The images of the December eruptive phase were taken from the same sites and the slope distances to the eruption vent ranged between 400 and 935 m respectively. In order to determine temperatures, we
Fig. 6. Sequence of the Dolomieu crater filling between November 20, 2008 (b) and March 23, 2009 (g). Photographs taken from the Dolomieu crater rim show the Dolomieu crater before the 2008 eruptions (a). Numbers indicate the elevation of the Dolomieu floor and marker points on the south crater wall as well as the elevation of the lava flow surface (b–g). Ellipses mark the area visible since the end of the September lava flow. (h) A sketch of the filling process showing consecutive events, pink: September lava flow, orange: early December lava flow, red: late December lava flow.
64 T. Staudacher / Journal of Volcanology and Geothermal Research 191 (2010) 60–68
T. Staudacher / Journal of Volcanology and Geothermal Research 191 (2010) 60–68 Table 1 Temperature evaluations for the September and December eruptive phases. Date
Time
Observation site
Distance (m)
Temperature (°C)
± (°C)
22/09/08 22/09/08 22/09/08 22/09/08 22/09/08 22/09/08 15/12/08 15/12/08 19/12/08 19/12/08
09:23 09:23 09:23 12:48 12:49 12:49 10:53 11:28 10:24 11:00
Bory — Sept 08 vent Bory — Sept 08 vent Bory — Sept 08 vent Soufrière — Sept 08 vent Soufrière — Sept 08 vent Soufrière — Sept 08 vent Bory — Dec 08 vent Soufrière — Dec 08 vent Soufrière — Dec 08 vent Bory — Dec 08 vent
385 385 385 695 695 695 935 430 400 935
1140 1100 1120 1125 1145 1150 930 900 930 1050
50 65 60 60 70 70 40 50 30 50
corrected for distance, relative humidity, emissivity and air temperature. Air temperatures were 10± 5 °C in September and 15± 5 °C in December; emissivity was 0.95 ± 0.05 and relative humidity was 70± 10% for both eruptions. In September the temperatures of six measurements ranged between 1100 and 1150 °C (Table 1). Such temperatures are typical for a fresh magma emitted by Piton de la Fournaise eruptions (Coppola et al., 2007; Vigouroux et al., 2009). They are in good agreement with temperatures of 1150–1190 °C determined by a plagioclase-glass equilibrium study on the Pele's hairs emitted from all the three 2008 eruption phases (Di Muro, personal communication). Temperatures of the December eruptive phase are significantly lower, ranging between 900 and 1050 °C. The later represent four measurements taken on December 15 and 19 (Table 1). Such low temperatures might be due to an older and colder magma batch, but could also be an artifact due to unsuitable conditions during recording. In particular they are not in agreement with temperature determination on the Pele's hairs from this eruptive phase (see above). In fact the December vent had the form of a hornito and direct view into the hot magma from above was almost impossible. In addition volcanic gases were much more abundant in December. Such conditions might have prevented the measurements of reliable temperatures and therefore they should only be taken as lower limits. 4. Cartography and volume estimations The high frequency of observations allowed us to follow the extrusion of the lava flow through time. Mapping of the lateral extent of lava flows had been performed by using pictures taken from helicopter at high altitude and observations from the Dolomieu crater
65
rim, as well as from Aster images after the end of the eruptive phases (M. Urai, pers. information). The final extension of the September lava flow was determined to be 300 × 160 m (Fig. 5a), with a total surface on the Dolomieu crater floor of 0.033 × 106 m2. The extension of the December lava flow was determined as 420 × 220 m (Fig. 5d), with a total surface on the Dolomieu crater floor of 0.092 × 106 m2. The estimation of the volumes of the different lava flows was performed using the mapped lava flow surfaces and their corresponding depths. The latter were based on changes in height of the lava flow surface relative to marker points (see Fig. 6). The horizontalness of the lava flow was checked by measurements of three other marker points around the lava flow. Fig. 6a shows the Dolomieu caldera on October 12, 2007. The caldera floor was located at 2155 m elevation, 340 m below the eastern boarder (not visible here) and 375 m below the southern rim of Dolomieu. Two striking points, a large bolder and the sharp lower end of a scree at 2214 m and 2246 m respectively are indicated, as well as the southern Dolomieu crater rim at 2530 m. After the September eruptive phase, the crater floor has risen up about 50 m to an elevation of 2205 m (Fig. 6b). The November eruptive phase (Fig. 6c) only partly over-topped the September lava flow, with a thickness of about 5 m on the southern part. In particular the area within the ellipse was not covered by the U-shaped lava flow (Fig. 6c). On December 19 (Fig. 6d) the lava flow reached 2212 m elevation and five days later it attained the bolder at 2214 m elevation (Fig. 6e). On January 20 (Fig. 6f) the surface has raised up to 2227 m elevation, while the area inside of the ellipse was still visible. This was still the case after the end of eruption (Fig. 6g). The lava flow surface is currently at 2230 m elevation and some of the “old” September lava flow surface is still present. New measurements in November 2009 directed on sulfur-rich deposits on the lava flow gave us an elevation of 2233 ± 2 m above sea level, in excellent agreement with the former value. The sketch in Fig. 6h shows a possible scenario for the evolution of lava emplacement in 2008 inside the Dolomieu crater. It seems that lava from the December eruptive phase undermined the September lava flow through lava tubes and emerged at the borders in the form of degassed lava, forming the pahoehoe-type lava plateaus (Fig. 7). During this process it uplifted the “old” surface which “floated” on a liquid reservoir. Of note is the lava flow surface marked by the ellipses in Fig. 6b–h that is visible from November 20, 2008 to March 23, 2009. From these observations and taking into account the visible lava flow surfaces we performed different estimation of the emitted volumes. We assumed that the pre-eruptive Dolomieu crater bottom looked like a) an
Fig. 7. Infrared (a) and visible (b) image of the lava flow on the Dolomieu crater floor, on January 23, 2009, seen from the Bory crater. In order to enhance temperature differences in the much lower temperature range, we used a different colour code than in Fig. 4.
