By: Luke Barnard
Most people are familiar with the fact the Sun emits a range of electromagnetic radiation (e.g. sunlight), and that this radiation is necessary to sustain life on Earth as we know it. What is less well known is that alongside the Sun’s electromagnetic radiation, it also generates a wind of plasma that continuously blows out through the Solar System, with speeds of 250 km/s to 750 km/s.
This solar wind impacts our everyday lives through its effects on the technology we increasingly depend on; particularly spacecraft in orbit around Earth. We rely on satellites for critical services such as communications, GPS, and weather forecasting. When services like these are disrupted, it can have both expensive and dangerous consequences .
During periods of intense solar wind activity, it squeezes and shakes Earth’s magnetic field. This produces energetic charged particles which are harmful to satellite electronics and can also make it difficult to maintain radio communications with them. Depending on how intense the solar wind is, satellites can be temporarily or permanently damaged, with knock on impacts to the services they provide.
Space Weather Forecasting grew out of the need to understand and predict when situations like this would occur. A key challenge in space weather forecasting is to be able to forecast the solar wind flow throughout the Solar System. This is difficult because there are only a handful of spacecraft able to measure the solar wind, and these only measure it at single points which are vastly separated. By way of analogy, it is like trying to forecast the weather at Reading, with only weather observations at a few other far away cities, like Exeter, Manchester and Brighton. The limited information is still useful, but there is a lot that can happen in between and a lot of uncertainty.
Our research  aims to help solve this problem by using images of the solar wind plasma to characterise the solar wind flow near the Sun. This would be an extra source of information on the solar wind flow, which we could use to help improve computer models that forecast the solar wind.
Figure 1: This shows the relative locations of Earth, STEREO-A and STEREO-B. The purple shaded regions show the field-of-view of the inner Heliospheric Imager camera on STEREO-A.
NASA’s STEREO mission consists of two spacecraft which are in Earth-like orbits, but they drift relative to Earth [Figure 1], so that they can observe the space between the Sun and Earth. On each spacecraft are a pair of cameras called the Heliospheric Imagers, which produce images of the solar wind plasma [Figure 2]. The cameras record visible sunlight that has scattered off of electrons in the solar wind. Interpreting the images is tricky because there are electrons and visible light everywhere in space, and so we don’t actually produce an image of a specific feature or object. But, because we understand the physics of sunlight well, and of how sunlight scatters off of electrons, we are able to use these images to identify regions where there are relatively more electrons, and a denser solar wind.
Figure 2: A movie of heliospheric imager images from July 2008. Movie obtained from the UK Solar System Data Centre
Our aim was to show that variability in the images could be statistically related to the direct single point measurements of solar wind flow observed by other spacecraft. This would be the first step in creating and calibrating a technique to estimate the solar wind flow directly from the images.
We compared the solar wind point measurements and images directly, computing the correlation between variability in the images recorded by STEREO-A with the solar wind speed measured directly at Earth, STEREO-A, and STEREO-B. We found that there is a strong correlation between variability in the images and the solar wind speed observations at the three spacecraft, but that the correlation was largest when a delay was applied between the image and solar wind observations. This delay was different for each pair of spacecraft, and changed in time in a way that can only be explained by the orbits of the spacecraft. Based on this statistical analysis we have concluded that we probably can trace the flow of the solar wind in the Heliospheric Imager data. Our next step is to investigate how to best compute a reliable estimate of the solar wind speed directly from the images.