Gradients and threshsolding
With more analysis of the gradient and derivative functions it appears that the deriv function is for 1D analysis moving through each row of pixels one at a time and only evaluating the derivative using pixels on either side but not including the pixels above and below. The gradient function uses all neighbouring pixels however.
For thresholding of the magnetogram plots I tried fitting a gaussian to a quiet sun region and then using the 3 sigma value as a threshold. This seems to give a value of around +/- 19 Gauss as the non-quiet sun threshold
Though this leads to too much of the magnetic field being picked up (below)
Trying to measure the flux of the whole sunspot may not be as useful as thresholding for just the umbra anyway though as the field in the penumbra will be more inclined to the line of sight and therefore will not be as accurate as the readings of the flux in the umbra. It may be possible to use Johanna’s umbra/penumbra thresholds from the whitelight images to threshold for just the umbral flux.
Videos of AR 11515: (Ignore any brief horizontal lines, x2jpeg is producing these artefacts for some unknown reason..)
Magnetogram 30 June – 2 July AR11515
http://www.youtube.com/watch?v=S8QxGQ3HHyo
Gradient of magnetogram 30 June – 2 July AR11515
http://www.youtube.com/watch?v=LkpWGE488pE
Magnetogram July 1 15:07 – 16:31
Magnetogram Gradient July 1 15:07 – 16:31
Derivatives of magnetogram data
I have been experimenting with different procedures and colour tables to try and produce the best gradient map of the magnetogram data.
Here are some resulting graphs:
Magnetogram data:
Gradient using ‘deriv’ function and changing negative values back to positive (clipped data between 0.25 and 1 gauss/km):
Gradient using ‘gradient’ function (clipped data between 0.25 and 1 gauss/km):
Also after investigating the whitelight/magnetogram plots further it seems that there was no delta present in this AR at this time
Whitelight/mag plots for AR 11861
Whitelight and Magnetogram plots side by side for AR 11861
Find_delta
I began testing the find_delta code on AR 11861 (seen below) and did not find any deltas, by eye to me there do appear to be deltas around the 11th/12th but maybe they do not quite fit the criteria of the procedure. Below are some plots (ignore the black horizontal lines – x2jepg is producing these for some reason) of whitelight/mag contours. Deltas were not found for any of these.
Updated Graphs
I have run the code again using a log plot for the histogram seen below:
There appears to be a change in gradient around the -650 and 650 points. I took these values as the new threshold for the umbra/penumbra.
This gives thresholds which look like this:
Plotting just the umbra i produced the following plots of Area Vs. Time and Flux Vs. Time
The values of flux obtained seem to be consistent with those in the literature (~10^22 Maxwells)
Area Vs. Time for active region 11861
Histogram of 11861 field of view
Plot of field of view with positive threshold of 100 and negative threshold -100
Plot of area vs. time for polarities in active region 11861
Active regions 11861, 11864 and 11865 in magnetogram and 171Å
Attached is a video of the magnetogram and 171Å data up until Oct 14. There appears to be a winding up of the new flux emergence in 11861 as the large area of negative polarity enlarges during the 11 Oct this rotation may correspond to the slowly rotating swirl which we observed on 3 consecutive days at the SST (image attached below). This winding up of the magnetic field along with the mixed polarities in the active region may have contributed to the numerous flares produced by the region over the observing time period.
http://www.youtube.com/watch?v=UnZ03OWZt3A
Flux Emergence
Latest HMI magnetogram data shows flux emergence in sunspot 11856 over the last few days as can be seen in the video below
phalpintcd 
Recent Comments