66
T. Staudacher / Journal of Volcanology and Geothermal Research 191 (2010) 60–68
Table 2 Estimation of emitted volumes and TADR of the December 2008 to February 2009 eruptions for different time intervals. Date
Time interval (days)
Emitted volume (106 m3)
TADR (m3 s− 1)
15/12/08 16/12/08 17/12/08 19/12/08 24/12/08 28/12/08 04/02/09 Total
0.42 1.60 0.73 2.25 5.08 3.96 37.5 51.5
0.05 0.01 0.05 0.06 0.32 0.52 0.48 1.49
1.41 0.11 0.81 0.33 0.72 1.51 0.16 0.33
inverted elliptical cone or b) an axial conoïd. Finally we estimated the emitted volumes of the December lava flows at different dates. From a) and b) we derived volumes of 0.9 ± 0.1 × 106 m3 for September eruptive phase and 2.2 ± 0.3 × 106 m3 for the September+ November + December eruptive phases. Consequently the December eruptive phase produced 1.3 ± 0.3 × 106 m3. The volume emitted by the November eruptive phase was small, estimated at ≤0.1× 106 m3. Intermediate volumes of the December eruptive phase had been estimated between December 15 and the end of eruption (Table 2). We used the estimated lava flow surface and the variation of the elevation of the uprising lava flow between these dates (see Fig. 6 and Table 2). The intermediate TADR (Time Averaged Discharge Rate (Harris et al. 2007)) ranged between 0.1 and 1.5 m3 s− 1 (Table 2, Fig. 8). A high value was determined for the first hours of the eruption, as had been often observed in the past (Coppola et al., 2005; Staudacher et al., 2009). The eruption rate slowed down to 0.11 m3 s− 1 on the second day and increased until the end of December, simultaneously with the increasing tremor intensity (Fig. 8). Between December and early February, the low eruption rate of 0.16 m3 s− 1 accompanied the episode with low seismic tremor (Fig. 8), in agreement with observations by Battaglia et al. (2005). The resulting total volume of 1.5 × 106 m3 (Table 2) estimated in this way agrees with the value of 1.3 × 106 m3 estimated above. 5. Influence of the Dolomieu caldera on the 2008 eruptive activity In April 2007, a major eruption occurred in the Grand Brûlé, about 8 km away from Piton de la Fournaise at low elevation (Staudacher et al., 2009). The lateral withdraw of magma lowered the pressure in the shallow magma chamber. One consequence of that pressure drop was the rupture of the roof of the magma chamber and the collapse of the overlying rock column, forming an about 1 km large and 340 m deep
slightly elongated caldera in the Dolomieu crater (Fig. 9a) (Urai et al. 2007; Michon et al., 2007; Staudacher et al., 2009). The size and the effective depth of the collapse were however not well known. Field observations immediately after the collapse in 2007 showed that a narrow band at about 2350 m elevation on the caldera wall was marked by vapor emanations and wet rock almost all around the Dolomieu crater (Fig. 9a). This zone was also characterized by elevated temperatures of up to 200 °C. High temperatures between 50 and 100 °C were still observed since then (Fig. 9b and d). These vapor emanations persist until now (Fig. 9c) and seem to correspond with the arrival of hydrothermal fluids, eventually through the ring fault formed by the 2007 collapse between the crumbled rock column and the bed rock, as already suggested by Michon et al. (2007) and Staudacher et al. (2009). The feeding dykes of the 2008 eruptive phases emerged at the same elevation (2340 m) as did the hydrothermal emanations. It seems that they followed the same pathway and that the ring fault acted as a channel for the dyke migration from the magma chamber to the surface as proposed by Opheim and Gudmundsson (1989) and Backström and Gudmundsson (1989). Fig. 10 suggests such a scenario, even though we do not know whether the 2008 eruptions were fed directly from the magma chamber along the ring fault or from residual magma stored in the volcanic massif or within the collapsed column. These observations might give new insights into the structure of the 2007 Dolomieu collapse, in particular the size of the collapsed “piston” (Fig. 10). If the elevation of the emanations and the 2008 magma correspond in fact to the limit between the collapsed rock column and the bedrock, then the size of the “piston” which slumped into the shallow magma chamber had a size of only about 690 × 500 m and does not cover the whole Dolomieu crater. Taking into account a volume of the caldera of 96× 106 m3 (Urai et al., 2007) to 90× 106 m3 (Staudacher et al., 2009) we can estimate an initial depth of the caldera of about 350 m. This depth is in excellent agreement with the depth of 340 m determined in May and June 2007, shortly after the collapse (Staudacher et al., 2009) and confirms the fact that no large breakdowns followed the April 5 to 6 collapse and that the final depth of the caldera was reached almost immediately. After the main collapse material from the Dolomieu crater walls slid down and only minor landslides of the caldera walls filled up some 10 m of the crater floor, forming the funnel-shaped caldera. 6. Summary (1) The September, November and December 2008 eruptive activity inside of the 340 m deep Dolomieu crater was inaccessible to direct measurements and probing. Aerial photographs and photographs from the Dolomieu crater rim, as well as determinations of the
Fig. 8. Histogram of intermediate TADR of the December 14 to February 4 eruptive period compared to the tremor intensity observed at the Bory seismic station.
T. Staudacher / Journal of Volcanology and Geothermal Research 191 (2010) 60–68
67
Fig. 9. View into the Dolomieu from the South-West. a) An image of the Dolomieu crater taken on April 11, 2007, some days after the collapse; b) an infrared image taken on May 5, 2008. The temperatures range between 95 °C (yellow colour) and 40 °C (red colour). Dark areas correspond to temperatures between 1 and 5 °C. c) The Dolomieu crater on February 18, 2009, after the 2008 eruptions; d) a superposition of an infrared and a visible image from March 30, 2009 (Coppola, pers. com.).
elevation of the lava flow surface of the three different eruptive phases in the Dolomieu crater, allowed for mapping the inundated surfaces and the determination of the corresponding emitted
volumes. During the third eruptive phase, estimations of intermediate volumes and TADR were obtained, the pattern of which match quite well with the eruption tremor. The average lava outflow of the three eruptive phases range between 0.3 and 1 m3 s− 1. They agree with normal eruption rates of typical summit eruptions at Piton de la Fournaise as observed before for five eruptions between 1999 and 2004 (Coppola et al., 2005; Staudacher et al., 2008; Coppola et al., 2009). (2) Numerous photographs of the lava flow allowed the observation of an uprising lava flow surface floating on a liquid lava “lake” during the December phase. (3) Highest magma temperatures of up to 1150 °C were measured by an infrared camera in September 2008. (4) The site of the eruptive vents and infrared observations of hydrothermal emanations in the Dolomieu caldera wall since its formation suggest that the collapsed rock column in April 2007 had a size of about 690 × 500 m, thus covering only part of the dimension of the Dolomieu crater. Acknowledgments
Fig. 10. Cross section of Piton de la Fournaise, indicating the size and depth of the collapsed column between the summit and the magma chamber.
