2022/07/14 Meeting notes

[Victoria-Samboco]

In today meeting we discuss about what we have been doing in the lasts days.

Because the MS provided by prof. Ian was not calibrated we was working in calibration (1GC) on the MS. So After discussing the results with Prof Oleg, the conclusion reached was to ask Prof. Ian to send the calibrated MS that would be the correct one to work with for what we are doing now. Prof Ian sent raw data thinking it was part of the tasks but no, what I'm doing starts with Selfcal.

The updating version of the Wsclean is done

Next steps:
Wait for the calibrated MS from Prof @IanHeywood and continue with the proceeds from Self_calibration split subtraction... until the Residual data imaging, to imaging the Sun.

We will also be doing 1GC in the meantime, so in addition to rephasing the MS to the sun with the 1GC MS, we will also be doing 1GC on the side so that I learn this tool as well, is also a good exercise for me too.

[Victoria-Samboco]

Greeting Prof @o-smirnov and @IanHeywood.

I apdated the wsclean to the last version of wsclean and when i run changecentre stil giving the same output (Old uvw: [0, 0, 0] (0) New [0, 0, 0] (0)) . In the last meeting it was said that the old uvw and new uvw are not supposed to be zeros.I don't know what the problem could be.

./chgcentre ~/solarkat/IMAGING/1645272078_calibrated/scans2/1645272078_sdp_l0_1024ch_J0333-2741_scan30.ms 0h1m54.71s 0d11m48.52s
A program to change the phase centre of a measurement set.
Written by André Offringa ([email protected]).

Processing field "J0333-2741": 03h33m37.33s -27d41m15s -> 00h01m54.71s 00d11m48.52s (57.8459 deg)

Old uvw: [0, 0, 0] (0)
New [0, 0, 0] (0)

Old uvw: [0, 0, 0] (0)
New [0, 0, 0] (0)

Old uvw: [0, 0, 0] (0)
New [0, 0, 0] (0)

Old uvw: [0, 0, 0] (0)
New [0, 0, 0] (0)

Old uvw: [0, 0, 0] (0)
New [0, 0, 0] (0)

Changing phase centre: 0%....10%....20%....30%....40%....50%....60%....70%....80%....90%....100%

[IanHeywood]

Perhaps the zeros that are reported are autocorrelations in the Measurement Set (where each antenna is correlated with itself). These are usually flagged as we generally don't make use of them. Their UVW coordinates won't be changed by a rephasing.

It's too late to see if the UVWs have changed because they will have been overwritten by chgcentre at this point, so I would just try imaging this Measurement Set and see if it's now pointing at the Sun.

[IanHeywood]

Initial UVWs are indeed zero:

In [1]: from pyrap.tables import table

In [2]: tt = table('temp.ms')
Successful readonly open of default-locked table temp.ms: 28 columns, 2655000 rows

In [3]: uvw1 = tt.getcol('UVW',startrow=0,nrow=10)

In [4]: uvw1
Out[4]:
array([[0., 0., 0.],
       [0., 0., 0.],
       [0., 0., 0.],
       [0., 0., 0.],
       [0., 0., 0.],
       [0., 0., 0.],
       [0., 0., 0.],
       [0., 0., 0.],
       [0., 0., 0.],
       [0., 0., 0.]])

[o-smirnov]

Does that correspond to ANTENNA1 == ANTENNA2?

In any case the real check is indeed to save the old UVWs (using numpy, perhaps), and compare them to the new UVWs. Too late for this MS, but next time you do the procedure @Victoria-Samboco, do that.

[IanHeywood]

Does that correspond to ANTENNA1 == ANTENNA2?

Yep

[IanHeywood]

It seems to change the UVWs despite the zeros. I'm fairly sure those are just reporting the autos in the first N-antenna rows.

Here's a before and after comparison:

In [7]: uvw1 = tt.getcol('UVW',startrow=55,nrow=10)

In [8]: uvw1
Out[8]:
array([[  0.        ,   0.        ,   0.        ],
       [  0.        ,   0.        ,   0.        ],
       [  0.        ,   0.        ,   0.        ],
       [  0.        ,   0.        ,   0.        ],
       [-14.99636078, -32.118492  ,   9.68685372],
       [ 25.33460278,  10.96100443,  -9.69320922],
       [ 40.33096356,  43.07949644, -19.38006294],
       [ 53.87532629, -17.90119659, -13.9915018 ],
       [ 68.87168707,  14.21729542, -23.67835552],
       [ 28.54072351, -28.86220102,  -4.29829258]])

In [3]: uvw2 = tt.getcol('UVW',startrow=55,nrow=10)

In [4]: uvw2
Out[4]:
array([[  0.        ,   0.        ,   0.        ],
       [  0.        ,   0.        ,   0.        ],
       [  0.        ,   0.        ,   0.        ],
       [  0.        ,   0.        ,   0.        ],
       [  1.09262004, -21.88178964,  29.50113495],
       [-18.15869721,   7.70675289, -21.60582552],
       [-19.25131725,  29.58854254, -51.10696048],
       [-53.99650203, -11.67638841, -19.15216611],
       [-55.08912207,  10.20540123, -48.65330106],
       [-35.83780481, -19.3831413 ,   2.45365941]])

[Victoria-Samboco]

In this case i will image the ones i have to see if its possible to see something there. Otherwise I will have to redo the split, subtration and before changecentre save the old uvw to compare later. For now I will try to see what i can get from imaging...?

[IanHeywood]

Hope to see the sun soon!

[Victoria-Samboco]

The file I'm using to run the tasks I just opened now and it's empty. When I try to open another and save it does not accept. And from the error I'm having, it looks like I ran out of space on the server, Is there a way to recover? The good part is that I still have another version of it in another folder, I just need to go back to editing if the same problem doesn't happen and that when editing doesn't accept to save it's giving an error.

[Victoria-Samboco]

Prof @o-smirnov Oleg can I ask for storage more space in aretha. I have only 103 G and Its full .Things are not working properly because I have no space. its not accepting to save any document, when I open the script iam working on its empty,...I had to delete some files and data that i was using previously but still the same thing. I tryed to go to another server but also its not acepting to copy files/data from aretha to the other server(simon).

[IanHeywood]

Sorry about this Victoria. I'm sure Oleg will sort this out but in the mean time there should be space on Nash if you have access to that machine ...

[o-smirnov]

I have also allocated you space on simon and Garfunkel. Aretha is too full -- you should switch to one of those machines.