AGN chapter presentability

[drphilmarshall]

@ohadshemmer @nielbrandt @tanguita

Happy New Year! Timo and I just looked through the AGN microlensing section together, in #250. Below are my notes from looking through the rest of the chapter - hope they are useful! The main things I see missing are Figures of Merit, and the discussion of their likely dependence on your diagnostic metrics - but in most cases I think you are close.

• [x] In the AGN census section, what are you trying to measure? Is it the luminosity function? If so, you should be able to come up with a global LF accuracy that could serve as a Figure of Merit for this science case.
• [ ] In the metrics subsection, you give four diagnostic metrics, some of which (eg astrometric redshift) might be dependencies of the FoM - but it will have others as well (eg photometric redshift, depth, etc). It could be worth adding words to the effect that you are investigating a FoM and its dependencies, and lay out the plans you have for this investigation.
• [x] Won't the number of z>6 quasars you find depend on the useable sky area as well? Or is color, variability and lack of proper motion enough that you have no identification/confusion problems?
• [x] Did you look at any other OpSim outputs as well as enigma1189 for the z>6 quasar count?
• [ ] The disk variability intro is very nice, and seems to be leading up to a Figure of Merit based on the PSD measurement - but then the metrics section doesn't go the next step. Do you plan to create a fiducial mock sample of objects, with some hyper-parameterized relationship between PSD parameters, and then try to recover those hyper-parameters? Figures of Merit are typically hyper-parameter accuracies.
• [ ] It sounds like your OpSim analysis plan is to identify runs that have a variety of cadences in the u and g-band, and compare them using your figures of merit - if this is true, you shoudl write this!
• [ ] For now, your discussion section could be used to briefly discuss what risks are involved (eg itll be a lot of work to do the simulations, the results will depend on how good you are at data analysis, etc). What do you think?
• [x] What does it mean for PRM to "work best"? Are you interested in the mean of the lag measurement accuracy across a realistic sample? Or something else? If you can pin this down you will have your Figure of Merit for this science case.

[tanguita]

Hi @ohadshemmer , can you check whether we can tick the first three check boxes so we can move this issue (the other four check boxes) to a different milestone? I guess there is no time to for the fourth checkbox regarding the newer metrics and the AGN census section.

[ohadshemmer]

Hi @tanguita , I think we can check boxes #1, 3, 4, and 8 from above. The others will have to wait for the "Future Work" section.

[tanguita]

Hi @ohadshemmer ,

Done... moved to the Tucson project and community meeting 2015