We are grateful to P. Boissier and F. Massin for their help of taking part of the infrared images and P. Boissier for the daily GPS evaluation and to the technical team for the maintenance of the different networks. A. Peltier, D. Coppola and a reviewer are thanked for improvement of the manuscript. Sassan Hafizi and Philipp St. from the University of Portsmouth are thanked for the corrections of the English. This is IPGP contribution No. 2588.
68
T. Staudacher / Journal of Volcanology and Geothermal Research 191 (2010) 60–68
References Battaglia, J., Aki, K., Staudacher, T., 2005. Location of tremor sources and estimation of lava output using tremor source amplitude on the Piton de la Fournaise volcano: 2. Estimation of lava output. J. Volcanol. Geotherm. Res. 147 (2005), 291–308. Backström, K., Gudmundsson, A., 1989. The grabens of Sveinar and Sveinagja, NE Iceland. Nordic Volcanol. Inst. Prof. Pap. 8901 38 pp. Coppola, D., Staudacher, T., Cigolini, C., 2005. The May–July 2003 eruption at Piton de la Fournaise (La Réunion): volume, effusion rates, and emplacement mechanisms inferred from thermal imaging and Global Positioning System (GPS) survey. In: Manga, M., Ventura, G. (Eds.), Kinematics and Dynamics of Lava Flows: Geological Society of America Special Paper, vol. 396, pp. 103–124. Coppola, D., Staudacher, T., Cigolini, C., 2007. Field thermal monitoring during the August 2003 eruption at Piton de la Fournaise (La Réunion). J. Geophys. Res. 112, B05215. doi:10.1029/2006JB004659. Coppola, D., Piscopo, D., Staudacher, Th., Cigolini, C., 2009. Lava discharge rate and effusive pattern at Piton de la Fournaise from MODIS data. J. Volcanol. Geoth. Res. 184, 174–192. doi:10.1016/j.jvolgeores.2008.11.031. Di Muro A., Garofalo K., Staudacher Th., Villemant B., in preparation. Garofalo K., Staudacher Th., Ferrazzini V., Kowalski P., Boissier P., Dupont A., Peltier A., Villemant B., Boudon G., (2009), Eruptive SO2-plume measurements at Piton de la Fournaise (Ile de la Réunion) by stationary NOVAC scanning MAX-DOAS instruments, EGU General Assembly 2009, Vienna. Harris, A.J.L., Dehn, J., Calvari, S., 2007. Lava effusion rate definition and measurement: a review. Bull. Volcanol. 70, 1–22. Michon, L., Staudacher, Th., Ferrazzini, V., Bachelery, P., Marti, J., 2007. April 2007 collapse of Piton de la Fournaise: a new example of caldera formation. Geophys. Res. Lett. 34 (L21301). doi:10.1029/2007GL031248. Opheim, J.A., Gudmundsson, A., 1989. Formation and geometry of fractures, and related volcanism, of the Krafla fissure swarm, Northeast Iceland. Geol. Soc. Am. Bull. 101, 1608–1622. OVPF/IPGP 2008, internal reports, http://www.ipgp.fr/pages/03030807.php.
Peltier, A., Staudacher, T., Bachelery, P., 2007. Constraints on magma transfers and structures involved in the 2003 activity at Piton de la Fournaise from displacement data. J. Geophys. Res. 112 (B03207). doi:10.1029/2006JB004379. Peltier, A., Famin, V., Bachèlery, P., Cayol, V., Fukushima, Y., Staudacher, Th., 2008. Cyclic magma storage and transfers at Piton de la Fournaise volcano (La Réunion hotspot) inferred from deformation and geochemical data. Earth Planet. Sci. Lett. 270 (3–4), 180–188. Peltier, A., Bachelery, P., Staudacher, Th., (2009a). Magma transport and storage at Piton de La Fournaise (La Réunion Island) between 1972 and 2007: a review of geophysical and geochemical data. J. Volcanol. Geotherm. Res., (2008), doi:10.1016/j.jvolgeores.2008.12.008 Peltier A., Staudacher Th., Bachèlery P., (submitted for publication), Inter-eruptive shallow magma migrations inferred in 2008 at Piton de La Fournaise (La Réunion) from GPS data, JVGR submitted. Sigmarsson, O., Condomines, M., Bachèlery, P., 2005. Magma residence time beneath the Piton de la Fournaise volcano, Reunion Island, from U-series disequilibria. Earth Planet. Sci. Lett. 234, 223–234. Staudacher, Th., Ruizé, L., Peltier, A., 2008. Historique des éruption du Piton de la Fournaise de 1998 à 2007. Graphica, St André, La Réunion, p. 98. Staudacher Th., Peltier A., (2010), Eruptive activity at Piton de la Fournaise from 1998 to 2008: description of eruptions, survey, forecast and monitoring of the volcano. Staudacher, Th., Ferrazzini, V., Peltier, A., Kowalski, Ph., Boissier, P., Catherine, Ph., Lauret, F., Massin, F., 2009. The April 2007 eruption and the Dolomieu crater collapse, two major events at Piton de la Fournaise (La Réunion Island, Indian Ocean). J. Volcanol. Geotherm. Res. 184 (1–2), 126–137. doi:10.1016/j.jvolgeores.2008.11.005. Urai, M., Geshi, N., Staudacher, Th., 2007. Size and volume evaluation of the caldera collapse on Piton de la Fournaise volcano during the April 2007 eruption using ASTER stereo imagery. Geophys. Res. Lett. 34 (L22318). doi:10.1029/2007GL031551. Vigouroux, N., Williams-Jones, A.E., Wallace, P., Staudacher, T., 2009. The November 2002 eruption of Piton de la Fournaise, Reunion: tracking the pre-eruptive thermal evolution of magma using melt inclusions. Bull. Volcanol.
