We are trying to clarify the convention used in the tables provided in the notebook. In particular, we would like to know whether the throughput distributions are expressed in terms of photon counts or energy, and which magnitude system is assumed (AB, ST, or VEGAMAG).
As explained in the Introduction of the 204.4 tutorial, the throughput shows the percentage of incident flux at the top of the atmosphere that is recorded by the detectors as a function of wavelength. Does this answer your question?
Thank you @galaxyumi331 for the reply! Just to be absolutely sure for our synthetic photometry calculations, could you please confirm these two specific points:
Since LSST CCDs are photon counters (as reported in the document: https://pstn-001.lsst.io/fluxunits.pdf), do the throughput curves S( λ) in the notebook represent the photon-counting transmission (meaning we should include the 1/λ factor in the integrand when integrating an energy-based flux density like F_\nu)?
Can you confirm that the assumed magnitude system for the calibrations in this tutorial is the AB magnitude system? I read on this page (Filters section): Key numbers | Rubin Observatory
This PSTN-011 document might be very helpful for you to understand the LSST photometry.
As shown in Section 2.5, when the LSST pipelines integrate a specific energy flux density Fν, the 1/λ dependency is already absorbed inside the definition of normalized system response. Thus, no, you do not need to manually append an extra 1/λ factor.
In the LSST Science Pipelines, instrumental fluxes are first converted into calibrated physical fluxes, and the calibration converts counts into nano-jansky (nJy) units. Finally, it maps directly to the AB magnitude system.
The expected-but-not-measured throughput curves in the throughputs repo ought to look very similar to the bandpass curves you retrieve from the butler, I believe.
You can see how the rubin_sim photometry code handles these throughput curves for synthetic photometry here (examples of using that code here if it’s helpful to understand the Bandpass and Sed classes and their interaction).