The keys to coping with (extreme) space weather events
Satellites’ vulnerability to space weather is by no means a thing of the past. Sure, the last major event took place in 2003: From 29 to 31 October, the so-called Halloween solar storms disrupted over 47 satellites and even took out one Japanese satellite costing US$640m sparking a wave of concerns among governments, satellite operators and other stakeholders like insurance companies. But whilst these events occurred almost 15 years ago, more recent ones like the malfunctions of Galileo atomic clocks or the loss of a Kazakh satellite in late March 2017 — which are still being investigated — may have been caused by space weather-related radiation. ‘Space weather can cause interruptions in service for periods of weeks or even months, which can be very expensive for satellite operators. In the case of the KazSat-2, we know that the incident happened when the radiation belts had been enhanced, and that’s just the latest example. After all these years and accumulated experience, the reality is that we still have satellite damage very likely caused by space weather events,’ says Prof Richard Horne, science leader at British Antarctic Survey and coordinator of the SPACESTORM (Modelling space weather events and mitigating their effects on satellites) project. SPACESTORM was born out of the observation that there is room for improvement in existing space weather prediction models, that we don’t know enough about space weather impacts in Low and Medium Earth Orbits, and that stakeholders should be able to make more informed decisions about how satellites should be designed. Finally, the consortium wanted to answer the question most stakeholders don’t dare to ask themselves: how many satellites would be lost should an extreme space weather event occur? To answer this question, Prof Horne and his team decided to focus in part on the Galileo constellation. ‘Information on the design of the Galileo spacecraft and how much radiation protection they include are all held in great confidence, just as for most commercial spacecraft, so it’s very difficult to assess them’ he explains. ‘So what we did is to try and work out what we believe is the environment in Medium Earth Orbit (MEO) where the Galileo satellites fly. We calculated the worst case electron spectrum, and looked at how much shielding might be required to protect the spacecraft.’ Similar calculations were made for the geostationary orbit, using a statistical analysis of existing data along with the project’s own physical model. ‘It’s remarkable that these two different approaches come together with a result that is very similar, which makes us quite confident in our approach,’ Prof Horne says, before continuing: ‘We actually found that the flux tends towards a limiting value of between 5 x 105 and 2 x 106 cm-2 s-1 sr-1. This corresponds to a current of between 1 and 4 pA cm-2 — which exceeds by a factor of 10 the NASA recommended guidelines.’ Based on these results, the team found that designers would have to almost double the shielding of their satellites to be confident of surviving an extreme space weather event. ‘The decision, of course, is in the hands of the business manager,’ Prof Horne points out. ‘It can be done, but that’s going to cost a huge amount of money in terms of launching the spacecraft. As the scientific community believes that the probability of an extreme event is one in 100 or 150 years, the question is, does your company want to plan for that kind of event or not?’ Digging into past event, forecasting future ones Another key contribution of the project is the understanding it brings of all space weather events, both from the past and in the future. The team has been working closely with business, Governments, and the European Space Agency to make sure they understand the risks and can make informed decisions about the best course of action. To make this possible, they developed radiation belt models and reconstructed 30 years of radiation environment for the whole of the outer radiation belt, including medium and low Earth orbit for which there was only little data available so far. All this, with much more accuracy than what was possible before. ‘We tested our reconstruction by comparing it with data we had for a few periods in medium Earth orbit (MEO), for example against the Giove data, and obtained a Heidke skill score of about 0.7. Heidke skill score determines how good your forecasting is 1: A score of 1 is a perfect forecast, 0 is a bad one. Ours is pretty good,’ Prof Horne enthuses. Insurance companies are already showing interest in these models to examine what the radiation environment was like when space weather events occurred in the past. Icing on the cake, the project created a forecasting website capable of planning changes in the radiation belt up to three hours ahead. Combined with predictions of impacts on engineering, the website can provide a risk indicator for the four main space weather risks affecting satellites. ‘If satellite operators know from a forecast that their satellites will be at high risk, they can perhaps postpone manoeuvres, delay a software update, bring additional people in or make more transmission capacity immediately available,’ Prof Horne says. Projects partners at ONERA in France went a step further with new experiments on the main materials being used on satellites. Their goal: to determine whether current lab methods of exposing materials to intense radiation for short periods can truly represent the effects of long — up to 15 years — exposure in space. ‘It quickly became apparent that electrical properties are very important,’ says Prof Horne. ‘Irradiating some materials like Kapton wires actually changes their conductivity, so radiation exposure tests in the lab should be complemented with more experiments on the material itself.’ One commercial company has already started using results from the project when deciding which type of satellite they should buy, and Prof Horne believes this will eventually provide incentives for the design of a new generation of materials. With new technologies like passive emitters and electric propulsion respectively changing how satellites become charged and how much time they remain exposed to radiation, SPACESTORM results should continue to prove very useful in the near future. Whilst the project is now completed, the forecasts will be developed further under a new project funded by the ESA, and the team is already working on improving them to be able to predict space weather for up to 24 hours ahead.
Keywords
SPACESTORM, GPS, Galileo, shielding, materials, engineering, satellite, low earth orbit, geostationery, space weather events, radiations, ESA, forecasting