Contents lists available at ScienceDirect
Journal of Volcanology and Geothermal Research j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j vo l g e o r e s
Field observations of the 2008 summit eruption at Piton de la Fournaise (Ile de La Réunion) and implications for the 2007 Dolomieu collapse Thomas Staudacher Observatoire Volcanologique du Piton de la Fournaise (OVPF), Institut de Physique du Globe de Paris, CNRS, UMR 7154 – Géologie des Systèmes Volcaniques, 14 RN3, le 27ème km, 97418, La Plaine des Cafres, La Réunion, France
a r t i c l e
i n f o
Article history: Received 22 July 2009 Accepted 8 January 2010 Available online 21 January 2010 Keywords: Piton de la Fournaise summit eruptions lava flow emplacement lava temperature dyke emplacement caldera ring fault
a b s t r a c t More than one year after the most important eruption ever observed at Piton de la Fournaise in April 2007, a new eruption started in 2008 at Piton de la Fournaise, with three different eruptive phases in September, November and December. They were located within the 340 m deep Dolomieu crater. Due to the steep (dip angle 45–80°) and unstable walls of the new crater formed in April 5, 2007, no measurements in situ of the 2008 eruption had been possible. Only observations from the Dolomieu crater rim and from helicopter were performed, using an infrared camera, a portable rangefinder and numerous photographs. These field observations allowed precise monitoring of the setting up of the lava flows with time. The total erupted volume of lava of the three phases was 2.2 × 106 m3 and the average flow rates ranged between 0.3 and 1 m3 s− 1. Lava temperatures of up to 1150 °C have been measured by an infrared camera. Overall, infrared images of the Dolomieu crater illustrate the control of the eruptive vents by the structure of the April 2007 Dolomieu collapse. © 2010 Elsevier B.V. All rights reserved.
1. Introduction Piton de la Fournaise is the active volcano in the South-East of Ile de la Réunion in the western Indian Ocean. It is located within the 8 km wide and 13 km long U-shaped Enclos Fouqué caldera, which is open to the sea (Fig. 1). Two craters crown the summit, the about 250 m wide Bory crater and the 1 km wide Dolomieu crater. Piton de la Fournaise produced over the last century about 150 eruptions and emitted more than 1000x 106m3 of lava (OVPF compilation). The last decade was particularly active with 27 eruptions producing 367 x 106m3, about 1/3 of the estimated volume emitted during the last century (Peltier et al., 2009; Staudacher and Peltier, 2010). Amongst this eruptive sequence, the April 2–May 1, 2007 eruption represents a major event. It was with >140 x 106m3 of lava flows the most voluminous and intense eruption for at least two centuries (Staudacher et al., 2009; Peltier et al., 2009). This voluminous eruption, which took place at a low elevation of 590 m in the Grand Brûlé area (Fig. 1), was the cause of the simultaneous Dolomieu caldera collapse. The rock column between the Dolomieu crater and the magma chamber collapsed into the latter on April 5 and 6. In 24 hours, a 1 km long, 800 m large and 340 m deep funnel-shaped caldera was formed (Urai et al., 2007; Michon et al., 2007; Staudacher et al., 2009). Even though the April 2007 eruption emptied between 30 and 40% of the shallow magma reservoir (Sigmarsson et al., 2005; Peltier et al., 2007, 2008) Piton de la Fournaise erupted again in 2008 within the
E-mail address: [email protected]. 0377-0273/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.jvolgeores.2010.01.012
Dolomieu crater. Three eruptive phases formed a 420 × 220 m large and about 75 m deep lava flow on the crater floor.
2. Overview of the 2008 summit eruption 2.1. September 21 eruptive phase Following the April 2007 eruption and the collapse of the central Dolomieu crater the whole Piton de la Fournaise massif was destabilized and needed to find a new state of structural stability. One of the consequences was the structural readjustment of the Dolomieu crater; its dimensions reduced about 3 m and 1.2 m in north–south in east–west directions respectively, and a subsidence of 40 to 99 cm of its border was recorded by the permanent GPS network (Staudacher et al., 2009). This subsidence lasted from April 2007 to July 2008. Between March and July 2008, a normal background seismicity of less than 10 volcano-tectonic seismic events per day was observed beneath the summit (OVPF internal report, 2008). On July 28 the general trend of subsidence reversed. The permanent GPS network recorded a new inflation of the summit cone (Fig. 2), with a crater elongation of about 5.6 cm on a north–south axis and 3.3 cm on a west–east axis. No significant tiltmeter or extensometer variations were recorded. The inflation was accompanied by an increase of seismicity beneath the summit and by repeated, 30 min to 3 h lasting seismic crises on August 4, 15 and 31 with 46, 471 and 241 events, and on September 8, 9, and 15 with 104, 514 and 296 events, respectively. Magnitudes of the seismic events ranged between −0.3
T. Staudacher / Journal of Volcanology and Geothermal Research 191 (2010) 60–68
61
Fig. 1. Map of Piton de la Fournaise volcano inside of the Enclos Fouqué caldera. Shown are the Bory and Dolomieu summit craters and the summit permanent GPS stations BONG (Bory north), SNEG (Soufrière north-east), DERG (Dolomieu east), DSRG (Dolomieu south) and BORG (Bory), as well as the Tunnel Cathérine tiltmeter station. Other tiltmeter stations are close to the corresponding GPS stations. The April to May 2007 lava flow is shown in dark grey in the Grand Brûlé and the 2008 lava flow is shown in black within the Dolomieu crater. Coordinates are Gauss Laborde in meters.
and 2.6, but most were smaller than Md 1 (OVPF internal report, 2008). The first appearance of new activity was a short period of volcanic tremor on September 12, between 6:00 and 16:00 (GMT). No magma arrived at the surface, but simultaneous release of gas from the southwestern Dolomieu crater wall at about 2350 m elevation was observed. SO2 was detected by the 3 km distant NOVAC network on the Enclos Fouqué caldera (Garofalo et al. 2009). On September 21, at 11:05, a 30 min seismic crisis preceded a new eruption in the Dolomieu crater. Only insignificant GPS horizontal and vertical variations of <1 cm accompanied this seismic crisis. The eruptive vent was located on the western crater wall about midway between the floor and the top of the crater at 2340 to 2380 m elevation (Gauss Laborde coordinates [m]: 179080/37260–179110/ 37200). After a peak of intensity on the first day of the eruption, the tremor decreased strongly and reached a somewhat constant value (Fig. 3a). A higher and irregular eruption tremor was recorded on
September 25. The eruption ceased on October 2, with the disappearance of the eruption tremor at 01:30. The lava flow, emitted during the nine days of eruption accumulated on the Dolomieu crater floor (Fig. 4a) and stretches to about 300 × 160 m (Fig. 5a). The elevation of the surface was measured from the Dolomieu crater border with a TruPulse 200 laser rangefinder to 2205 m above sea level (Fig. 6b). These measurements had been made from different sites around the Dolomieu crater on bright rocks lying on the lava flow or on gravel adjacent to the lava flow border. Successive measurements allowed us to estimate an error of ±3 m. Based on the depth of 340 m of the Dolomieu crater measured after the April 2007 collapse (Staudacher et al., 2009), this technique allowed us to estimate the maximum thickness of the lava flow to about 50 m. The emitted volume was estimated between 0.8 and 1.0 × 106 m3 and the average flux was 0.9 m3 s− 1 (Section 4). As the lava flow was inaccessible for direct sampling, Pele's hairs have been collected for geochemical analysis (Di Muro et al., in
Fig. 2. Curves of four permanent summit GPS stations, showing pre, co- and post eruptive ground variations between January 2008 and April 2009. Data points are daily averaged values and show an uncertainty of about ±0.5 cm. Eruptions are shown in red straight lines and yellow background. Easting is represented in blue, northing in red. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
62
T. Staudacher / Journal of Volcanology and Geothermal Research 191 (2010) 60–68
Fig. 3. Eruption tremor recorded by the Bory seismic station of September 21 (a), November 27 (b) and December 14 (c) eruptive phases.
