Probing an active, exceedingly shallow rhyolite reservoir with experiments and petrological tools: the case of Krafla IDDP1

The discovery of a rhyolitic magma body at a depth of 2.1 km during drilling of the first well in the Iceland Deep Drilling Project (IDDP-1) at Krafla volcano, NE Iceland, presented an unprecedented opportunity to explore shallow magma properties and the root of geothermal systems. Yet, to safely access this near-magma energy, we require an in-depth understanding of magmas’ response to drilling activity and power plant operations, which we target here using the natural samples retrieved in-situ from the magma body during IDDP-1. The glassy fragments display a spectrum of colors (light brown to black), different vesicularities and crystals of plagioclase, pyroxene, Ti-magnetite, and apatite; some crystals exhibit zonation and embayment.

In this study we experimentally explore the stability of the rhyolitic magma to different P-T-X conditions to assess magma response to perturbations prompted by drilling. We tested pressures of 16, 35, and 45 MPa (between hydrostatic pressure and the pressure as estimated by dissolved H₂O-CO₂ concentration), temperatures ranging from 880 to 920°C, and durations spanning from 6 to 48 hours. All experiments were carried out under water-saturated conditions, with oxygen fugacity fixed by Ni and Co filler-rods to NNO+1 (more oxidized) or QFM (more reducing), respectively. We used the original glass chips (without remelting them) to see how the texture would evolve when subjected to different P, T, X. The main difference observed is when comparing experimental products at NNO+1 or QFM conditions which influenced glass color: darker hues appeared under NNO+1 condition and lighter hues prevailed under QFM condition. Whilst at NNO+1 the phases present in the original mineralogical assemblage were generally relatively stable (with one exception; see below). At 45 MPa, we observe no dissolution of the original phases but overgrowth of pyroxene in all charges. By reducing temperature from 900 ºC to 880 ºC or by increasing the oxygen fugacity (to NNO+1) we observed an increase in microlite content of the same original phases (except for apatite which did not crystallize). At the lower pressure of 35 MPa, the microlite content was higher than at 45 MPa; yet, the original mineralogical assemblage remained, whereby no dissolution took place and pyroxene overgrowth occurred. Again, by decreasing temperature from 920 ºC to 880 ºC or by increasing the oxygen fugacity the microlite content increased. Importantly, quartz crystallized at 880 °C, 35 MPa and under NNO+1 conditions; indicating that these conditions were likely not met during drilling. At 16 MPA, the experiments failed as the Au capsules ruptured due to pressure from excess fluids. Our initial findings suggest that the magma may reside in the crust at a minimum temperature of 900 °C, if at 45 MPa, or 920 °C, if at 35 MPa. This work establishes a “reference frame” for understanding shifts in magmatic parameters that may be triggered by drilling into active systems.