IAVCEI/COSIV Working Group on Volcanic Record Completeness

Chairs (pro-tem)

• Mark Bebbington and Susanna Jenkins
• The organization and leadership of the working group will be decided at the IAVCEI 2017 pre-meeting workshop (see below)

Upcoming Workshop

IAVCEI 2017 pre-meeting workshop Statistics in Volcanology: Workshop on Volcanic Record Completeness August 13, 2017, Portland.

Initial Questions and Thoughts

The workshop will consider and debate these questions, and any others, in order to frame the future activities of the working group. We have initially divided the questions into two categories: a) how to determine (where?) a record is complete, and b) what to do with the result.

Definitions

• Notation
• R = observed record
• H = actual eruptions (the ‘history’). Note: R is a subset of H
• P(x,t,m) = probability that an eruption at volcano x, of age t and size m, is in R
• With a certain amount of notation abuse, R = P × H
• Naive Definition: A record is complete from age T at magnitude M if every post-T eruption of magnitude M or larger is present: P(t,m) = 1 for all t <= T, m >= M
• More Statistical definition: A record is complete from age T at magnitude M if every post-T eruption of magnitude M or larger is present with probability P > 1-e, for some e = o(1): P(t,m) > 1-e, e = o(1), for all t <= T, m >= M

Determining Completeness

1. Are separate approaches required for the ordinal VEI records (e.g. GVP) and the cardinal magnitude records (e.g. LaMEVE) records? If so, how do we merge the results?
• Magnitude is much better behaved. The VEI scale is multiply defined, has discontinuities at VEI 1/VEI 2, and at approximately AD1600, before which most VEIs are increased by one, and VEI 2 is a ‘default’ class for unknown sizes. But, and it’s a big but, magnitude is available for << 10% of known eruptions.
• Certain (how many?) volcanoes have records which are more detailed than in either or both of GVP and LaMEVE. How do we meld these into a system?
2. Is there a valid distinction between, or differing levels of confidence in records from, instrumental observation, written accounts, oral traditions and geological records?
• For certain they all have different scope and reliabilities. On the other hand, one way of getting more eruptions (other than waiting around) is via geological exploration. So these are ‘targeted’ (in some sense, biased).
3. To what extent can closeness in space be used to borrow strength? (e.g. if one volcano is complete far in the past, then the neighbouring volcano is likely to be as well, even if there are no eruptions).
• This applies particularly to written and oral records, and also to geological investigation, although it can be a nuisance there.
• Deposits are usually credited to volcanoes on the basis of their present geochemistry. Is this valid for 10 ka deposits?
4. To what extent can we group volcanoes based on their typology, tectonic or geographical setting?
• The use of such analogue volcanoes is being advocated in monitoring and forecasting. Can it be applied to completeness?
5. Is there a way of getting a handle on the long-term periodicities of volcano activity (if they exist) in order to better discriminate between missing data and quiescence?

Dealing With Incompleteness

1. In estimating frequency-magnitude (stationary hazard) relationships, how do we account for extreme events that are rarely captured in any one volcano's record? (Related to Qs 4 and 5 above.
2. What are the pros and cons of using short complete records versus longer incomplete records?
• Using a short record is accurate within the bounds of what’s in the record, but you may underestimate the chances of a large (hence infrequent) eruption
• Averaging across a long, but incomplete, record can underestimate the frequency of poorly recorded smaller eruptions
3. What about when using time-varying hazard models, i.e., when we consider trends and/or cycles in volcano activity, and the time since the last eruption is a datum. How do we identify prior quiescent (as opposed to incomplete recording) periods and hence inform the distribution of the remaining time to eruption?
4. How do we appropriately carry through uncertainty bounds from underlying data to our frequency-magnitude and hazard estimates?
• Is it possible to estimate the downstream biases and quantify our uncertainty in hazard estimation e.g. tephra fall, PF probabilities, etc. from using an incomplete record?