preparation) on a 1 m2 fabric placed previously on the Bory crater floor. Lava temperatures up to 1150 °C were measured by a FLIR ThermaCAM PM695 from the Dolomieu and the Bory crater rims (Section 3; Fig. 4b and Table 1). 2.2. November 27 eruptive phase During October and November, the seismicity remained high with 20 to 100 events per day, amongst which an event of Md 3.3 occurred on November 12, 21 h37. Several seismic crisis occurred on October 20 and 31 with 191 and 226 events, respectively (OVPF internal report, 2008). The latter was recorded on the tiltmeter network with significant variations at La Soufrière station (radial component + 22 μrad; tangential component +5 μrad), at Bory station (radial component − 5 μrad; tangential component −6 μrad) and Tunnel Catherine station (radial component − 9.5 μrad; tangential component −3.5 μrad). Station locations are shown in Fig. 1. Two further crises happened on November 6 and 21 (114 and 239 events respectively). The permanent GPS network recorded a Dolomieu crater opening between November 5 and 25 of 2.6 cm and 2.2 cm on a
north–south and west–east trending axis, respectively (Fig. 2). On November 27, a new seismic crisis started at 07:25, simultaneously with a north–south extension of the Dolomieu crater of 7 cm and a west–east extension of 3.5 cm, recorded by the permanent GPS stations (Fig. 2). At 07:49 an eruption tremor appeared (Fig. 3b). The eruptive vent was adjacent to the September vent (Fig. 5b). A Ushaped lava flow partly covered the lava flow emitted in September and October on the Dolomieu crater floor (Figs. 5b and 6c). The eruptive phase stopped next day, after only 26 hours of activity. Less than 0.1 × 106 m3 of lava were emitted at an average flux of <1 m3 s− 1. As for the September eruptive phase, Pele's hairs have been collected on a fabric exposed in the Bory crater. 2.3. December 14 eruptive phase Immediately after the November eruptive phase, seismicity increased again with up to 40 low energy events (most of which were smaller than Md < 1, (OVPF internal report, 2008)) per day. A slight summit inflation was also recorded by the radial component of the Soufrière tiltmeter (87 μrad). The permanent GPS network continued to
Fig. 4. Visible and infrared images of the lava flow in the Dolomieu crater, recorded on September 22 from the northern Dolomieu boarder, close to “La Soufrière”. Temperatures are given in [°C].
T. Staudacher / Journal of Volcanology and Geothermal Research 191 (2010) 60–68
63
Fig. 5. Mapping of the three different eruptive phases. a) First phase: September 21 to October 2 lava flow in red. The dashed line indicates the limit of the lava flow on the crater floor and the crater wall; b) second phase: November 27 and 28 lava flow in yellow; c) and d) third phase: December 14 to 19 and December 24 to February 2 in different shades of blue, the darkest one being the most recent lava flow extension. Black lines display the eruptive vents. The sites where the infrared images were taken are shown in Fig. 5a; B = Bory; S = La Soufrière. The dotted oval curve denotes the proposed ring fault of the caldera. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
record summit inflation (Fig. 2). Between November 28 and December 14, a Dolomieu crater elongation of 5.1 and 4.8 cm on the north–south and west–east axis, respectively, was recorded. Compared to August– November, the inflation center shifted from the west to the east of the Dolomieu (Peltier et al., submitted for publication). On December 14, at 04:10, a 19 h-long lasting seismic crisis started, with a total of 1006 earthquakes. Such a long seismic crisis preceding a summit eruption is unusual for the Piton de la Fournaise volcano. It was accompanied by insignificant surface deformations (<1 cm) recorded by the permanent GPS network. At 23:00, an eruption tremor appeared beneath the summit and announced a third eruptive phase. Two vents opened in the Dolomieu crater at about 2340 m elevation (Figs. 5c and 7b). The main vent was located at 179550/37610 [m] (Gauss Laborde coordinates) on the northern crater wall; a secondary vent was located at 179750/37430 [m] on the north-eastern crater wall. This second vent was active for a few days and produced only an insignificant amount of lava. It is noticeable that the elevation of the eruptive vents is identical to the elevation of the vents of the September and November eruptive phases. The intensity of eruption tremor was quite irregular and increased until December 24. It stayed at a high level until January 1, 2009 (Fig. 3c). At 10:00 on the same day the tremor started to decrease and disappeared completely at 17:00. It reappeared on January 2 at 06:10 and continued at a low and very stable intensity until February 2, before disappearing on February 4. September and November lava flows on the Dolomieu crater floor were first flooded from the east by aa-type lava. In addition, from December 17 gas poor lava around the former lava flow started to appear, forming pahoehoe-type lava (Fig. 5c). Two weeks after the beginning of the eruptive phase, most of the lava flows emitted from the northern vent flowed down through channels beneath the surface of the Dolomieu crater floor. Resurgences of gas poor lava from below formed large pahoehoe plateaus on the boarder of the lava flow (hot area in Fig. 7a and
grey area in Fig. 7b). The lava flow and its elevation grew up continuously (Fig. 6d–g); the final elevation of the surface was determined after the end of eruption by use of the TruPulse 200 rangefinder to 2230 m and by the end of the December to February eruptive event, the surface has raised up for a further 25 m (Fig. 6g). Frequent pictures taken from the Dolomieu crater rim revealed that older lava flow surface emitted during the September eruptive phase was still visible in the western part (area inside of the ellipses in Fig. 6b–g). The total volume of the lava flows emitted during the September 2008 to February 2009 was estimated to 2.2 × 106 m3 (see Section 4). The December eruptive phase produced therefore 1.3 × 106 m3, with an average flow rate of 0.3 m3 s− 1. Again we determined the temperature of the lava flows with a FLIR ThermaCAM PM695 camera from Bory crater and from the Dolomieu border just above the active vent. Only very few Pele's hairs for geochemical analysis were found on the surface close to the former “La Soufrière” site just above the eruptive vent and on a new fabric in the Bory crater. 3. Lava temperature measurements During the September and December 2008 eruptive phases, infrared images had been taken using a FLIR ThermaCAM™ PM695. The temperatures were determined by the ThermaCAM Researcher Pro 2.9 program. The infrared images of the September eruptive phase were taken always in two ranges (0–500 °C and 350–1500 °C) from the Bory crater just above the September eruptive vent (B in Fig. 5a) and from the Dolomieu crater rim, close to La Soufrière (S in Fig. 5a) at slope distances from the vent of 385 m and 680 m respectively. For the lava flow itself (Figs. 4b and 7a) slope distances were estimated for each individual point. The images of the December eruptive phase were taken from the same sites and the slope distances to the eruption vent ranged between 400 and 935 m respectively. In order to determine temperatures, we
Fig. 6. Sequence of the Dolomieu crater filling between November 20, 2008 (b) and March 23, 2009 (g). Photographs taken from the Dolomieu crater rim show the Dolomieu crater before the 2008 eruptions (a). Numbers indicate the elevation of the Dolomieu floor and marker points on the south crater wall as well as the elevation of the lava flow surface (b–g). Ellipses mark the area visible since the end of the September lava flow. (h) A sketch of the filling process showing consecutive events, pink: September lava flow, orange: early December lava flow, red: late December lava flow.
