DISCUSION* (Deruelle y Brousse, 1984, Rev. Geol. , No. 22, p. 345)

"NUEE ARDENTE" DEPOSITS AT : A RE-INTERPRETATION

PETER w. FRANCIS Lunar and Planetary Institute, 3303 NASA Road 1, Houston, Texas 77058, USA

CARLOS F. RAMIREZ Servicio Nacional de Geología y Minería, Casilla 10465, Santiago, Chile.

ABSTRACT

The "nuée ardente" deposits at Tata Sabaya volcano are re-interpreted on morphological evidence as being those of a ¡arge volcanic debris avalanche, similar to that experienced by Mt. Sto Helens, May 1980. The prismat­ ic jointed blocks and deposits described by DerueIle and Brousse (1984) are clear evidence oí magmatic activity, but tbe stratigraphic relations between these deposits and those of the debris avalanche are not clear.

RESUMEN

En base a evidencias morfológicas, se interpretan los depósitos de "nube ardiente" del volcán Tata Sabaya (oeste de ), como depósitos de una gran avalancha volcánica de detritos, similar a la ocurrida en el volcán Saint Helens (noroeste de EE.UU.), en Mayo de 1980. Los bloques con diaclasamiento prismático y los depósitos de pómez descritos por Deruelle y Brousse (1984), son evidencias de actividad magmática, pero las relaciones es­ tratigráficas entre estos depósitos y aquéllos de la avalancha de detritos no son claras.

Deruelle and Brousse (1984) discussed the very May 1980. At 08:32 on the morning of May 18, interesting deposits of the Tata Sabaya volcano, the northero flank of the volcano collapsed, form­ Bolivia. While their contribution is very useful in ing a mejor debris avalan che with a volume of 2.8 3 drawing attention to what is cIearIy a remarkably km , and leaving a large "collapse " or am­ impressive deposit, we fee! that their interpreta­ phitheatre. Failure of the flank of the volcano ex­ tions are incomplete and misleading, particularly posed a "cryptodome" of hot dacitic in their emphasis on "nuée ardente" deposits. It is the core of the volcano, triggering a violent lateral­ matter of regret that DerueIle and Brousse provid­ Iy directed blast, which devastated an area of sorne 2 ed few details of the physical characteristics of this 500 km • Subsequently, a sustained plinian erup­ important deposit, and concentrated instead on tion column became established, causing ash fall petrological aspects. We have not yet had the ben­ out over a large area. Intennittent collapse of the efit of field studies at Tata Sabaya, but as the re­ eruption column caused the emplacement of a se­ sult of studies of volcano, north Chile, ries of pumiceous pyrocIastic flows () which experienced a similar type of eruptíon which accumulated to fonn a pum ice plain, over­ (Francis et al., 1985) and remote sensing study of Iying the debris avalanche deposits. These details Tata Sabaya using Landsat Thematic Mapper im­ have been fully documented in a publication of agery (Fig. 1), we fee! that we are in position to the United States Geological Survey (Hoblitt et al., offer a fuller interpretation. 1981). The key to understanding the Tata Sabaya de­ Although similar eruptions have been described posits lies in the eruption of Mount Sto Helens of

• Nota del Editor: por acuerdo el Comité Editor y haciendo uso de la posibilidad contemplada en nota de ContraportadilIa, se presenta aquí un comentario al artículo mencionado.

Revista Geológica de Cbile No. 24, p. 10 7· 110, 1985. 108 DISCUSION: "NUEE ARDENTE" TATA SABAYA

FIG.1. Landsat Thematic Mapper image (band 4) ofTata Sabaya vo\cano and its debris avalanche deposit, Bolivia. Large hummocks southeast of vo\cano are s1umped blocks. Amphitheatre left by collapse has been largely filled by youngcr , though traces of scarp are visible on east f1ank of the volcano.

previousJy (for example, the eruption of Shive­ those of a large debris avalanche deposit (Fig. 1). luch; Gorshkov and Dubik, 1970) the eruption Amongst the chief diagnostic features are the ex­ of Mt. Sto Helens focussed attention on this kind tensive tracts of hummocky topography with nu­ of eruption, which is now seen as a relatively com­ merous small hills and dosed depressions, longitu­ mon, almost normal, stage in the evolution of dinal and transverse ridges, and a systematic de­ a large volcano (Kerr, 1984), and a number of oth­ crease in hummock size away from the source vol­ er major volcanic debris avalanche deposits have eano (Siebert,1984; Ui,1983; Voight et al.,1981). been recognized: Mt. Shasta, USA (Crandell et al., The largest hummocks at Tata Sabaya are about 1984); Colima, Mexico (Luhr and Prestegaard, 100 m high. Taking an average thickness of 20 m 1985); Popocatepetl, Mexico (Robin and Boudal, gives a very conservative estimate for the volume 3 1984); and Socompa, Chile (Francis et al., 1985). of the whole deposit of 4 km . The amphitheater " Siebert (1984) has catalogues approximately 100 left by the collapse event has been largely obscur­ other examples in the geological literature. Sorne ed by post-collapse extrusion of flows (such as of these catastrophic collapse events appear to is currently under way.at Mount Sto Helens), but have also involved a violent bIast and magmatic parts of the bounding scarps can be seen in places, eruption (for example, Socompa), sorne involved particularly on the lower eastern flank oí the vol­ onIy phreatic eruptions (e.g. Bandai-San, 1888), cano. while others may have becn simply "cold" col­ Thus, we condude that Tata Sabaya represents Iapses or large . a further example of large scale vol cano collapse The morphology of the Tata Sabaya deposits, and debris avalanche formation. But was the erup­ which cover nearly 200 km' are unequivocally tion magmatic, phreatic or "cold"? Here the field P. Francis C.F. Ram írez y 109