Members/Mailing list

```Ben Andrews "AndrewsB@si.edu";
Willy Aspinall "Willy.Aspinall@bristol.ac.uk";
Lucia Dominguez Barragan "Lucia.Dominguez@unige.ch";
Mark Bebbington "m.bebbington@massey.ac.nz";
Maria Benito Saz  "mabenito@fomento.es";
Sonia Calvari "sonia.calvari@ingv.it";
Kathy Cashman "glkvc@bristol.ac.uk";
Chuck Connor "cbconnor@usf.edu";
Elizabeth Cottrell "cottrelle@si.edu";
Natalia Deligne "N.Deligne@gns.cri.nz" ;
Ciro Del Negro "ciro.delnegro@ingv.it";
Benjamin Edwards "edwardsb@dickinson.edu";
Victor-Hugo Forjaz "vforjazovga@gmail.com";
Jean-Robert Grasso  "grassojr@yahoo.fr";
Olivier Jaquet "olivier.jaquet@in2earth.com";
Susanna Jenkins "Susanna.Jenkins@bristol.ac.uk" ;
Leif Karlstrom "leif@uoregon.edu";
Koji Kiyosugi "kkiyosug@mail.usf.edu";
Christine Lane "christine.lane@manchester.ac.uk";
Einat Lev "einatlev@ldeo.columbia.edu";
Christina Magill "christina.magill@mq.edu.au";
Sue Mahoney "sue.mahoney@bristol.ac.uk";
Tamsin Mather "Tamsin.Mather@earth.ox.ac.uk";
Ana Teresa Mendoza Rosas "ateresa@geofisica.unam.mx";
Gary Michelfelder "GaryMichelfelder@MissouriState.edu";
Maurizio Mulas "mmulas@espol.edu.ec";
Sarah Ogburn "sogburn@usgs.gov";
Carolyn Parcheta  "Carolyn.E.Parcheta@jpl.nasa.gov";
David Pieri "dave.pieri@jpl.nasa.gov";
Richard Robertson "richie.robertson@gmail.com";
Jonty Rougier "J.C.Rougier@bristol.ac.uk";
Melody Runge "mhum400@aucklanduni.ac.nz";
Laura Sandri "laura.sandri@ingv.it";
Anja Schmidt "A.Schmidt@leeds.ac.uk";
Tom Sheldrake "Thomas.Sheldrake@unige.ch";
Steve Sparks "Steve.Sparks@bristol.ac.uk";
Karen Strehlow "ks12548@bristol.ac.uk";
Ed Venzke "venzkee@si.edu";
Kristi Wallace "kwallace@usgs.gov";
Ting Wang "ting.wang@otago.ac.nz";
John Wardman "john.wardman@bios.edu";
Sebastian Watt  "S.Watt@bham.ac.uk";
Patrick Whelley "patrick.l.whelley@nasa.gov";
Alana Weir "aw1300@my.bristol.ac.uk";
Gordon Woo "Gordon.Woo@rms.com";
Indira Zuluaga "indirazuluaga@gmail.com";
```

Bibliography

A starter list with some of the relevant literature to date. Please suggest additions.

Completeness

• Coles SG, Sparks RSJ (2006) Extreme value methods for modelling historical series of large volcanic magnitudes. In: Mader HM, Coles SG, Connor CB, Connor LJ (eds) Statistics in Volcanology. Special Publications of IAVCEI, Geological Society, London
• Deligne NI, Coles SG, Sparks RSJ (2010) Recurrence rates of large explosive eruptions. J Geophys Res 115, B06203
• Furlan C (2010) Extreme value methods for modelling historical series of large volcanic magnitudes. Statist. Modelling 10, 113-132.
• Guttorp P, Thompson ML (1991) Estimating second-order parameters of volcanicity from historical data. J Amer Statist Assoc 86, 578-583.
• Hill BE, Aspinall WP, Connor CB, Godoy AR, Komorowski J-C, Nakada S (2009) Recommendations for assessing volcanic hazards at sites of nuclear installations. In Connor CB, Chapman NA, Connor LJ (eds) Volcanic and Tectonic Hazard Assessment for Nuclear Facilities. Cambridge University Press, pp. 566-592.
• Jenkins S, Magill M, McAneney J, Blong R (2012) Regional ash fall hazard I: a probabilistic methodology. Bull Volcanol 74, 1699-1712.
• Kiyosugi K, Connor C, Sparks RSJ, Crosweller HS, Brown SK, Siebert L, Wang T, Takarada S (2015) How many explosive eruptions are missing from the geologic record? Analysis of the quaternary record of large magnitude explosive eruptions in Japan. J Appl Volcanol 4, 17.
• Mead S, Magill C (2014) Determining change points in data completeness for the Holocene eruption record. Bull Volcanol 76, 874.
• Rougier J, Sparks SR, Cashman KV (2016) Global recording rates for large eruptions. J Appl Volcanol 5, 11.
• Simkin T (1993) Terrestrial volcanism in space and time. Ann Rev Earth Planet Sci 21, 427-452
• Wang T, Bebbington M (2012) Estimating the likelihood of an eruption from a volcano with missing onsets in its record. J Volcanol Geotherm Res 243-244, 14–23.
• Whelley PL, Newhall CG, Bradley KE (2015) The frequency of explosive volcanic eruptions in Southeast Asia. Bull Volcanol 77, 1.