64 T. Staudacher / Journal of Volcanology and Geothermal Research 191 (2010) 60–68
T. Staudacher / Journal of Volcanology and Geothermal Research 191 (2010) 60–68 Table 1 Temperature evaluations for the September and December eruptive phases. Date
Time
Observation site
Distance (m)
Temperature (°C)
± (°C)
22/09/08 22/09/08 22/09/08 22/09/08 22/09/08 22/09/08 15/12/08 15/12/08 19/12/08 19/12/08
09:23 09:23 09:23 12:48 12:49 12:49 10:53 11:28 10:24 11:00
Bory — Sept 08 vent Bory — Sept 08 vent Bory — Sept 08 vent Soufrière — Sept 08 vent Soufrière — Sept 08 vent Soufrière — Sept 08 vent Bory — Dec 08 vent Soufrière — Dec 08 vent Soufrière — Dec 08 vent Bory — Dec 08 vent
385 385 385 695 695 695 935 430 400 935
1140 1100 1120 1125 1145 1150 930 900 930 1050
50 65 60 60 70 70 40 50 30 50
corrected for distance, relative humidity, emissivity and air temperature. Air temperatures were 10± 5 °C in September and 15± 5 °C in December; emissivity was 0.95 ± 0.05 and relative humidity was 70± 10% for both eruptions. In September the temperatures of six measurements ranged between 1100 and 1150 °C (Table 1). Such temperatures are typical for a fresh magma emitted by Piton de la Fournaise eruptions (Coppola et al., 2007; Vigouroux et al., 2009). They are in good agreement with temperatures of 1150–1190 °C determined by a plagioclase-glass equilibrium study on the Pele's hairs emitted from all the three 2008 eruption phases (Di Muro, personal communication). Temperatures of the December eruptive phase are significantly lower, ranging between 900 and 1050 °C. The later represent four measurements taken on December 15 and 19 (Table 1). Such low temperatures might be due to an older and colder magma batch, but could also be an artifact due to unsuitable conditions during recording. In particular they are not in agreement with temperature determination on the Pele's hairs from this eruptive phase (see above). In fact the December vent had the form of a hornito and direct view into the hot magma from above was almost impossible. In addition volcanic gases were much more abundant in December. Such conditions might have prevented the measurements of reliable temperatures and therefore they should only be taken as lower limits. 4. Cartography and volume estimations The high frequency of observations allowed us to follow the extrusion of the lava flow through time. Mapping of the lateral extent of lava flows had been performed by using pictures taken from helicopter at high altitude and observations from the Dolomieu crater
65
rim, as well as from Aster images after the end of the eruptive phases (M. Urai, pers. information). The final extension of the September lava flow was determined to be 300 × 160 m (Fig. 5a), with a total surface on the Dolomieu crater floor of 0.033 × 106 m2. The extension of the December lava flow was determined as 420 × 220 m (Fig. 5d), with a total surface on the Dolomieu crater floor of 0.092 × 106 m2. The estimation of the volumes of the different lava flows was performed using the mapped lava flow surfaces and their corresponding depths. The latter were based on changes in height of the lava flow surface relative to marker points (see Fig. 6). The horizontalness of the lava flow was checked by measurements of three other marker points around the lava flow. Fig. 6a shows the Dolomieu caldera on October 12, 2007. The caldera floor was located at 2155 m elevation, 340 m below the eastern boarder (not visible here) and 375 m below the southern rim of Dolomieu. Two striking points, a large bolder and the sharp lower end of a scree at 2214 m and 2246 m respectively are indicated, as well as the southern Dolomieu crater rim at 2530 m. After the September eruptive phase, the crater floor has risen up about 50 m to an elevation of 2205 m (Fig. 6b). The November eruptive phase (Fig. 6c) only partly over-topped the September lava flow, with a thickness of about 5 m on the southern part. In particular the area within the ellipse was not covered by the U-shaped lava flow (Fig. 6c). On December 19 (Fig. 6d) the lava flow reached 2212 m elevation and five days later it attained the bolder at 2214 m elevation (Fig. 6e). On January 20 (Fig. 6f) the surface has raised up to 2227 m elevation, while the area inside of the ellipse was still visible. This was still the case after the end of eruption (Fig. 6g). The lava flow surface is currently at 2230 m elevation and some of the “old” September lava flow surface is still present. New measurements in November 2009 directed on sulfur-rich deposits on the lava flow gave us an elevation of 2233 ± 2 m above sea level, in excellent agreement with the former value. The sketch in Fig. 6h shows a possible scenario for the evolution of lava emplacement in 2008 inside the Dolomieu crater. It seems that lava from the December eruptive phase undermined the September lava flow through lava tubes and emerged at the borders in the form of degassed lava, forming the pahoehoe-type lava plateaus (Fig. 7). During this process it uplifted the “old” surface which “floated” on a liquid reservoir. Of note is the lava flow surface marked by the ellipses in Fig. 6b–h that is visible from November 20, 2008 to March 23, 2009. From these observations and taking into account the visible lava flow surfaces we performed different estimation of the emitted volumes. We assumed that the pre-eruptive Dolomieu crater bottom looked like a) an
Fig. 7. Infrared (a) and visible (b) image of the lava flow on the Dolomieu crater floor, on January 23, 2009, seen from the Bory crater. In order to enhance temperature differences in the much lower temperature range, we used a different colour code than in Fig. 4.
