Figure 4 (a):

data is saved in Figure 4(a) sub folder. Each file is organised in columns as follows:

temperature (C)	photon counts on P 1	photon counts on P 2	photon counts on P 3	photon counts on P 4	photon counts on P 5	photon counts on P 6	photon counts on P 7	photon counts off P 1	photon counts off P 2	photon counts off P 3	photon counts off P 4	photon counts off P 5	photon counts off P 6	photon counts off P 7	laser power (mW) P 1	laser power (mW) P 2	laser power (mW) P 3	laser power (mW) P 4	laser power (mW) P 5	laser power (mW) P 6	laser power (mW) P 7	


The columns of intersts are temperature (c), photon counts on P 1, photon counts off P 1 and laser power (mW) P 1

each files is taken at a different laser power.



Calibration:

The temperature column referes to the temperature of the etalon and is realted to frequency through the calibration:

f = a1 + a2 T + a3 T^2 + a4 T^3

where f is the frequency and T is the temeprature. The coefficeint a1, a2, a3, a4 are found in file "etalon_calibration_parameters.csv"


Processing the data:



points where the laser power was greater than 1% of the average laser power are discarded. 

the data with no microwaves (photon counts off P 1) is binned into 1 GHz binned. This is fitted with a single lorentzian. The amplitude of the lorentzian are the purple triagles in fig 4 (a)

To study the microwave effect, the difference photon counts on P 1 - photon counts off P 1 is taken. Again this is binned into 1 GHz bins. The binned data is fitted with the sum of three lorentzians, one for each feature in the data. The amplitudes of these features give the other points in Fig 4(a).



Model:

The solid lines in figure 4 (a) are quadrative fits to the plotted data. 

#############################

Figure 4 (b)

data is saved in Figure 4(b) sub folder. Each file is organised in columns as follows:

temperature (C)	photon counts (on)	photon counts (off)	power (mW)

each file is a at a different microwave power. The microwave power is saved in the first line of the file.


Calibration:

The temperature is calibrated in the same way as Fig. 4 (a)



Processing:

points where the laser power was greater than 1% of the average laser power are discarded.

the difference between the photon counts on and photon counts off columns are taken. The data is binned into 1 GHz bins

The data is fitted with a sum of three lorentzians, the amplitudes of these lorentizans are the plotted points in figure 4(b)



Model:

The solid line in figure 4(b) is from the model described in the text.

The code for the model was written in mathematica. The code is in this folder.

The parameters used in the model are in the folder for figure 3.