Size Distribution

• Bebbington MS (2014) Long-term forecasting of volcanic explosivity. Geophys J Int 197, 1500-1515.
• De La Cruz-Reyna S (1991) Poisson-distributed patterns of explosive eruptive activity. Bulletin of Volcanology 54, 57-67
• Hurst T, Smith W (2010) Volcanic ashfall in New Zealand - probabilistic hazard modelling for multiple sources. NZ J Geol Geophys 53, 1-14.
• Marzocchi W, Zaccarelli L (2006) A quantitative model for the time-size distribution of eruptions. J Geophys Res 111, B04204
• Newhall CG, Self S (1982) The volcanic explosivity index (VEI): An estimate of the explosive magnitude for historical eruptions. J Geophys Res 87, 1231-1238.
• Pyle DM (1998) Forecasting sizes and repose times of future extreme volcanic events. Geology 26, 367-370
• Pyle DM (2000) Sizes of volcanic eruptions. In Sigurdsson H et al. (eds.) Encyclopedia of volcanoes. Academic Press, pp 263-269.

Analogue Volcanoes

• Jenkins S, Magill M, McAneney J, Blong R (2012) Regional ash fall hazard I: a probabilistic methodology. Bull Volcanol 74, 1699-1712.
• Rodado A, Bebbington M, Noble A, Cronin S, Jolly G (2011). On selection of analogue volcanoes. Math Geosci 43, 505-519.
• Sheldrake T (2014) Long-term forecasting of eruption hazards: a hierarchical approach to merge analogous eruptive histories. J Volcanol Geotherm Res 286, 15-23.
• Whelley PL, Newhall CG, Bradley KE (2015) The frequency of explosive volcanic eruptions in Southeast Asia. Bull Volcanol 77, 1.

Time Varying Hazard

• Bebbington M (2010) Trends and clustering in the onsets of volcanic eruptions. J Geophys Res 115, B01203, doi:10.1029/2009JB006581.
• Bebbington MS (2013) Models for temporal volcanic hazard. Statistics in Volcanology, 1, 1-24.
• Green R, Bebbington MS, Cronin SJ, Jones G (2013) Geochemical precursors for eruption repose length. Geophys J Int 193, 855-873.
• Marzocchi W, Zaccarelli L (2006) A quantitative model for the time-size distribution of eruptions. J Geophys Res 111, B04204
• Mulargia F, Gasperini P, Tinti S (1987) Identifying regimes in eruptive activity: An application to Etna volcano. J Volcanol Geotherm Res 34, 89-106.
• Wadge G (1982) Steady state volcanism: Evidence from eruption histories of polygenetic volcanoes. J Geophys Res 87, 4035-4049.

Uncertainties

• Anderson JG, Biasi GP (2016) What is the basic assumption for probabilistic seismic hazard assessment? Seismol Res Lett 87, 323-326

Page maintained by

Mark Bebbington
• email: m.bebbington@massey.ac.nz
• tel: +64 6 951 7641
• fax: +64 6 350 5632
• postal address: Institute of Fundamental Sciences - Statistics, Massey University, Private Bag 11222, Palmerston North, New Zealand