Predicting solar storm arrivals at Earth

During the last years, alerts of solar storms on their way to Earth have been frequently sent out by the media. Solar storms or so-called coronal mass ejections (CMEs), are formations consisting of charged particles and an embedded magnetic field structure. While slow CMEs need three to five days, the fastest can reach the Earths magnetosphere within one day or less, having impact speeds of up to 10 million kilometers per hour. The consequences of these impacts are geomagnetic storms, which can damage satellites as well as lead to large-scale power outages on the ground, to name only two possible effects. Accurately predicting arrival times and speeds of CMEs is quite difficult. Because of limited observational possibilities, errors in the arrival time of 1020 hours are common. Besides the high prediction errors, false alarms are an even more important issue. False positive alarms are alerts where CMEs predicted to arrive Earth actually miss, false negative alarms are CMEs that are not predicted to arrive but actually hit. The goal of this project is the enhancement of a CME prediction tool, that currently assumes an elliptical shape of the CME front and a uniform, unstructured background solar wind, which causes a deceleration or acceleration of the CME. The basis of this prediction tool are observations from the NASA mission Solar TErrestrial RElations Observatory (STEREO) and its heliospheric imagers. These heliospheric imagers are wide-angle cameras that provide a side view on the CME during its journey through interplanetary space. The aim of this project is to uncouple the tool from the rigid ellipse shape and to include a variable background solar wind speed. By allowing a variation of the CME shape during propagation, possible influences of high speed solar wind streams or other CMEs can be taken into account when forecasting a CME arrival. Another important improvement is the applicability of the tool to observations of polarized light that can be directly related to the shape of the CME, which is further incorporated into the prediction utility. We expect a significant reduction of the prediction errors in CME arrival time and speed at Earth as well as a decrease of todays false alarm rate.

Solar storms can have far-reaching effects on technological systems in space and on earth, they can damage pipelines, induce high currents in electricity lines and disrupt communication channels. In order to be able to predict these effects, it is necessary to gain a better understanding of the propagation of these mass ejections from the solar corona (CMEs for "coronal mass ejections"). The project "Predicting Solar Storms at Earth" used data from instruments that can observe CMEs from the side as they propagate from the Sun to the Earth orbit. These data are the basis for a propagation model that was developed at the Space Research Institute of the Austrian Academy of Sciences and has now been improved. This model not only allows the data from these wide-angle cameras to be used, it also includes the interaction of CME and solar wind. The NASA mission "STEREO" consists of two identical satellites that deliver these wide-angle observations from different angles. This made it possible to investigate whether data from different observation positions lead to different results. In fact, it has been shown that there can be deviations in the arrival time at Earth of up to 18 hours - depending on the vantage point from which the solar storm is observed. Based on this knowledge, the propagation model was further developed and the shape of the CME, previously assumed to be rigid, became malleable. Now it is possible for the front of the CME to adapt to the local conditions, such as solar wind speed and density, and this could lead to a deformation of the front. If the solar storm hits regions in the solar wind that are slower than it is itself, it will be slowed down by the drag at this point and thus will be deformed. Conversely, it is also possible that a slow solar storm is accelerated and dragged along by the surrounding faster solar wind. Another innovation that this model now includes is the estimation of the mass of the CME. This is a first step towards predicting the density of a solar storm, which is partly responsible for the increased interaction with the Earth`s magnetosphere. Another important point in predicting the time of arrival of a solar storm on Earth (or at other locations in the solar system) is the quality of the data available. The wide-angle observations already mentioned are in real time only available in limited quality. A study within the project examined the effects that the use of these real-time data has on the prediction result. In fact, the use of images with a lower temporal and spatial resolution leads to a deterioration in the accuracy of the model.