Using Statistics to Quantify and Communicate Uncertainty During Volcanic Crises

Abstract

For decades, and especially in recent years, there has been an increasing amount of research using statistical modelling to produce volcanic forecasts, so that people could make better decisions. This research aims to add confidence by arming users with quantitative summaries of the chaos and uncertainty of extreme situations, in the form of probabilities-that is to say the measure of the likeliness that an event will occur.

Partly Cloudy With a Chance of Lava Flows: Forecasting Volcanic Eruptions in the Twenty‐First Century

“…Encouragingly, volcano forecasts and hazards assessments have become far more probabilistic in recent years. Of course, probabilistic forecasts come with their own set of challenges, including how to communicate uncertainties with land managers and the public (e.g., Dolye et al, ), how to think about and account for the uncertainty in uncertainties (Sobraledo & Martí, ), and how to evaluate success. This latter challenge is critical because, in some regards, a probabilistic forecast cannot be “wrong” unless the probability of an event is incorrectly assigned 0 or 100% (even very unlikely events do occur in nature).…”

Small Data's Big AI Potential

“…For instance, imagine building an algorithm predicting disease risk for a set of people that do not have electronic health records, or forecasting the likelihood of eruptions in volcanoes with long eruptive recurrence. 19 Small data approaches can provide us with a principled way to deal with this lack or absence of data. It can do so by transferring knowledge from a related problem, making use of both labeled and unlabeled data.…”