NuSTAR small flare of 12 September 2020

Basic information

NuSTAR observation time (orbit 8): 2020-09-12 20:22:00 - 21:07:00

Instrument	Wavelength regime	Cadence	Resolution/px
NuSTAR	SXR	20s	low
XRT Be-Thin	SXR	200s	1”
AIA	EUV	12 s	0.6”
STEREO-A EUVI	EUV	5 min	1.6”

Flare Location

AIA 335 image with rectangle indicating where fluxes were calculated — Rectangles describing regions where lightcurve fluxes were calculated. The 40" square box shown above was used for XRT, AIA, and STEREO-A (not shown). The 160" square box shown below was used for NuSTAR. Coordinates of box centers are given below.

NuSTAR camera A low-energy image with rectangle indicating where fluxes were calculated — Rectangles describing regions where lightcurve fluxes were calculated. The 40" square box shown above was used for XRT, AIA, and STEREO-A (not shown). The 160" square box shown below was used for NuSTAR. Coordinates of box centers are given below.

Flare location determined by first aligning XRT to AIA (~10” offset from location in XRT header) then finding the position of the center of mass of the 90% contour of the XRT submap around the flare area.

[<SkyCoord (Helioprojective: obstime=2020-09-12 20:27:01, rsun=695508000.0 m, observer=<HeliographicStonyhurst Coordinate (obstime=2020-09-12 20:27:01): (lon, lat, radius) in (deg, deg, m)
( 0.,  7.22972093,   1.50532432e+11)>): (Tx, Ty) in arcsec
(-883.70237088,  221.77460855)>]

Flare time evolution (AIA and STEREO)

3-minute integrated AIA 94 images

Source	Image	Base Difference	Sobel Filter
AIA 94
STEREO-A 195

Lightcurves

AIA at native cadence except for AIA 193 (3 minutes currently, will be updated later)

STEREO at native cadence (5 minutes).

NuSTAR images at 20s cadence (made by Sam), adjusted to counts.

Masks

Find regions of brightening/dimming by comparing the change in the flare region vs the quiet Sun region. A pixel is included in ‘brightening’ mask if the value of that pixel in the difference image (integrate flare X-ray peak minus integrated pre-flare) is >3x greater than standard deviation of the average flux difference in Quiet Sun (𝜎__QSdiff) in those same time ranges (details in code below). Likewise with ‘dimming’ mask, only the pixel value is less than -3x 𝜎__QSdiff. The total mask is sum of both masks in each channel.

#pre-flare time range (according to XRT, NuSTAR) 20:22-20:32
#select time ranges to integrate images
pfs=dt.strptime('20200912_202200','%Y%m%d_%H%M%S')
pfe=dt.strptime('20200912_203200','%Y%m%d_%H%M%S')
#peak time range (according to XRT, NuSTAR) 20:39-20:49
#integrate whole image. de-rotation doesn't shift more than ~1px, safe enough to do difference images like this
kfs=dt.strptime('20200912_203900','%Y%m%d_%H%M%S')
kfe=dt.strptime('20200912_204900','%Y%m%d_%H%M%S')

mask_total=np.ma.masked_inside(m.data,-3*t,3*t)
mask_plus.append(np.ma.masked_less(m.data,3.*t).mask) #1 means masked, 0 means un-masked, this is why it masked_less and masked_greater are used where they are
mask_minus.append(np.ma.masked_greater(m.data,-3.*t).mask) 

Masks for all AIA channels

Lightcurves with masks

DEM

DEMreg results using 35 temperature bins between logT= 5.7 and logT = 7.1, 3x3 pixel binning (masked pixels do not contribute to the mean)

Solutions DEMreg results

DEM evolution

Difference in DEM evolution relative to start of observation. NuSTAR low-energy count time profile and AIA 193 are shown (not to scale) on the x-z axis, and the AIA response functions (not to scale) on the y-z axis.

NuSTAR small flare of 12 September 2020 - orbit 8

Categories

Tags