Driving space weather forecasts with real data

Space weather, the result of variable conditions in the near-Earth solar wind, has a number of adverse effects on Earth- and space-based technologies, particularly power infrastructure, and communication, Earth-observation and GPS satellites. Our increasing reliance on such technologies drives an increasing need to reliably forecast space weather. NERC has a central role in the science of forecasting and mitigating such natural hazards, and the UK Met Office (UKMO) is currently developing its own space-weather forecasting system.

Advanced space-weather forecasting relies on a chain of numerical models to simulate the propagation of disturbances from the Sun to the Earth. Currently, this coupled system of models is driven by observations of the magnetic field at the solar photosphere. This is used to model the magnetic field structure in the solar corona. Unfortunately this provides only very indirect information about the speed or density of disturbances, meaning estimates of their arrival time and characteristics are not well constrained.

The recently developed Heliospheric Imager (HI) instruments allow the first continual observations of the solar wind structure between the Sun and the Earth. We propose to use these direct solar wind measurements to drive the chain of space-weather forecast models, which should provide a vast improvement of the skill in which the forecast of near-Earth solar wind conditions. This new forecast scheme will require the solar wind speed and density to be estimated from HI data. Solar wind magnetic field information will still be derived by the existing method of coronal magnetic field modelling. Density will be derived from the absolute brightness of structures observed by HI. Speed will require solar wind features to be tracked through consecutive HI images. This will initially be performed by human observers, as part of the citizen science project, Solar Stormwatch. Later in the project, this tracking process will be fully automated, allowing analysis of a much greater quantity of HI data and for the scheme to be more readily transitioned into operational forecast use.

The proposed 3-year work plan covers the full development and testing of the new forecast scheme. At the end of the project, the scheme will be transitioned to the UK Met Office. It is then expected to be adopted internationally through the UKMO's close collaboration with the Space Weather Prediction Center in the US.

The UK Met Office and other international space-weather forecasting agencies will benefit directly from the outcome of this research, as will the many end-users of their forecasts, including NASA, ESA and the UKSA.

Our research would also inform those seeking to understand the risks posed by space-weather. The insurance industry is seeking to quantify the risks; Lloyds of London commissioned a briefing report into space weather and its impact on Earth and implications for business (http://www.lloyds.com/News-and-Insight/360-Risk-Insight/Research-and-Reports/Space/Space-Weather). The UK government recently included space weather in the national risk register. The space industry, spacecraft operators and space agencies also take a keen interest in space weather both to protect existing assets in space and also to inform the development of new technologies that would mitigate against such risks. Our research will improve our understanding of the Earth's space environment and provide better forecasting of the space environment around Earth.

The ability to better predict solar wind conditions will provide more accurate input to those running models of the Earth's upper atmosphere and ionosphere. The UK Met Office is currently working with modellers at University College, London to provide such a service. Solar wind transients are known to disturb the Earth's upper atmosphere. Particle precipitation, variable magnetic fields and strong electrical currents generate frictional heating that causes the atmosphere to expand and the ionosphere to become disturbed and weaken. The results of such activity are;
- An expanded thermosphere that results in increased satellite drag
- Loss of HF radio and enhanced radiation that requires long-haul flights to divert from polar regions
- Geomagnetically induced currents can cause unexpected surges in power grids
- GPS navigation systems can become less accurate and signal from spacecraft can be lost
Better forecasting of such events would enable more cost-effective action to be taken by vulnerable industries. While the aircraft industry is aware of the limitations of GPS, other users such as the maritime industries, surveyors, drivers and a host of other users with an over-reliance on GPS timing and positioning will be affected.
Other users who depend on reliable HF radio for communication are those operating in remote areas such as the military, search and rescue services and emergency aid programmes. Improved forecasts of enhanced solar wind conditions at Earth will also benefit are tourists and tour operators offering trips to see the spectacular northern lights.

There is great public interest in the Sun and its influence on our planet as demonstrated by the success of the citizen science project Solar Stormwatch (www.solarstormwatch.com) for which Chris Davis is the project scientist. This project has so far attracted over 20,000 interested members of the public, globally, to help analyse solar wind data and to provide real-time space-weather forecasts. Solar Stormwatch members have already proved that they are very capable of tracking solar wind transients and we intend to harness their enthusiasm once more in analysing solar wind data in more detail. Automatic identification of indistinct features is a complex problem but one that has been shown to benefit from an initial training set provided by manual identifications.

We will promote our project to the general public through the stormwatch blog and discussion forum, by writing articles for popular science magazines and by providing input to other interested media outlets. We anticipate that any improvements resulting from this research will become an integrated part of the forecasting process within 5 years.

Staff working on project will gain skills in image processing, statistical data analysis, communications, computer programming, problem solving, an understanding of plasma physics, project management and media.