How to convert DN unit of AIA images to SI (or cgs) irradiance values?

[Harsh Mathur]

Dear SunPy Community,

I hope you are doing well. I have a question regarding AIA image data. Specifically, I would like to convert the data from DN (Data Number) units to irradiance in SI or CGS units.

I understand that the units of AIA images (for ex: 171) are in DN, and there is also an instrumental degradation factor to account for. Please guide me on how to perform this conversion and account for the degradation factor in the process.

Any advice or pointers to relevant documentation would be greatly appreciated!

Warm regards,
Harsh Mathur
harshmat...@gmail.com

“Perseverance is the hard work you do after you get tired of doing the hard work you already did."

[will.t...@gmail.com]

Hi Harsh,

The AIA instrument team may have a slightly different perspective, but I will try to weigh in from the user perspective.

tl;dr: Correcting for degradation is straightforward. Converting from detector to physical units is not. Out of curiosity, what is your use case for wanting to change units here?

There are two different questions here:

1. Regarding the degradation correction, this is just a matter of applying the appropriate (scalar) degradation factor. You can do this with aiapy: https://aiapy.readthedocs.io/en/stable/generated/api/aiapy.calibrate.correct_degradation.html#aiapy.calibrate.correct_degradation.
2. Regarding converting an AIA intensity to units of irradiance, I assume you mean going from units of DN pix^{-1} s^{-1} to units of erg cm^{-2} s^{-1} sr^{-1}. Recall that the AIA channels are narrowband and multi-thermal such that the intensity in any given pixel for a particular channel is the summation of many spectral lines covering a range of temperatures. This makes a straight conversion between instrument units and units of irradiance impossible, at least without some assumptions. You could go about this a few ways:

• The most rigorous way of doing this would be to first compute a differential emission measure (DEM) by inverting the integral over temperature of the DEM multiplied by the appropriate temperature response function (see Section 2.5 of this paper). Once you have the DEM, you can combine that with the contribution function (or the combination of many contribution functions) to get something in units of irradiance. 
• Alternatively, though far less rigorously, you could assume that any given AIA channel is dominated by contributions from a single atomic transition and simply divide the AIA intensity in instrument units by the wavelength response of that channel (see this aiapy gallery example for how to compute these) at the wavelength of the chosen transition. However, this is difficult due to the fact that most channels are dominated by multiple lines.
  

I hope that helps!

Best,

Will