66
T. Staudacher / Journal of Volcanology and Geothermal Research 191 (2010) 60–68
Table 2 Estimation of emitted volumes and TADR of the December 2008 to February 2009 eruptions for different time intervals. Date
Time interval (days)
Emitted volume (106 m3)
TADR (m3 s− 1)
15/12/08 16/12/08 17/12/08 19/12/08 24/12/08 28/12/08 04/02/09 Total
0.42 1.60 0.73 2.25 5.08 3.96 37.5 51.5
0.05 0.01 0.05 0.06 0.32 0.52 0.48 1.49
1.41 0.11 0.81 0.33 0.72 1.51 0.16 0.33
inverted elliptical cone or b) an axial conoïd. Finally we estimated the emitted volumes of the December lava flows at different dates. From a) and b) we derived volumes of 0.9 ± 0.1 × 106 m3 for September eruptive phase and 2.2 ± 0.3 × 106 m3 for the September+ November + December eruptive phases. Consequently the December eruptive phase produced 1.3 ± 0.3 × 106 m3. The volume emitted by the November eruptive phase was small, estimated at ≤0.1× 106 m3. Intermediate volumes of the December eruptive phase had been estimated between December 15 and the end of eruption (Table 2). We used the estimated lava flow surface and the variation of the elevation of the uprising lava flow between these dates (see Fig. 6 and Table 2). The intermediate TADR (Time Averaged Discharge Rate (Harris et al. 2007)) ranged between 0.1 and 1.5 m3 s− 1 (Table 2, Fig. 8). A high value was determined for the first hours of the eruption, as had been often observed in the past (Coppola et al., 2005; Staudacher et al., 2009). The eruption rate slowed down to 0.11 m3 s− 1 on the second day and increased until the end of December, simultaneously with the increasing tremor intensity (Fig. 8). Between December and early February, the low eruption rate of 0.16 m3 s− 1 accompanied the episode with low seismic tremor (Fig. 8), in agreement with observations by Battaglia et al. (2005). The resulting total volume of 1.5 × 106 m3 (Table 2) estimated in this way agrees with the value of 1.3 × 106 m3 estimated above. 5. Influence of the Dolomieu caldera on the 2008 eruptive activity In April 2007, a major eruption occurred in the Grand Brûlé, about 8 km away from Piton de la Fournaise at low elevation (Staudacher et al., 2009). The lateral withdraw of magma lowered the pressure in the shallow magma chamber. One consequence of that pressure drop was the rupture of the roof of the magma chamber and the collapse of the overlying rock column, forming an about 1 km large and 340 m deep
slightly elongated caldera in the Dolomieu crater (Fig. 9a) (Urai et al. 2007; Michon et al., 2007; Staudacher et al., 2009). The size and the effective depth of the collapse were however not well known. Field observations immediately after the collapse in 2007 showed that a narrow band at about 2350 m elevation on the caldera wall was marked by vapor emanations and wet rock almost all around the Dolomieu crater (Fig. 9a). This zone was also characterized by elevated temperatures of up to 200 °C. High temperatures between 50 and 100 °C were still observed since then (Fig. 9b and d). These vapor emanations persist until now (Fig. 9c) and seem to correspond with the arrival of hydrothermal fluids, eventually through the ring fault formed by the 2007 collapse between the crumbled rock column and the bed rock, as already suggested by Michon et al. (2007) and Staudacher et al. (2009). The feeding dykes of the 2008 eruptive phases emerged at the same elevation (2340 m) as did the hydrothermal emanations. It seems that they followed the same pathway and that the ring fault acted as a channel for the dyke migration from the magma chamber to the surface as proposed by Opheim and Gudmundsson (1989) and Backström and Gudmundsson (1989). Fig. 10 suggests such a scenario, even though we do not know whether the 2008 eruptions were fed directly from the magma chamber along the ring fault or from residual magma stored in the volcanic massif or within the collapsed column. These observations might give new insights into the structure of the 2007 Dolomieu collapse, in particular the size of the collapsed “piston” (Fig. 10). If the elevation of the emanations and the 2008 magma correspond in fact to the limit between the collapsed rock column and the bedrock, then the size of the “piston” which slumped into the shallow magma chamber had a size of only about 690 × 500 m and does not cover the whole Dolomieu crater. Taking into account a volume of the caldera of 96× 106 m3 (Urai et al., 2007) to 90× 106 m3 (Staudacher et al., 2009) we can estimate an initial depth of the caldera of about 350 m. This depth is in excellent agreement with the depth of 340 m determined in May and June 2007, shortly after the collapse (Staudacher et al., 2009) and confirms the fact that no large breakdowns followed the April 5 to 6 collapse and that the final depth of the caldera was reached almost immediately. After the main collapse material from the Dolomieu crater walls slid down and only minor landslides of the caldera walls filled up some 10 m of the crater floor, forming the funnel-shaped caldera. 6. Summary (1) The September, November and December 2008 eruptive activity inside of the 340 m deep Dolomieu crater was inaccessible to direct measurements and probing. Aerial photographs and photographs from the Dolomieu crater rim, as well as determinations of the
Fig. 8. Histogram of intermediate TADR of the December 14 to February 4 eruptive period compared to the tremor intensity observed at the Bory seismic station.
T. Staudacher / Journal of Volcanology and Geothermal Research 191 (2010) 60–68
67
Fig. 9. View into the Dolomieu from the South-West. a) An image of the Dolomieu crater taken on April 11, 2007, some days after the collapse; b) an infrared image taken on May 5, 2008. The temperatures range between 95 °C (yellow colour) and 40 °C (red colour). Dark areas correspond to temperatures between 1 and 5 °C. c) The Dolomieu crater on February 18, 2009, after the 2008 eruptions; d) a superposition of an infrared and a visible image from March 30, 2009 (Coppola, pers. com.).
elevation of the lava flow surface of the three different eruptive phases in the Dolomieu crater, allowed for mapping the inundated surfaces and the determination of the corresponding emitted
volumes. During the third eruptive phase, estimations of intermediate volumes and TADR were obtained, the pattern of which match quite well with the eruption tremor. The average lava outflow of the three eruptive phases range between 0.3 and 1 m3 s− 1. They agree with normal eruption rates of typical summit eruptions at Piton de la Fournaise as observed before for five eruptions between 1999 and 2004 (Coppola et al., 2005; Staudacher et al., 2008; Coppola et al., 2009). (2) Numerous photographs of the lava flow allowed the observation of an uprising lava flow surface floating on a liquid lava “lake” during the December phase. (3) Highest magma temperatures of up to 1150 °C were measured by an infrared camera in September 2008. (4) The site of the eruptive vents and infrared observations of hydrothermal emanations in the Dolomieu caldera wall since its formation suggest that the collapsed rock column in April 2007 had a size of about 690 × 500 m, thus covering only part of the dimension of the Dolomieu crater. Acknowledgments
Fig. 10. Cross section of Piton de la Fournaise, indicating the size and depth of the collapsed column between the summit and the magma chamber.
