NOTE THAT THE SCRIPT USED TO CREATE THESE PLOTS WAS UPDATED ON OCTOBER 16. 
PLOTS FOR DATA TAKEN AFTER OCTOBER 12, 2024, WERE GENERATED WITH THE NEW 
VERSION OF THE SCRIPT.

THIS IS HOW TO READ PLOTS FOR DATA TAKEN AFTER OCT 12, 2024

Each directory contains figures made from night-time autocorrelation data collected over 
a period of about 10 hours and extracted every 10 minutes. 

The antenna status is based on autocorrelation spectra recorded between 30-80 MHz. 
Before analyzing the spectra, RFI are removed by applying a median filter with a kernel 
size of 25 channels. The choice of 25 channels is empirical. 

Each directory should contain 6 figures plus a text file summarizing the antenna status 
for that night. If the directory is empty, it means that my scripts failed. The most 
common reason for this to happen is that some subbands are missing in the ms files. 

Below I explain what each of these figures show: 

- [date]_antenna_status_xx.png and [date]_antenna_status_yy.png contain antenna 
flags for the xx and yy polarizations. The vertical axis indicates the antenna 
correlator number, while the x axis indicates the number of the ms file. Files are 
spaced by 10 minutes. The colors have the following meaning: 
1. GREEN: the antenna looks good
2. BLACK: the recorded power is less than 2 Correlator Power Units (CPU)
3. RED:  the recorded power is above 40 CPU. These signals are saturated. 
3. BLUE: the flux shows large time variability as compared to the temporal profile of 
the median flux 
4. YELLOW: the shape of the spectrum differ substantially from the median spectrum
5. PINK: sometime even good antenna have bad moments. A pink box indicates a moment 
in time when the autocorrelation spectrum is substantially different from the median 
spectrum of that specific antenna meausured over the entire time interval. 

Notes: 
- Black antennas are usually "Dead" signals, though some show some flux. 
- Yellow antennas are not necessarily "bad" antennas, they might just have 
a spectral shape that differs from the median. However, sometime yellow antenna 
show spectra that are severely corrupted by RFI
- Blue antennas are generally really bad and might affect other antennas
- Vertical pink lines indicate times at which something happened that affected all the 
antennas. Isolated pink pixels are generally ok. 

- [date]_dynSpec.pdf contains dynamic spectra. The title of each plot reports
the antenna name, correlator number, polarization, and the antenna Flag 
(with the expection of PINK flags). These plot should be self explanatory.

- [date]_medFluxVsTime contains the time profile of the median flux measured 
between 30-80 MHz. The solid lines show the median flux of the antenna listed in 
the plot title, while the dashed line show the global median calculatued on 
all antennas. The plot title also list the antenna flags.

- [date]_normSpectra_xx.pdf and [date]_normSpectra_yy.pdf show normalized antenna spectra. 
In particular, blue lines show spectra for each antenna at each time, normalized by the 
median spectra of the same antenna calculated over the entire time interval (red curve). 
The black curve is the normalized global median, i.e., the median spectrum calculated over all
antenna and are times. The normalization values for the red and black curves are reported 
in the legend.