Abstract
A large lava flow field, essentially constituted of "pahoehoe" flows, forms a relatively monotonous area on the north-west flank of the terminal cone of Piton de la Fournaise and on a large part of the Enclos floor. We named this unit "champ de lave de l'Enclos Fouque" (CLEF). To the east, more recent lava flows almost completely cover the CLEF. Mapping of the CLEF has been attempted using satellite imagery. Radar (SIR-C) and SPOT images have been used. Both types of images lead to comparable results for the surface of the CLEF, estimated to 11 km 2 . On satellite images, only a general map of the facies of the CLEF can be done. Air photos and ground observations allow to characterize more precisely the facies defined from satellite images. On the slopes of the Central Cone, the surface of the CLEF is mostly composed of tumuli that may produce small lava flows at their downhill base. On the floor of the Enclos, the CLEF is constantly formed by a multitude of intricate small flows. Except for small "aa" patches, surfaces are "pahoehoe". Tumuli, pressure ridges and collapsed plates are common features of the CLEF. Near the NW border of Enclos, the surface becomes less regular, with a higher proportion of "aa" patches and tilted slabs. This is probably due to the increase of slope in this area. In the vicinity of Bory crater contemporaneous welded scoria cover a zone extending 150-200 m from the crater. Several elements, found in various historical documents collected by Lacroix [1936 and 1938], bring evidence that the CLEF may result from a succession of events between the years 1750's to the 1790's. During this period, the volcano has been virtually continuously in eruption. Bory crater was the only active crater until 1766, when a new summit crater was formed by collapse, 400 m east of Bory crater. In the following years, this 1766 crater was filled by lava emissions which erected a gigantic tumulus, 50 m high, called Mamelon central. Several large lava flows, some of them reaching the sea, are unambiguously described as emitted from the summit craters. One of the main event appears to be the phase that occurred in 1753 (or 1759 according to different authors). This phase was associated to felt earthquakes and widespread ash-falls in the island. Lava flows covered most the western part of Enclos and reached the sea to the east. Whereas the activity of 1753 (1759 ?) appears as paroxysmal, extensive lava flows are described until 1794. In 1791, a new crater formed south of the 1766 crater. In 1801, Bory de St-Vincent [1804] observed an active lava lake inside this 1791 crater. In summary, it appears that quasi-continuous activity took place during the second half of the 18th century at Piton de la Fournaise, with lava flows covering almost all the Enclos and Grand Brule areas. It is obvious that the presently outcropping surface of the CLEF is only a fraction of its former extension. From the above-mentioned historic descriptions, it is reasonable to assume the CLEF may have recovered the entire surface of the Enclos Fouque and the Grand Brule depression. The surface extension of the CLEF would thus lie between 11 km 2 , its presently observed area, to ca. 90 km 2 . The compounded thickness of the CLEF has not been directly observed. In the rim of Dolomieu crater, a series of lava flows thought to represent the CLEF is about 25 m thick, but this thickness is that of the shield built around the vents and not that of the lava field. Using statistical laws for strombolian cones morphometry from Wood [1980], the thickness of the CLEF may be approached considering the crater diameter of 3 cones located in the north-western part of Enclos and partially flooded by the CLEF. We obtain estimates of 5 to 15 m. If Enclos has been entirely covered by the CLEF and for a mean thickness of 5 to 10 m, then the volume of the CLEF eruption is 450 to 900X10 6 m 3 . Mineralogical and chemical compositions of the CLEF lavas are similar to those of the transitional olivine basalts of the historic period [steady-state basalts, Albarede et al, 1997]. The chemical compositions vary somewhat due to variable modal olivine phenocrysts abundances and are characteristic of a rapid transfer of magma from deep reservoirs with virtually no low-pressure evolution. Major changes of the summit craters were associated to this eruption as a result of the unusually prolonged magmatic activity at Piton de la Fournaise during decades. The occurrence of similar phases at Piton de la Fournaise in the future can be appraised with reference to Kilauea. Holcomb [1987] reports 7 long-lived eruptions at Kilauea in the last 3 centuries. Thus, although not frequent, this type of event is not uncommon for Kilauea. By analogy, it must be considered as probable that Piton de la Fournaise will experience new long-lived eruptions in the future.
Evidence for a lava lake on Mt. Michael volcano, Saunders Island (South Sandwich Islands) from Landsat, Sentinel-2 and ASTER satellite imagery
