Soft X-ray irradiance study
using Hinode/XRT synoptic composite images

Sample pages from the XRT Synoptic Composite Image Archive (SCIA)   (http://solar.physics.montana.edu/HINODE/XRT/SCIA/)

1. Introduction

• Hinode's X-Ray Telescope :
  • launched in 2006, September.
  • grazing incidence telescope with aperture size ≈ 35cm
  • 9 X-ray analysis filters
  • observing wavelength range : 6 - 200 Å
  • CCD : back-illuminated 2048 × 2048 pix
  • field of view : 2040 × 2040 arcsec.

• Full-Sun observations by Hinode/XRT --- Synoptic program ---
  • 2007.2 --- 2008.2 : 4 times/day with 2048 × 2048 pix
  • 2008.2 --- present : 2 times/day with 1024 × 1024 pix
  • 2018.2 --- present : additional synoptic images as HOP 349.
    
    
  • synoptic composite images : short/medium/long exposure images taken closely in time are processed into a single composite image to increase the dynamic range. These products are accessible from the XRT Synoptic Composite Image Archive (SCIA).
    (for detail, see Takeda et al. (2016) Sol.Phys. 291, 317)

• XRT synoptic composite image statistics
  
  Number of synoptic composite images obtained with selected filters
  (as of 30 September, 2021). Colored cells indicate the data used in this study.
  
  

  year	Al_mesh	Ti_poly	Al_poly	thin_Be	Alpl/Tipl	med_Al	thick_Be
  2006	0	2	366	138	0	3	0
  2007	1150	1129	1614	368	285	24	0
  2008	426	919	360	38	1	1	0
  2009	383	727	379	0	0	117	0
  2010	843	849	6	1	6	2	0
  2011	714	669	41	371	7	6	0
  2012	759	764	87	451	6	2	0
  2013	618	629	88	444	0	0	0
  2014	690	692	54	641	53	52	50
  2015	604	450	337	606	109	112	79
  2016	645	0	667	658	142	171	97
  2017	664	0	830	652	153	153	114
  2018	1214	0	1214	1239	144	144	127
  2019	1639	0	1644	1632	134	135	135
  2020	1593	0	1592	1602	126	126	135
  2021	716	0	744	721	65	63	63
  Total	12658	6830	10023	9562	1231	1111	792

2. Method of Analysis

• Step 1 : Derive filter-ratio Te and EM (isothermal corona assumed).
  • collect soft X-ray full-Sun signals (< 1.1R_⦿) from 2 different analysis filters.
    
    • 2008 - 2015.5 : Al_mesh and Ti_poly
      
    • 2015.6 - 2021.6 : Al_mesh and Al_poly
  • apply Filter Ratio Method to derive Te and EM.
    • traditional coronal elemental abundance (Feldman 1992) assumed.
• Step 2 : Calculate isothermal coronal spectrum using Te and EM.
  • CHIANTI atomic DB ver.9.0.1 (isothermal.pro)
  • traditional coronal elemental abundance (sun_coronal_1992_feldman_ext.abund)
  • CHIANTI's default ionization equilibrium data (chianti.ioneq)
  
  
• Step 3 : Integrate coronal spectrum over a certain wavelength range.
  • 5 -- 60 Å as the XRT irradiance.
  • 1 -- 8 Å for comparison with GOES/XRS
  • etc.

3. Calibration challenges and difficulties

• Visible stray light contamination --- caused by a crack developed on the pre-filter.
  
  

  event	date	open area fraction	note
  1st	9-May-2012	~0.1%	Ti_poly and C_poly filters started to be affected.
  2nd	14-Jun-2015	~3.3%	Ti_poly filter single use disabled.
  3rd	27-May-2017	~7.5%	Al_mesh filter started to be affected.
  4th	29-May-2018	~9.9%

  
  
  • Open/Ti_poly image in 2014 (after the 1st pre-filter failure).
    
    

    Before stray light subtraction.	Stray light image obtained in eclipse seasons.	Stray light subtracted.

    
    
  • How to correct
    1. Subtraction of visible light component from the X-day images.
    2. Modification of XRT's effective area function due to the open area of the pre-filter.
    (How to best correct the XRT effective area functions is still under discussion among the XRT team.)
  
  
• Growth of contaminant layers on CCD and X-ray analysis filters.
  --- caused by the evaporation of some material used in the telescope.
  
  
  • Previous calibration by Narukage et al.(2011) needs to be modified.
  
