Community Research and Development Information Service - CORDIS

H2020

PROGRESS Report Summary

Project ID: 637302
Funded under: H2020-EU.2.1.6.

Periodic Reporting for period 1 - PROGRESS (Prediction of Geospace Radiation Environment and solar wind parameters)

Reporting period: 2015-01-01 to 2015-12-31

Summary of the context and overall objectives of the project

Just as weather can be expressed as a set of atmospheric parameters that are important not only for our comfort but also determine the conditions for the operation of technological systems on the ground and in the atmosphere, space weather is expressed by the set parameters relating to the near Earth environment that determine important conditions for many modern technological systems operating on the terrestrial surface (e.g. power grids), in the atmosphere (aviation) and in the space (satellites, manned missions). Functions provided by spacecraft (communication, navigation) are critical for our modern post industrial society. Even the global financial industry requires spacecraft services both for communications and the time synchronisation of transactions, relying heavily of GNSS as a reference clock. With the exception of galactic cosmic rays, processes that occur on our nearest star, the Sun, drive space weather and may result in events such as magnetic storms, and drastic enhancements of the energetic particles fluxes in the near Earth space that are hazardous to the operations of technological systems. The advanced accurate forecast of these hazards is essential for the mitigation of their effects. The major advance of the current space weather forecasting capabilities is the major target of PROGRESS.

The overall aim of the project PROGRESS is to exploit the synergy of the complementary expertise available within the partner groups, the available spacecraft and ground based data combined with state of art data assimilation methodologies in order to develop an accurate and reliable forecast of space weather hazards. This aim is split in to the following objectives for the Project.

Objective 1: Develop a European numerical MHD based model that will enable the advanced forecast of solar wind parameters at L1. This will give a direct simulation connection between observed photospheric drivers and solar wind parameters at L1.

Objective 2: Use state of the art system science methodologies to develop new forecasting tools for geomagnetic indices and to assess the prediction efficiency of these new tools alongside those currently available to identify the most reliable techniques to predict the geomagnetic state of the magnetosphere, as expressed by geomagnetic indices, in relation to the solar wind input conditions.

Objective 3: Construct a new set of statistical wave models to describe the plasma wave environment of the inner magnetosphere that will accurately reflect the physics of the dynamics of the radiation belts under the influence of the solar wind. These novel wave models will lead to more realistic tensors of diffusion coefficients that are critical for physics based models of the radiation belts.

Objective 4: Incorporate forecasting capabilities into the physics based numerical model for low energy electrons IMPTAM that currently is able to provide a now-cast only.

Objective 5: Develop a novel, reliable, and accurate forecast of the radiation environment in the region of radiation belts exploiting the fusion between data based models for high energy fluxes at geostationary orbit SNB3GEO, IMPTAM, the most advanced model for high energy electrons in the radiation belts – VERB, and state of the art data assimilation methodology.

Objective 6: To combine the prediction tools for geomagnetic indices and radiation environment within the magnetosphere with the forecast of solar wind parameters at L1 and upstream of the magnetosphere to significantly increase the advance time of the forecast.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

During the first reporting period (2015-01-01 to 2015-12-31) the work performed toward the stated objectives and the main results are as follows.

Objective 1.

• Convert the Cartesian geometry Lagrangian-remap code Lare3d to spherical geometry and release through a source control management system. The new code is called SWIFT.
• In parallel to this to extend the AWSoM code to be able to run time-accurate and predictive by solving along selected field lines in the lower atmosphere.
• Details of these activities have been reported in deliverable D2.1.

Objective 2.

• A survey of existing operational models forecasting Kp, Dst, and AE has been carried out to determine the current availability of models. The results are available in deliverable D3.1.
• PROGRESS aims to develop a number of different models for the geomagnetic indices, based on different methodologies. Since these modes are driven by specific parameters it was decided to identify and collect relevant data sets for use within the project. Details of the data sets collected and how to access them are documented in deliverable D3.2.
• Key problems associated with the models include their validation and verification, and determination of the predictive efficiency. A survey on the methodologies that have been used to assess these criteria has been compiled. The results are contained in deliverable D3.3.

Objective 3.

• Data from THEMIS and Cluster were analysed to estimate the spatial distribution and quantity of measurements available in order to define the spatial grid to be used as a starting point for the spatio-temporal modeling of the magnetic field wave amplitudes using the Error Reduction Ratio methodology. The results were reported in deliverable D4.1.
• A set of databases containing the wave amplitudes and there location of observation for three types of wave emission commonly observed in the inner magnetosphere (chorus, hiss, and magnetosonic) were created for the satellites Cluster 4, and THEMIS A, D, and E. In addition, similar datasets from the satellites CRRES, DE, and Polar are available. However, there is currently no cross-calibration of these data sets with those of Cluster and THEMIS and so it was decided not to use them in the subsequent ERR analysis. The data sources and contents of the databases are described in deliverable D4.2.
• The wave amplitude datasets, together with auxiliary data sets containing observations of the solar wind (density, velocity, and pressure) and geomagnetic activity in the form of the geomagnetic indices Dst and AE were used as the output and inputs for the ERR systems analysis methodology. The results detail which of the input parameters have the greatest influence of the wave amplitudes. The results were reported in deliverable D4.3.

Objective 4.

• A new empirical model for boundary conditions for low energy electrons at L=6-11 dependent on solar wind and IMF parameters is now constructed based on the extensive analysis of THEMIS ESA (eV-30 keV) and SST (25 keV –10 MeV) data during 2007-2013.

Objective 5.

