Hard X-ray Pulsations via Gaussian Decomposition

From RHESSI Wiki

Revision as of 11:11, 29 January 2023 by Hhudson (Talk | contribs)
Jump to: navigation, search


Nugget
Number: 443
1st Author: Hannah COLLIER and Laura HAYES
2nd Author:
Published: January 30, 2023
Next Nugget: TBD
Previous Nugget: A slow HOPE with microwave context
List all



Contents

Introduction

ESA's Solar Orbiter mission carries the STIX hard X-ray imaging spectrometer. In its deep-space orbit, STIX has a very stable background and continuously observes the full solar disk, making it an excellent instrument suited to the analysis of time-varying behaviour on short timescales.

The hard X-ray emission from a solar flare often displays pronounced pulsations and time-varying signatures, with timescales ranging from seconds to tens of seconds. Such time variations are of great interest as they may be related to the fundamental timescales occurring in solar flares, such as energy release and particle acceleration processes, as well as magnetohydrodynamic (MHD) waves and oscillations in or around the flare site. These variations, often termed quasi-periodic pulsations (QPPs), have been observed in hard X-ray emission for many decades now (Ref. [1]). QPPs are present in flare emission at many wavelengths and on many time scales. Multiple periods of oscillations have been observed in the same flare and the fast-time variability has been shown to be present in both the impulsive and decay phase of flares, which challenges the standard solar flare model. To date, at least fifteen proposed models exist which attempt to explain the fundamental processes behind such time variability in flares (e.g. the earlier Nugget.) These models typically involve either a fundamentally periodic acceleration process, MHD or electromagnetic waves which modulate the observed emission or MHD waves which modulate the efficiency of energy release. However, it is often difficult to unambiguously determine the underlying mechanism(s) at play in a given flare.

The high-cadence X-ray observations from STIX provide a new opportunity to characterise fast-time variations and QPPs in flare emission. We have recently reported (Ref. [2]) a new method for detecting and characterising fast (1 s) timne variations in the hard X-ray (HXR) emission from solar flares. year of Solar Orbiters science operations.

Methodology

Our method assumes that the non-thermal HXR flare signal is made up of multiple time-series Gaussian profiles via a linear combination. This simple approach has several advantages, and we make use of machine learning to implement it. Figure 1 illustrates the steps involved.

Figure 1: The steps involved in the application of this method for an example M1.2 GOES class flare SOL2022-05-04, reference Earth time 15:16:12~UT. the fit is estimated as the standard deviation of the residual (difference between observed and fitted curve, normalised by the error on the observed data).

From the Gaussian decomposition, each flare pulsation can be identified and characterised. Thus we can derive key characteristics such as the periodicity, full width at half maximum (FWHM), time evolution, and amplitude and compare them quantitatively with models. The timing information obtained can also be used to perform localised imaging spectroscopy of distinct particle populations, a powerful capabililty with STIX data.

Results

An estimate of signal (quasi-) periodicity can be derived from peak times of the individual Gaussian components. Figure 2 shows the time of each Gaussian component against the peak \number for the SOL2022-05-04 example flare. The slope of the line fit gives an estimate for the period P5.55s. This value agrees with the period obtained using the fourier-based AFINO method (Ref. [3]) which gives P=4.96(-0.54,+0.66)s. This shows that STIX can detect variations on short timescales, which is something that was much more difficult on this timescale with its predecessor RHESSI.

Figure 2: Mean time of each Gaussian component against peak number for the example M1.2 GOES class flare of SOL2022-05-04. The slope of the line fit gives an estimate of periodicity in the signal. A stronger fit with multiple lines indicates that there is non-stationarity in the signal.

Conclusions

We have applied our method (Ref. [2]) now to four M- and X- GOES-class flares from the first year of Solar Orbiter science operations. This represents the first STIX characterisation and detection of rapid variations in the HXR time profile of solar flares with on timescales between 4-128 s. This method adds to the set of current QPP detection techniques, with the additional ability of characterising non-QPP, fast-time-varying signatures.

References

[1] "Sixteen-Second Periodic Pulsations Observed in the Correlated Microwave and Energetic X-Ray Emission from a Solar Flare"

[2] "Characterising fast-time variations in the hard X-ray time profiles of solar flares using Solar Orbiter's STIX"

[3] "A Large-scale Search for Evidence of Quasi-periodic Pulsations in Solar Flares"

Personal tools
Namespaces
Variants
Actions
Navigation
Toolbox