I'm a Lecturer in Volcanology in the School of Geosciences at The University of Edinburgh and researcher at CEMPEG, Uppsala University, Sweden.

My research focuses on the evolution of magmas in the sub-volcanic plumbing system. The formation of magma (composed of molten rock and solid particles termed crystals) is often complex and multi-stage. Although ultimately sourced from the mantle, the magmatic composition is often modified during assent through a combination of assimilation of foreign crustal material, mixing of magmas, crystallisation. These processes creates chemical heterogeneities within the magma, although subsequent magmatic evolution may erase these signatures. An alternative is to study the hosted crystals that have the potential to preserve a record of the changing magmatic conditions within their crystal structure. Crystals are interrogated through a combination of micro-analytical techniques (SEM, (FEG)-EPMA, LA-ICPMS, TOF-SIMS, NanoSIMS) to forensically fingerprint individual crystal zones and melt inclusions to decipher the magmatic evolution.

Crystal specific studies

Crystals hosted within magmatic bodies are often zoned providing a record of their history. Individual zone have subtly different chemical compositions, reflecting changes in physical conditions within the magma chamber. Chemical fingerprinting of individual zones allows the processes that created them to be constrained, retrospectively.  In addition, the chemical exchange of elements between two adjacent zones is a time-dependant process that can be modelled through diffusion chronometry methods. This allows a petrological time series to be created that can be linked to the contemporaneous, pre-eruptive volcano monitoring record. One of the enduring challenges is to link the monitored signals at active volcanoes to the actual magmatic processes. This is not simple as a sample of magma is required that can only be gained through a subsequent eruption. 

Orthopyroxene crystals from the 1980-86 eruption of Mount St. Helens were characterised through a combination of high-resolution imaging of crystals combined with chemical fingerprinting. This revealed the present of multiple crystal populations that were generated through genetically related magmas. Concentric zoning was observed in over 100 crystals with either Fe-rich or Mg-rich rims which were modelled through Fe-Mg diffusion chronometry to determine the time lapsed between crystal growth and volcanic eruption. Surprisingly both Fe-rich and Mg-rich rims grew at the same time and generally within 12 months prior to eruption. Integrating this petrologic time series with the seismic and SO2 flux revealed that peaks in crystal growth correlate with extremely well with increased seismicity and SO2 flux recorded at Mount St. Helens. Establishing that important but elusive evidence that clearly shows there is a relationship between seismicity and magma movement, that crystal growth is commensurate with the volcano monitoring record, and that geophysical monitoring techniques do record an accurate record of the arrival of magma pulses.  Such a correlation as shown in this study strongly suggests that petrologic time series from ancient or recently re-active volcanic centres where monitoring is limited can provide fundamental insights in pass volcanic activity and can be used to help with mitigation for future eruptions.For full study see Saunders et al. (2012). 

Melt inclusions 

Melt inclusions are small bubbles of melt that are entrapped within a crystal during growth. They provide a window into earlier stages of the magmatic evolution compared to the groundmass glass composition that reflects the final melt composition prior to eruption. 

Geochemical forensics of an active volcanic arc: Sumatra, Indonesia. 

Andesitic volcanoes are typically located at destructive plate margins where two of the Earth’s tectonic plates converge, with one plate sinking or subducting into the Earth’s mantle. This leads to the formation of volcanic arcs on the Earth’s surface composed of a chain of dominantly explosive andesitic volcanoes. Petrogenesis of andesites however is contentious with arguments for the primary generation of andesitic melts from the mantle to mixing and interaction of basalt and rhyolitic magmas in the crust.. Studies of arc volcanism are increasingly providing evidence of multiple crystal populations derived from polygenetic sources revealing a single liquid line of descent of magmas is improbable

The Indonesian Archipelago (intra oceanic subduction system) is formed by the subduction of the Indian-Australian plate beneath the Eurasian plate.  The western and central segment (Sumatra, Java) of the Sunda Arc is an oblique subduction regime beneath relatively thick (ca. 25 km) continental crust. The continental crust in the western Sunda arc is hence variable in composition, consisting of mixtures of volatile rich sediments, sandstones, granites and carbonates. The diversity in crustal material makes this is an ideal place to test the nature and amount of contribution continental crust poses to arc volcanism.

The Toba supereruption (74 ka) and the 1883 eruption of Krakatau are two well-documented explosive and destructive eruptions originating from the western Sunda Arc. These two volcanoes located to the north and south of Sumatra sandwich over 19 further volcanoes that could be potentially as dangerous: although few detailed studies of these other volcanic complexes have been conducted. Detailed sampling of 16 of these volcanoes (Bual Buali, Sorik Marapi, Marapi, Talang, Kerici, Berang, Kaba, Dempo, Sikincau, Sekibcau, Ratai, Rajabasa, Krakaktau) and two located to the north of Toba (Sibayak, Sinabung) was conducted in 2010, providing a comprehensive sample set on which to base this study. Despite these volcanoes being located in close proximity to two highly explosive and destructive volcanoes (Toba and Krakatau), comparatively few studies on their magma genesis and the potential hazards their eruptions pose to the local communities and for global climate exist. A detail study of both whole rock and in-situ micro-analytical techniques are being applied to interrogate the genesis of these andesitic volcanoes.

This is an on-going study with researchers based at Uppsala University and is funded through a Swedish Research Council Grant to Saunders and Troll. A selection of preliminary results of this study can be located here.

Analytical Development

Advances in analytical techniques have allowed the quantitative analysis of geological materials on a micrometre resolution. However, many of the components of volcanic samples of interest now require a nanoscale spatial resolution. In the last few years in collaboration with labs worldwide, I have been developing analytical protocals to allow the analysis of zoned crystals at spatial resolutions down to 200 nm. 

Results of a recent FEG-EPMA study of plagioclase and pyroxene crystals on a sub-micron resolution can be read here.

Current research interests

          1. the relationship of basalt, andesite and rhyolitic magmas in arc settings and their petrogenesis.
          • application of crystal forensics methods to decipher the history of magmas.
          • application of diffusion chronometry methods to investigate the timescales of magmatic processes prior to eruption.
          • linking crystal specific studies to monitoring signals at active volcanoes
          • development of protocols of the analysis of geological materials at nanoscale spatial resolutions. 
Please do contact me if you are interested in PhD's or postdocs and we can discuss potential opportunities and funding.