S. Tonarini et al., GEOCHEMICAL AND ISOTOPIC MONITORING OF MT-ETNA 1989-1993 ERUPTIVE ACTIVITY - BEARING ON THE SHALLOW FEEDING SYSTEM, Journal of volcanology and geothermal research, 64(1-2), 1995, pp. 95-115
Two flank eruptions took place at Mt, Etna during 1989-1993. The forme
r (September-October 1989) produced lavas among the most primitive of
this century while the latter (December 1991-March 1993) was unusually
long lasting and the largest of the last three centuries. Alkali basa
lts of 1989 displayed initial enrichments of Rb and K accompanied by a
n higher value of radiogenic-Sr and isotopic disequilibrium between ho
st-rock and clinopyroxene (0.70364 and 0.70356, respectively). Subsequ
ently, within one week, K, Rb and Sr-87/Sr-86 ratio decreased to const
ant values (Sr-87/Sr-86 = 0.70355). Nd isotopic ratio did not show sig
nificant variations. These features can be explained by a selective co
ntamination process involving essentially Rb, K and radiogenic-Sr. The
bulk from the 1991-1993 eruption hawaiites, were erupted between Janu
ary and May 1992. In this period, lavas (here called JML) were charact
erised by uniform geochemical and isotopic compositions and Sr-isotopi
c equilibrium between whole rock and pyroxene (0.70355). In other peri
ods (December 1991 and June 1992) we observed lower and variable Sr is
otopic compositions coupled with disequilibria between lavas and their
clinopyroxenes (0.70351 in the w.r. and 0.70345 in the cpx of the sam
ple of 15 December 1991). Starting from June 1992, the products seem t
o reveal fractionation effects as suggested by the continuous rise in
concentrations of incompatible elements. When the observed variability
cannot be ascribed to fractionation effects, we are not able to discr
iminate mixing from selective contamination. However, lavas erupted du
ring December 1991 evolved towards the isotope composition of JML that
was never exceeded, moreover the regular behaviour of Rb, K and radio
genic-Sr of 1989 lavas was not observed during the 1991-1993 eruption,
Thus, we prefer to interpret these variations as mainly due to mixing
processes between an end-member of JML composition and another one ha
ving the same, or lower, isotopic composition of the clinopyroxene fou
nd in the sample of December 1991. Several petrogenetic processes are
contemporaneously active at Mt. Etna; they encompass magma mixing, sel
ective contamination and fractional crystallization. Their relative im
portance varies according to the original volume of each magma batch a
nd to the status of the shallow feeding system. The complexity of thes
e processes, acting in the short time span of a few months, claims cau
tion about the geochemical representativity of a single lava sample fo
r each eruption.