Evaluation of non-conventional geothermal potential in a volcanic island

Madeira, is an intraplate volcanic island, located at the eastern North Atlantic Ocean, with an emerged area of 737 km2 and maximum altitude of 1861 m. Although there are no historical eruptions, the existence of recent volcanism (6 my) with well preserved volcanic cones and thermal evidences, such as the occurrence of hot water rich in CO2, suggest a heat source at subsurface and the existence of rocks/water with significant temperatures at depths likely to be exploitable for economic generation of electricity.
In a volcanic geothermal system the heat source comes from magma emplacement at relatively shallow levels, thus knowledge of magma chamber(s) depth is one of the keys to geothermal reservoir assessment. In the evaluation of Madeira island geothermal potential, geophysical methods are being applied but, these are most useful for locating chambers beneath active volcanoes. Petrological and geochemical methods supported on whole rock analysis and mineral chemistry can be a helpful tool to constraint the crystallization temperatures and pressures/depths of magmas and to unveil the physical-chemical crystallization history of selected phenocrystals, hence the depth of magma chambers. Volcanic rocks (effusive and explosive) in Madeira are predominantly alkaline basalts. Generally they are holocrystaline presenting porphyritic texture with phenocrystals of olivine and clinopyroxene and, sometimes, calcic plagioclase in a groundmass composed by plagioclase microlites, clinopyroxene, oxides and occasionally interstitial glass. Petrographic and chemical criteria show that phenocrystals had a polibaric crystallization, suggesting that they paused in crustal magma chambers prior to eruption, and that the first minerals to crystallize were olivines and clinopyroxenes. One of the methods to determine the pressure of crystallization is the Ol-Cpx-Plag cotetic method that requires glass composition, or an assumption of it. We applied selected geothemobarometers, based on olivine-liquid and clinopyroxene-liquid equilibrium, to core-mantle analysis of olivine and clinopyroxene phenocrysts to estimate P, which is related to the depth at which magma resides in a chamber, and T of equilibrium crystallization. When required, oxygen fugacity (gO2) was calculated from oxides and whole rock composition. The first results seem to point out a concentration of core crystallization occurring between 2 to 4 kb/6 to 12 km, which is a good indication of a single and wide magma chamber beneath the island.