This work focuses on a rigorous analysis of the\ud physical–chemical, compositional and textural relationships\ud of amphibole stability and the development of new\ud thermobarometric formulations for amphibole-bearing calcalkaline\ud products of subduction-related systems. Literature\ud experimental results (550–1,120C, \1,200 MPa, -1 B\ud DNNO B ?5), H2O–CO2 solubility models, a multitude of\ud amphibole-bearing calc-alkaline products (whole-rocks and\ud glasses, representing 38 volcanoes worldwide), crustal and\ud high-P (1–3 GPa) mantle amphibole compositions have\ud been used. Calcic amphiboles of basalt-rhyolite volcanic\ud products display tschermakitic pargasite (37%), magnesiohastingsite (32%) and magnesiohornblende (31%) compositions with aluminium number (i.e. Al# = [6]Al/\ud AlT) B 0.21. A few volcanic amphiboles (*1%) show high\ud Al# ([0.21) and are inferred to represent xenocrysts of\ud crustal or mantle materials. Most experimental results on\ud calc-alkaline suites have been found to be unsuitable for\ud using in thermobarometric calibrations due to the high Al#\ud ([0.21) of amphiboles and high Al2O3/SiO2 ratios of the\ud coexisting melts. The pre-eruptive crystallization of consistent\ud amphiboles is confined to relatively narrow physical–\ud chemical ranges, next to their dehydration curves. The\ud widespread occurrence of amphiboles with dehydration\ud (breakdown) rims made of anhydrous phases and/or glass,\ud related to sub-volcanic processes such as magma mixing\ud and/or slow ascent during extrusion, confirms that crystal\ud destabilization occurs with relatively low T–P shifts. At the\ud stability curves, the variance of the system decreases so that\ud amphibole composition and physical–chemical conditions\ud are strictly linked to each other. This allowed us to retrieve\ud some empirical thermobarometric formulations which work\ud independently with different compositional components\ud (i.e. Si*, AlT, Mg*, [6]Al*) of a single phase (amphibole),\ud and are therefore easily applicable to all types of calcalkaline\ud volcanic products (including hybrid andesites). The\ud Si*-sensitive thermometer and the fO2–Mg* equation\ud account for accuracies of ±22C (rest) and 0.4 log units\ud (maximum error), respectively. The uncertainties of the\ud AlT-sensitive barometer increase with pressure and\ud decrease with temperature. Near the P–T stability curve, the\ud error is\11% whereas for crystal-rich (porphyritic index\ud i.e. PI[35%) and lower-T magmas, the uncertainty\ud increases up to 24%, consistent with depth uncertainties of\ud 0.4 km, at 90 MPa (*3.4 km), and 7.9 km, at 800 MPa\ud (*30 km), respectively. For magnesiohornblendes, the\ud [6]Al*-sensitive hygrometer has an accuracy of 0.4 wt%\ud (rest) whereas for magnesiohastingsite and tschermakitic\ud pargasite species, H2Omelt uncertainties can be as high as\ud 15% relative. The thermobarometric results obtained with\ud the application of these equations to calc-alkaline amphibole-\ud bearing products were finally, and successfully,\ud c...

The following article presents constraints of the stability of Mg-rich (Mg/(Mg ? Fe 2? ) [ 0.5) calcic amphibole in both calc-alkaline and alkaline magmas, testing of previous thermobarometers, and formulation of new empirical equations that take into consideration a large amount of literature data (e.g. more than one thousand amphibole compositions among experimental and natural crystals). Particular care has been taken in choosing a large number of natural amphiboles and selecting quality experimental data from literature. The final database of experimental data, composed of 61 amphiboles synthesized in the ranges of 800-1,130°C and 130-2,200 MPa, indicates that amphibole crystallization occurs in a horn-like P-T stability field limited by two increasing curves (i.e. the thermal stability and an upper limit), which should start to bend back to higher pressures. Among calcic amphiboles, magnesiohornblendes and tschermakitic pargasites are only found in equilibrium with calc-alkaline melts and crystallize at relatively shallow conditions (P up to *1 GPa). Kaersutite and pargasite are species almost exclusively found in alkaline igneous products, while magnesiohastingsite is equally distributed in calc-alkaline and alkaline rocks. The reliability of previous amphibole applications was checked using the selected experimental database. The results of this testing indicate that none of the previous thermobarometers can be successfully used to estimate the P, T and fO 2 in a wide range of amphibole crystallization conditions. Multivariate leastsquare analyses of experimental amphibole compositions and physico-chemical parameters allowed us to achieve a new thermobarometric model that gives reasonably low uncertainties (T ± 23.5°C, P ± 11.5%, H 2 O melt ± 0.78wt%) for calc-alkaline and alkaline magmas in a wide range of P-T conditions (up to 1,130°C and 2,200 MPa) and DNNO values (±0.37 log units) up to 500 MPa. The A K- [4] Al relation in amphibole can be readily used to distinguish crystals of calc-alkaline liquids from those of alkaline magmas. In addition, several chemometric equations allowing to estimate the anhydrous composition of the melts in equilibrium with amphiboles of calc-alkaline magmas were derived.

No abstract

No abstract

No abstract