  

  contamination thickness on CCD [plotted by K. Yoshimura (MSU)]	thickness of contamination on X-ray filters from Narukage et al. (2011)

  
  
  1. The CCD contaminant layer thickness is now measured allowing the existence of the residual layer (since 2017).
  2. The additional growth of the FILTER contaminant layer after 2007 is likely.
     (Installation of the FILTER contaminant correction after 2007 is under discussion among the XRT team.)

4. Results

4.1 Mission-long XRT irradiance (5-60A), filter-ratio Te and EM

4.2 Comparison with GOES/XRS

• Comparison of GOES/XRS 1-8Å and Hinode/XRT 1-8Å irradiance.
  * GOES/XRS data source:
  • 2008 --- 2010 : Seamless GOES/XRS Archive (1-min averaged) in Yohkoh Legacy data Archive (YLA@MSU).
  • 2011 --- 2020 : 1-min averaged data from GOES15 reprocessed science-quality data (NCEI@NOAA).
  
  * XRT data : integrate the isothermal spectrum calculated from the Te and EM over the range, 1 to 8Å
  
  
  - XRT irradiance (1-8Å) is roughly 2 orders of magnitude lower!
  • XRT irradiance do not contain flare data.
  • difference in time resolution (XRT: a few image/day, GOES: 1-min.)
  • GOES/XRS response function (log energy channel)

• Comparison of GOES/XRS 1-8A and Hinode/XRT 0.3-10A irradiance.

• Scatter plot of GOES/XRS 1-8A and Hinode/XRT 0.3-10A irradiance.

4.3 Comparison with Yohkoh/SXT

• Yohkoh/Soft X-ray Telescope :
  • Grazing incidence telescope operated in 1991-2001.
  • Observing wavelength range: 3-30Å (cf. XRT: 5-60Å)
  • Irradiance study performed by Takeda, Acton & Albanese (2019) ApJ 887:225.
  
  
• Comparing the solar cycle evolution of Te and EM.
  
  
  • years 2008 though 2014 (rising phase of solar cycle 24)
    [Note: dotted lines in each plot is the equi-irradiance curve in the unit of W/m² in log scale.]
    
    
  • years 2015 though 2021 (descending phase of solar cycle 24)
    
    
  • all years 2008 though 2021
    
    
  • comparison with SXT (AlMg and thin_Al filter-ratio Te and EM)
    
    
    • XRT plot includes 2 solar minimum periods and 1 max period.
    • SXT plot includes 1 solar minimum period and 2 max periods.
    • SXT poorly detects coronal signals in solar minimum.
    • XRT filters sample the cooler portion of the coronal plasma than SXT filters.
    • XRT samples more dominant portion of coronal plasma because of its higher EM.
    • Significantly high frequency on XRT's lowest Te and EM histogram bin(s) might be showing that XRT observes the very bottom of the solar minimum condition.

4.4 Comparison with SDO/HMI

• Solar Dynamics Observatory/Helioseismic and Magnetic Imager
  • Data availability : 2010 through present.
  • Unsigned total magnetic field.
  • Measured within 60° from Disk Center, above 75G.
  • Data processed by D. Longcope (MSU).

• Hinode/XRT irradiance calculated for 5-60Å
  
  
• Previous study by Pevtsov et al. (2003) using Yohkoh/SXT and Kitt Peak magnetogram.

HMI vs. XRT irrad 5 - 60 Å	plot from Pevtsov et al. (2003)

• possible cause of the "knee":
  • data noise included in old data sets (SXT and KP mag.).
  • HMI vs. XRT 0.3-10Å
    ⟶ weak magnetic component that does not contribute to the X-ray brightness.

4.5 Use of spatial information

• Applying the Filter Ratio Method to images, instead of full-Sun integrated signals.

  2D-Te ⟶ EM weighted average Te	v.s.	full-Sun integrated Te
  2D-EM ⟶ total EM	v.s.	full-Sun integrated EM

  
  
  • EM-weighted Te is lower than full-Sun integrated Te.
  • total of space resolved EM is higher than full-Sun integrated EM.
  • possible indicator of thermal ununiformity of the corona.
  
  
  
• 3-temperature analysis:

  XRT 3-temperature movie period : 2008 Feb --- 2021 Jun. cool component (red) : below 1.5MK. medium component (green) : 1.5 to 2.5MK. hot component (yellow) : above 2.5MK.

  
  

  Solar cycle variation of cool/med/hot components
  

5. Summary