We are grateful to P. Boissier and F. Massin for their help of taking part of the infrared images and P. Boissier for the daily GPS evaluation and to the technical team for the maintenance of the different networks. A. Peltier, D. Coppola and a reviewer are thanked for improvement of the manuscript. Sassan Hafizi and Philipp St. from the University of Portsmouth are thanked for the corrections of the English. This is IPGP contribution No. 2588.
68
T. Staudacher / Journal of Volcanology and Geothermal Research 191 (2010) 60–68
References Battaglia, J., Aki, K., Staudacher, T., 2005. Location of tremor sources and estimation of lava output using tremor source amplitude on the Piton de la Fournaise volcano: 2. Estimation of lava output. J. Volcanol. Geotherm. Res. 147 (2005), 291–308. Backström, K., Gudmundsson, A., 1989. The grabens of Sveinar and Sveinagja, NE Iceland. Nordic Volcanol. Inst. Prof. Pap. 8901 38 pp. Coppola, D., Staudacher, T., Cigolini, C., 2005. The May–July 2003 eruption at Piton de la Fournaise (La Réunion): volume, effusion rates, and emplacement mechanisms inferred from thermal imaging and Global Positioning System (GPS) survey. In: Manga, M., Ventura, G. (Eds.), Kinematics and Dynamics of Lava Flows: Geological Society of America Special Paper, vol. 396, pp. 103–124. Coppola, D., Staudacher, T., Cigolini, C., 2007. Field thermal monitoring during the August 2003 eruption at Piton de la Fournaise (La Réunion). J. Geophys. Res. 112, B05215. doi:10.1029/2006JB004659. Coppola, D., Piscopo, D., Staudacher, Th., Cigolini, C., 2009. Lava discharge rate and effusive pattern at Piton de la Fournaise from MODIS data. J. Volcanol. Geoth. Res. 184, 174–192. doi:10.1016/j.jvolgeores.2008.11.031. Di Muro A., Garofalo K., Staudacher Th., Villemant B., in preparation. Garofalo K., Staudacher Th., Ferrazzini V., Kowalski P., Boissier P., Dupont A., Peltier A., Villemant B., Boudon G., (2009), Eruptive SO2-plume measurements at Piton de la Fournaise (Ile de la Réunion) by stationary NOVAC scanning MAX-DOAS instruments, EGU General Assembly 2009, Vienna. Harris, A.J.L., Dehn, J., Calvari, S., 2007. Lava effusion rate definition and measurement: a review. Bull. Volcanol. 70, 1–22. Michon, L., Staudacher, Th., Ferrazzini, V., Bachelery, P., Marti, J., 2007. April 2007 collapse of Piton de la Fournaise: a new example of caldera formation. Geophys. Res. Lett. 34 (L21301). doi:10.1029/2007GL031248. Opheim, J.A., Gudmundsson, A., 1989. Formation and geometry of fractures, and related volcanism, of the Krafla fissure swarm, Northeast Iceland. Geol. Soc. Am. Bull. 101, 1608–1622. OVPF/IPGP 2008, internal reports, http://www.ipgp.fr/pages/03030807.php.
Peltier, A., Staudacher, T., Bachelery, P., 2007. Constraints on magma transfers and structures involved in the 2003 activity at Piton de la Fournaise from displacement data. J. Geophys. Res. 112 (B03207). doi:10.1029/2006JB004379. Peltier, A., Famin, V., Bachèlery, P., Cayol, V., Fukushima, Y., Staudacher, Th., 2008. Cyclic magma storage and transfers at Piton de la Fournaise volcano (La Réunion hotspot) inferred from deformation and geochemical data. Earth Planet. Sci. Lett. 270 (3–4), 180–188. Peltier, A., Bachelery, P., Staudacher, Th., (2009a). Magma transport and storage at Piton de La Fournaise (La Réunion Island) between 1972 and 2007: a review of geophysical and geochemical data. J. Volcanol. Geotherm. Res., (2008), doi:10.1016/j.jvolgeores.2008.12.008 Peltier A., Staudacher Th., Bachèlery P., (submitted for publication), Inter-eruptive shallow magma migrations inferred in 2008 at Piton de La Fournaise (La Réunion) from GPS data, JVGR submitted. Sigmarsson, O., Condomines, M., Bachèlery, P., 2005. Magma residence time beneath the Piton de la Fournaise volcano, Reunion Island, from U-series disequilibria. Earth Planet. Sci. Lett. 234, 223–234. Staudacher, Th., Ruizé, L., Peltier, A., 2008. Historique des éruption du Piton de la Fournaise de 1998 à 2007. Graphica, St André, La Réunion, p. 98. Staudacher Th., Peltier A., (2010), Eruptive activity at Piton de la Fournaise from 1998 to 2008: description of eruptions, survey, forecast and monitoring of the volcano. Staudacher, Th., Ferrazzini, V., Peltier, A., Kowalski, Ph., Boissier, P., Catherine, Ph., Lauret, F., Massin, F., 2009. The April 2007 eruption and the Dolomieu crater collapse, two major events at Piton de la Fournaise (La Réunion Island, Indian Ocean). J. Volcanol. Geotherm. Res. 184 (1–2), 126–137. doi:10.1016/j.jvolgeores.2008.11.005. Urai, M., Geshi, N., Staudacher, Th., 2007. Size and volume evaluation of the caldera collapse on Piton de la Fournaise volcano during the April 2007 eruption using ASTER stereo imagery. Geophys. Res. Lett. 34 (L22318). doi:10.1029/2007GL031551. Vigouroux, N., Williams-Jones, A.E., Wallace, P., Staudacher, T., 2009. The November 2002 eruption of Piton de la Fournaise, Reunion: tracking the pre-eruptive thermal evolution of magma using melt inclusions. Bull. Volcanol.