i4ds / ecallistovalidation Goto Github PK

The reference data for testing SNR is two-fold. First, the data should be compared to observations from a reference instrument. We consider currently a radio spectrometer operated at the Nançay radio station, ORFEES , as a reference.
Therefore, starting from this reference, the station testing needs to be organized in such a way that the reference can be transferred to stations at more western longitudes, eventually reaching those that observe while the reference station is in the night.
The data availability from ORFEES to cross-calibrate e-Callisto is limited. However, we use the available data of this station to determine which radio bursts in which time intervals should be used in the e-Callisto database. The stations that have been observed during these intervals are selected and the absolute flux (in solar flux units SFU) of the radio bursts is computed.

• Understand how to make a spectrogram from the ORFEES data.

• Generate the spectrograms for each STOKES parameter, with frequencies and time information, pack it into a spectrogram

• Create an ORFEES data reader that returns a spectrogram, just like the one from e-callisto.

• Correct the position of Orfess.

• Correct the time axis and its format when plotting it.

• Rebinning the ecallisto data to be the same as Orfees (https://scipy-cookbook.readthedocs.io/items/Rebinning.html)

• Make the Spectrograms of Orfees and eCallisto have the same shape.

System mit Sternen implementieren z.B:
1 = schlechte Station
5 = gute Station

Input => date Interval
Output => Stationen mit Sternen *

These values should be able to use:

• SNR & STD(see fits tables) => if SNR is large & STD is small => good (check if vice versa is good)

• Calculate the SNR & STD.

• Create a Function to convert the Values to Stars.

• Create a function that takes the start_time & end_time as a parameter and returns a list of stations with the rating system.

• Making the rating system more efficient.

• Split Observations Time in DB => start time & end time.

• Specify the amount of included instruments in the rating function.

• Add the STD to the ranking system.

• Change the format of the start time & end time in DB.

• Correct the ephemeris times and the entire DB.

• Write a function to call a list of Spectrograms.
• Interpolation of all array. data for all Spectrograms in the List.
• Calculate the factor.
• At the end they should have the same length.

Create all 10'000 plots from September 2017 by using the python code constbacksub function (based on the IDL code from André).

The end result should be like that:

First column: Original data plot
Second column: background subtracted plot (based on the constbacksub function from the IDL time)
Third column: histogram

At the end, we will check which background subtraction code is more suitable for the future (Code from radiospectra or the code from on André which was first written in IDL)

• clone ssa-ecallisto wp1 branch
• call function get_fits_data(fits_path) (just copy paste the code from test.py into your Jupyter Notebook) => data for first column
• call function "constbacksub(data)" from the constbacksub.py & "elimwrongchannels()" from elimwrongchannels.py (See comment below) => data for second column
• Set our custom colormap to the plots (See comment below!)
• calculate histograms => third column
• show / plot all 3 plots in a row (as describe above)
• Jupyter Notebook to pdf

• Download the files from the Server.
• Calculate the Coordinates and save them as an Excel file.
• Read the Coordinates from the excel file.
• Plot the files.
• convert the pixel_tp_world (Physical Coordinates).
• records Videos for the Obs Files
• Save the data, data_times, and Coordinates as Numpy files.
• Correct the Time axes.
• Fix the Vertical Lines in the Plots while Rotation.
• Create a Folder for each Obs and save their plots, Numpy Files, and records.

@Beckicious said:
Instead of interpolating every line of the spectrogram by itself use the following function to interpolate the whole 2D array.
https://docs.scipy.org/doc/scipy/reference/generated/scipy.interpolate.interp1d.html#scipy.interpolate.interp1d