Home

Welcome to the PyVOLCANS wiki!

Here you can find more details about PyVOLCANS, including background information, statement of need and advanced analogue searches that can be performed with the program. Please see the wiki links for more details.

Please feel also free to contribute to the code, report any bugs or problems you encounter and/or seek support on the use and development of PyVOLCANS through the Issues tab of the repository, or through our GitHub profiles: @PTierz, @mobiuscreek, @volcan01010.

Many thanks for your interest and happy analogue volcanoes searching!

Background

PyVOLCANS is an open-access Python tool that generates data-driven sets of analogue volcanoes for any Holocene volcano listed in the Global Volcanism Program (GVP) Volcanoes Of The World database (v. 4.6.7), based on the VOLCANS (VOLCano ANalogues Search) method presented by Tierz, Loughlin and Calder (2019).

The main goal of PyVOLCANS is to help alleviate data-scarcity issues in volcanology, and contribute to developments in a range of topics, including (but not limited to): quantitative volcanic hazard assessment at local to global scales; investigation of magmatic and volcanic processes; and even teaching and scientific outreach. Please see this presentation for examples of the application of VOLCANS/PyVOLCANS to quantitative volcanic hazard at local to regional scales, within the RiftVolc project.

We hope that future users of PyVOLCANS will include any volcano scientist or enthusiast with an interest in exploring the similarities and differences between volcanic systems worldwide.

The VOLCANS method uses five volcanological criteria (tectonic setting, rock geochemistry, volcano morphology, eruption size and eruption style), and a structured combination of them, to quantify overall multi-criteria (or total) volcano analogy among any two volcanoes in the GVP database. The data used by the method are extracted from the GVP database as well as from a merged database of volcano morphology (after Pike and Clow, 1981; Grosse et al., 2014).

PyVOLCANS provides its user with full flexibility to identify customised sets of analogue volcanoes, by exploring three main variables:

    (1) target volcano (or volcano of interest);
    (2) weighting scheme (i.e. set of weights given to each of the five
    volcanological criteria to calculate multi-criteria, total analogy);
    (3) number of 'top' analogue volcanoes (i.e. those with the highest
    value of analogy with the target volcano).

In addition, PyVOLCANS allows the user to compare the values of total analogy computed for 'a priori analogues' (i.e. volcanoes thought to be good analogues to the target volcano by other strands of evidence, e.g. expert knowledge) with those computed for the rest of volcanoes in the GVP database. This permits investigation of sets of analogue volcanoes for varied purposes, and makes PyVOLCANS a useful complementary method to expert-derived analogue volcanoes. Please see Tierz et al. (2019) for more details on the VOLCANS method.

Analogy matrices

In the original VOLCANS article, the analogy matrices were calculated in Matlab. Users can find the scripts that generated those matrices in the VOLCANS_matlab_scripts folder, inside the PyVOLCANS GitHub repository. Please carefully check the README_VOLCANS file while exploring the Matlab scripts and before using them. Many thanks.

In the VOLCANS_csv_files folder, users can find the source datasets that were used to run the VOLCANS scripts and compute the analogy matrices. Please check the README_NUMBERING file while exploring and/or using these source datasets. Many thanks.

The initial releases of PyVOLCANS make use of pre-calculated analogy matrices. This represented the quickest way of making VOLCANS results available to users. In a future release, we aim to include a Python version of the code that calculates the analogies based on volcano characteristics, as this will make the method more open and widely usable.

Please also note that a few minor modifications have been implemented on some of the matrices, as a result of research developed since the original VOLCANS paper was published. These changes are the following:

• Sinabung, Indonesia (261080): (1) lava flows and lahars from the 2013–2018 eruption are included; and (2) the 2013–2018 eruption is updated to VEI 4 (GVP, 2013, database version 4.7.4). Please see Tierz et al. (2019).

• Alutu, Ethiopia (221270): rock type 'Dacite' is removed from the GVP profile of Aluto volcano. Please see Tierz et al. (2020).

• Quetrupillán, Chile (357121): the following rock types are used instead of those in the GVP 4.6.7 profile: Major (Trachyte), Minor (Basalt, Basaltic andesite, Rhyolite). A crater diameter of 1.37 km (equivalent to the value of summit width in Grosse et al., 2014) is used for Quetrupillán. Please see Simmons et al. (2020) and Simmons (2021).

Please stay updated on the new releases of PyVOLCANS and do not hesitate to contact us if you would like to know more about the VOLCANS method. You can also check the methodology section of the VOLCANS article, where careful and thorough descriptions of how the analogy matrices are calculated can be found. Thank you very much.