Four New Horsemen of an Apocalypse? Solar Flares, Super-volcanoes, Pandemics, and Artificial Intelligence

If economists have largely failed to predict or prevent the Global Financial Crisis in 2008, and the more disastrous economic collapse associated with the pandemic of 2020, what else is the profession missing? This is the question that motivates this survey. Specifically, we want to highlight four catastrophic risks – i.e., risks that can potentially result in global catastrophes of a much larger magnitude than either of the 2008 or 2020 events. The four risks we examine here are: Space weather and solar flares, super-volcanic eruptions, high-mortality pandemics, and misaligned artificial intelligence. All four have a non-trivial probability of occurring and all four can lead to a catastrophe, possibly not very different from human extinction. Inevitably, and fortunately, these catastrophic events have not yet occurred, so the literature investigating them is by necessity more speculative and less grounded in empirical observations. Nevertheless, that does not make these risks any less real. This survey is motivated by the belief that economists can and should be thinking about these risks more systematically, so that we can devise the appropriate ways to prevent them or ameliorate their potential impacts.

M-Class Solar Flares in Solar Cycles 23 and 24: Properties and Space Weather Relevance

A comprehensive statistical analysis on the properties and accompanied phenomena of all M-class solar flares (as measured in soft X-rays) in the last two solar cycles (1996–2019) is presented here with a focus on their space weather potential. The information about the parent active region and the underlying sunspot (Hale) type is collected for each case, where possible, in order to identify photospheric precondition as precursors for the solar flare eruption or confinement. Associations with coronal mass ejections, solar energetic particles, and interplanetary radio emissions are also evaluated and discussed as possible proxies for flare eruption and subsequent space weather relevance. The results show that the majority (∼80%) of the analyzed M-class flares are of β, β-γ, and β-γ-δ magnetic field configuration. The M-class population of flares is accompanied by CMEs in 41% of the cases and about half of the flare sample has been associated with radio emission from electron beams. A much lower association (≲10%) is obtained with shock wave radio signatures and energetic particles. Furthermore, a parametric scheme is proposed in terms of occurrence rates between M-class flares and a variety of accompanied solar phenomena as a function of flare sub-classes or magnetic type. This study confirms the well-known reduced but inevitable space weather importance of M-class flares.

Probabilistic hazard assessment: Application to geomagnetic activity

Probabilistic Hazard Assessment (PHA) provides an appropriate methodology for assessing space weather hazard and its impact on technology. PHA is widely used in the geosciences to determine the probability of exceedance of critical thresholds, caused by one or more hazard sources. PHA has proved useful where there are limited historical data to estimate the likelihood of specific impacts. PHA has also driven the development of empirical and physical models, or ensembles of models, to replace measured data. Here we aim to highlight the PHA method to the space weather community and provide an example of it could be used. In terms of space weather impact, the critical hazard thresholds might include the Geomagnetically Induced Current in a specific high voltage power transformer neutral, or the local pipe-to-soil potential in a particular metal pipe. We illustrate PHA in the space weather context by applying it to a twelve-year dataset of Earth-directed solar Coronal Mass Ejections (CME), which we relate to the probability that the global three-hourly geomagnetic activity index K p exceeds specific thresholds. We call this a ‘Probabilistic Geomagnetic Hazard Assessment’, or PGHA. This provides a simple but concrete example of the method. We find that the cumulative probability of K p > 6-, > 7-, > 8- and K p = 9o is 0.359, 0.227, 0.090, 0.011, respectively, following observation of an Earth-directed CME, summed over all CME launch speeds and solar source locations. This represents an order of magnitude increase in the a priori probability of exceeding these thresholds, according to the historical K p distribution. For the lower Kp thresholds, the results are distorted somewhat by our exclusion of coronal hole high speed stream effects. The PGHA also reveals useful (for operational forecasters) probabilistic associations between solar source location and subsequent maximum Kp .

A Survey of Uncertainty Quantification in Machine Learning for Space Weather Prediction

With the availability of data and computational technologies in the modern world, machine learning (ML) has emerged as a preferred methodology for data analysis and prediction. While ML holds great promise, the results from such models are not fully unreliable due to the challenges introduced by uncertainty. An ML model generates an optimal solution based on its training data. However, if the uncertainty in the data and the model parameters are not considered, such optimal solutions have a high risk of failure in actual world deployment. This paper surveys the different approaches used in ML to quantify uncertainty. The paper also exhibits the implications of quantifying uncertainty when using ML by performing two case studies with space physics in focus. The first case study consists of the classification of auroral images in predefined labels. In the second case study, the horizontal component of the perturbed magnetic field measured at the Earth’s surface was predicted for the study of Geomagnetically Induced Currents (GICs) by training the model using time series data. In both cases, a Bayesian Neural Network (BNN) was trained to generate predictions, along with epistemic and aleatoric uncertainties. Finally, the pros and cons of both Gaussian Process Regression (GPR) models and Bayesian Deep Learning (DL) are weighed. The paper also provides recommendations for the models that need exploration, focusing on space weather prediction.

Multipoint Interplanetary Coronal Mass Ejections Observed with Solar Orbiter, BepiColombo, Parker Solar Probe, Wind, and STEREO-A

We report the result of the first search for multipoint in situ and imaging observations of interplanetary coronal mass ejections (ICMEs) starting with the first Solar Orbiter (SolO) data in 2020 April–2021 April. A data exploration analysis is performed including visualizations of the magnetic-field and plasma observations made by the five spacecraft SolO, BepiColombo, Parker Solar Probe (PSP), Wind, and STEREO-A, in connection with coronagraph and heliospheric imaging observations from STEREO-A/SECCHI and SOHO/LASCO. We identify ICME events that could be unambiguously followed with the STEREO-A heliospheric imagers during their interplanetary propagation to their impact at the aforementioned spacecraft and look for events where the same ICME is seen in situ by widely separated spacecraft. We highlight two events: (1) a small streamer blowout CME on 2020 June 23 observed with a triple lineup by PSP, BepiColombo and Wind, guided by imaging with STEREO-A, and (2) the first fast CME of solar cycle 25 (≈1600 km s−1) on 2020 November 29 observed in situ by PSP and STEREO-A. These results are useful for modeling the magnetic structure of ICMEs and the interplanetary evolution and global shape of their flux ropes and shocks, and for studying the propagation of solar energetic particles. The combined data from these missions are already turning out to be a treasure trove for space-weather research and are expected to become even more valuable with an increasing number of ICME events expected during the rise and maximum of solar cycle 25.