evidence of Deruelle and Brousse (1984) for a The same situation may be true at Tata Sabaya. "nuée arden te" must be examined. The prismatic Finally, we emphasize the need for use of ap­ jointed blocks that they describe and illustrate are propriate terminology. While the deposits describ­ typical of so called "hot avalanche" deposits ed by Deruelle and Brousse clearly include prisma­ (Francis et al., 1974; Wright et al., 1980), and tic jointed blocks similar to those found in "nuée clearly indicate sorne post-collapse magmatic activ­ ardente" deposits, this does not mean that the ity. The pumice deposits briefly mentioned are whole deposit should be so described. Typical further evidence for this. Unfortunately, however, "nuée ardente" deposits have small volumes, usu­ Deruelle and Brousse do not present sufficient ally much less than 0.1 km3 (Wright et al., 1980), field evidence to determine whether the prismatic which is trivial compared to the volume of the Ta­ blocks and pum ice deposits were related to the ta Sabaya deposits. While "nuée ardente" deposits emplacement of the debris avalanche, in a "Mt. St. may sometimes be formed by " hot avalanches" Helens" type of eruption, or whether they were from a growing dacitic dome or plug, these avalan­ emplaced at sorne later stage, associated with the ches are also small, and result from the mechanical extrusion of the lava flows and domes that have instability of a growing dome, and not from the filJed the amphitheater. Our observations of the collapse of a large sector of the entire volcano. The field relations at Socompa show that prismatic Tata Sabaya deposits may include features reminis­ jointed blocks were emplaced both during the cat­ cent of "nuée ardente" deposits, but these are astrophic tollapse-related eruption, and also later merely one aspect of the phenomena resulting (over a much smaller area), during the emplace­ from the initiaI mejor collapse evento ment of the dacitic domes in the amphitheater.

REFERENCES

CRANDELL, D.R.; MILLER, C.D.; GLICKEN, H.X.; et No. 1250, p.401-420. al. 1984. Catastrophic debris avalanche from an­ KER R, R.A. 1984. Landslides from volcanoes seen as cestral Mount Shasta volcano, California. Geology, common. Science, April, 275-276. No. 12, p. 143-146. LUHR, J.; PRESTEGAARD, K. 1985. Caldera formation DERUELLE, S. 1978. The Negro de Aras nuée ardente at Volcán Colima: a large Mt. St. Helens-type ava­ deposits, a cataclysmic eruption of Socompa vol­ lanche event 4,300 years ago. EOS, No. 66, p. 410. cano ( of Chile). Bull. Vol cano!., No. 439, p. REICHE, P. 1937. The toreva block - A distinctivc.land­ 175-186. slide type. J. Geo!., No. 45, p. 538. DERUELLE, s.; SROUSSE, R. 1984. "Nuée ardente" de­ ROSIN, C.; SOUDAL, C. 1984. Une eruption remarqua­ posits at Tata Sabaya volcano (Bolivian-Chilean ble par son volume: I'evenement de type Saint-He­ Andes): Pumice and lava blocks crystallization lens du Popocatepetl. C.R.Acad. Se. Paris, No. 299, from single magma at different depths. Rev. Geol. p. 881-886. Chile, No. 22, p. 3-15. SIESERT, L 1984. Large volcanic debris avalanches: FRANCIS, P.W.; ROOSOL, M.J. ; WALKER, G.P.L etal. characteristics of so urce are as, deposits, and associ­ 1974. The San Pedro and volcanoes of ated eruptions. J. Volcano!. Geotherm. Res., No. north Chile and their hot avalan che deposits. Geol. p. Rundsch., No. 63, p. 357-388. UI, T. 1983. Volcanic dry avalanche deposits: identifica­ FRANCIS, P.W.; GARDEWEG, M.; RAMIREZ, C.F.; et tion and comparison with non-volcanic debris al. 1985. Cataclastic debris avalanche deposits of stream deposits. J. Volcano!. Geotherm. Res., No. the Socompa volcano, northern Chile. Geology, 18, p. 135-150. Vo!. 13, No. 9, p. 600-603. VOIGHT, s.; GLICKEN, H.; JANDA, R.J. et aL 1981. GROSHKOV, G.S.; DUSIK, J.M. 1970. Gigantic directed Catastrophic rockslide avalanche of May 18. In The blast at Shivcluch volcano (Kamchatka). Bull. Vol­ 1980 eruption of Mt. St. Helens, Washington. (Lip­ cano!., No. 34, p. 262-288. man, P.W.; Mullineaux, D.R.; eds.). USo Geol. Surv., HOSLITT, R.P.; MILLER, C.D.; VALLANCE, J.W. 1981. Prof. Pap., No. 1250, p. 347-377. Origin and stratigraphy of the deposit produced by WRIGHT, J.V.; SMITH, A.L; SELF, S.A. 1980. Awork­ the May 18 directed blast. In The 1980 eruptions ing terminology of pyroclastic deposits. J. Volca­ of Mt. St. Helens, Washington. (Lipman, P.W.; Mul­ no!. Geotherm. Res., No. 8, p. 315-336. lineaux, D.R.; eds.). US o Geo!. Surv., Prof. Pap.,