1 of 5

ISWAT - International Space Weather Action Teams H3-01 SEP Model Validation

SEPEM Reference Data set Evolution

Piers Jiggens

Space Environments and Effects Section

28/09/2021

ESA ESTEC

ESA UNCLASSIFIED – For ESA Official Use Only

1

2 of 5

What is the SEPEM RDS?

  • The SEPEM Reference Data Set (RDS) has existed in some form since ~2009.
  • It was intended as a long-term contiguous and coherent data set for SEP specification modelling. It is the data set used for the SAPPHIRE model amongst others.
  • Initially it was only solar protons but now also includes solar helium
  • Version 1 used a simple cross-calibration of GOES fluxes against IMP-8/GME fluxes interpolated to the GOES detector geometric mean energies for each bin.
  • It was realised that the issue was more likely an issue of the shape of responses over broad energy bins folded with the radiation environment, i.e. the appropriate bin energies were not well confined.
  • Sandberg et al. (2014) @ SPARC investigated this and found another problem in the IMP-8/GME channels at intermediate energies ~20 MeV. The “Effective Energies” for each GOES/EPS were attained. Later these were re-computed by Heynderickx (DH Consultancy) and extended to solar helium.
  • Version 2.1 (2.01) includes background subtraction based on quite times for 3 quiet days either side of SEP events.
  • Version 3 will extend the data set to HEPAD energies following on from a bow-tie analysis and correction for rear-penetrating particles (Raukunen &Vainio @ Univ. Helsinki)

2

3 of 5

A preview of RDS v3 – Release in Q1 2022

Good agreement Case

3

4 of 5

A preview of RDS v3 – Release in Q1 2022

Divergence in v2 extrapolation region

4

5 of 5

Contamination Issue (S. Mutch, ESA)

  • We have long known that low energy GOES channels exhibit significant contamination from higher energies during SPE onset for hard, well-connected events.
  • This vary rarely impacts the peak value or significant alters the fluence for any given channel. January 2005 is probably the most extreme example.
  • If you use the RDS for SEP forecast validation it can be an issue.
  • We are employing an inverse VDA method to identify “exclusion zones” where low energy SEPs should not appear.
  • We then do a regression to higher energies to parameterise the contamination.
  • In practice we need to do this on the original GOES channels.
  • Correction will then be applied data set wide (you’ll only see it at a few points).

5