Adding support for EOVSA spectrograms

[binchensun]

EOVSA total-power and cross-power spectrogram, in FITS format, are now available under http://ovsa.njit.edu/fits/synoptic/

• Example for total-power spectrogram on 2020 June 26: http://ovsa.njit.edu/fits/synoptic/2020/06/26/EOVSA_TPall_20200626.fts
• Example for cross-power spectrogram on 2020 June 26: http://ovsa.njit.edu/fits/synoptic/2020/06/26/EOVSA_XPall_20200626.fts

I have created a very preliminary class for EOVSA spectrograms under my fork at: https://github.com/binchensun/radiospectra/blob/master/radiospectra/sources/eovsa.py. For now only the "read" method is implemented. The spectrogram can be displayed successfully. Please kindly let me know if anything needs to be considered to be consistent with the general structure.

Also a question: it looks like in LinearTimeSpectrogram the frequency axis has to be reversed (higher frequency first). If I comment out lines 44-45, it gives me an error. Is this intentional?

[samaloney]

@binchensun Could you make a pull request for this?

[samaloney]

Once #44 is merged this will be partially resolved. Currently not sure how to convert the time and frequency data in the fits files so it compatible with pcolormesh the plots works but are not totally correct.

@binchensun would it be safe to assume any gaps in time or frequency above certain threshold correspond to missing data and if so do you have an idea what those thresholds would be?

from matplotlib import pyplot as plt
from matplotlib.colors import LogNorm

from radiospectra import net
from sunpy.net import Fido, attrs as a

query = Fido.search(a.Time('2020/10/05 00:00', '2020/10/06 00:00'),
                                  a.Instrument('EOVSA'), net.PolType.cross)
query
<sunpy.net.fido_factory.UnifiedResponse object at 0x111946da0>
Results from 1 Provider:

2 Results from the EOVSAClient:
       Start Time               End Time        Provider Instrument PolType
----------------------- ----------------------- -------- ---------- -------
2020-10-05 00:00:00.000 2020-10-05 23:59:59.999    EOVSA      EOVSA   Cross
2020-10-06 00:00:00.000 2020-10-06 23:59:59.999    EOVSA      EOVSA   Cross


from radiospectra.spectrogram2 import Spectrogram
dl = Fido.fetch(query[0][0])
spec = Spectrogram(dl[0])
spec.plot(norm=LogNorm())

[binchensun]

@samaloney This is fantastic! I will look into this after the RHESSI Workshop. Also, let me tag @sjyu1988, who is now the primary developer of our software (suncasa). We have a "dspec" class for visualizing EOVSA total-power and cross-power spectrogram (but no query features). Let's compare notes and hopefully merge the efforts.