• The current set of NARMAX flux prediction models operated by USFD has been extended to cover all energies measured by the GOES 13 Energetic Particle Sensor (EPS) instrument. The results and stability of the model are still under long-term evaluation prior to making them live. These results were reported in deliverable 6.1.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

During the First Reporting Period the project PROGRESS has achieved a number of advances beyond the current state of the art.

WP 2

Currently Europe does not possess its own modeling/forecast system for the forecast of conditions at L1 based of our observations of the solar disk. Any forecasts currently made within Europe are based on the US supported WSA/ENLIL model. Within PROGRESS two models are being developed and will be later coupled to fulfil this gap.

AWSoM, the Alfven Wave Solar atmosphere Model belonging to the University of Michigan, uses magnetograms of the solar disk to drive a model of the solar atmosphere from the soar surface out to 25Rs. This code, which in its initial form was highly computationally intensive, has had to be significantly modified to ensure that it runs in real time using a reasonable level of computing resources.

From 25Rs to L1 the propagation of the solar wind is computed using a new, European based code called SWIFT, being developed at the University of Warwick. This code, based on the successful Lare3D model, has been modified to use a more suitable geometry and in such as way as to allow the inclusion of new physics.

This real time version of AWSoM will, when coupled with SWIFT, provide timely predictions of the solar wind parameters at L1 and hence enable forecasts of the state of the magnetosphere and its level of geomagnetic activity with a lead time of a few days, rather than hours that are currently available.

WP 3

As pointed out during the initial tasks of WP 3 there are currently a number of models that exist online to predict the evolution of the geomagnetic indices Dst and Kp. THe methodologies employed include empirical models, Neural Networks, Relevance Vector Machine (RVM) and Moving Average (MA), Auto-Regressive Moving Average (ARMA) and Nonlinear Auto-Regressive Moving Average with exogenous inputs (NARMAX) data driven models. These indices are used by many numerical codes to define the level of geomagnetic activity.

PROGRESS aims to build a suite of further models, based on recent updates to the above methodologies. New models for Dst and Kp are currently under development. Participants at USD are applying the state of the art NARMAX system identification methodology for the prediction of Dst and Kp whilst participants from SRI-Kiev have taken a new approach to structural- parameter identification investigating the combination of Robust NARMAX Models using geomagnetic indices and their local Lyapunov exponents. This development will result in two new classes of model: (a) a recursive, robust bilinear dynamical model; (b) the Guaranteed NARMAX Model (GNM). These models will be supplemented with current and new models from IRF based in different methodologies such as Neural Networks.

WP 4

Current wave models used for the characterisation of the interaction between plasma wave modes and electrons are based on based on the wave amplitudes organised by measurement location and the current geomagnetic state of the magnetosphere. This current system accounts for neither the changes observed in the solar wind that drive the whole magnetospheric system nor the evolution of the magnetospheric system itself.

To address these current drawbacks PROGRESS teams at LPC2E and USFD have performed studies using the Error Reduction Ratio to determine which of the solar wind parameters are most influential on the observed wave activity and also to determine the time lags involved within the system. This study provides a new basis on which to asses the interaction between plasma wave modes and electrons within the radiation belts, and will redefine the sets of diffusion coefficients that are required by numerical codes to properly incorporate these processes and hence provide a mode realistic picture of the electron environment within the radiation belts and enable the forecast of its evolution.

WP 5

The forecast of the fluxes of low energy (from a few 10's to hundreds of keV) within the inner magnetosphere is strongly dependent upon the boundary conditions used. The current boundary conditions used in the Inner Magnetosphere Particle Transport and Acceleration (IMPTAM) model are based on the work of Tsyganenko and Mukai in 2003 and imply a number of limitations. The goal of the initial task of WP 5 was to generate a new empirical model of the boundary conditions in the range L=6-11 that is dependent upon the solar wind conditions and IMF parameters, based on observations from THEMIS by participants at FMI. By using lagged and time averaged solar wind parameters it was possible to build in the historical evolution of the system. The results showed that the plasma sheet density depended most strongly on the southward component of the IMF averaged on time scales corresponding to the main phase of the storm (~6hr) rather than substorm growth phase time scales (~45m). The electron perpendicular temperature was found to correlate best with a southward IMF Bz averaged over ~45 minutes, the substorm growth phase time scale, and lagged by 30 minutes in contrast to periods of northerly IMF Bz which required a longer lag time and averaging period (~1 and ~2 hours respectively). Thus, this WP has generated a vastly improved characterisation of how the proton and electron boundary conditions vary in relation to the solar wind conditions.

Recent analysis of the hiss wave mode by participants at SIST and UCLA have refined the properties of this wave mode and thus improved our understanding of its interaction with the local electron population. This has led to the calculation and dissemination of a set of updated diffusion tensors related to this was mode and their inclusion within the VERB and IMPTAM models of the radiation belt electron environment.

WP 6

Satellites traversing regions containing large fluxes of electrons (such as the outer radiation belt) are subjective to surface and internal (deep) charging effects. These processes can result damage or even loss of crucial subsystems. Thus, an accurate nd timely forecast of these fluxes is essential for satellite operators in their scheduling of the tasks performed by such satellites.

In order to characterise and forecast the fluxes of electron radiation environment at geostationary orbit NARMAX models for the fluxes of electrons in the energy range 30-500keV have been developed and put online at USFD. These models complement the already existing models for higher (>800keV and >2MeV) electrons that were available before PROGRESS began. These systems models require large amounts of data and so their results are only applicable to geostationary orbit. The main advantage of such models is that they are mode accurate than current numerical models.

Related information

Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top