I’m in the LSST’s “Transients and variable stars”, in the microlensing subgroup. I’m mailing you because we are looking for coordinates and parameters of the microlensing events in DP0 . When we did not find the MLT true light curves in the truth table, following a conversation on the Rubin forums, we tried to run the function write_star_variability_stats using the code write_star_variability_stats.py with stars_db_file='global/cfs/projectdirs/lsst/groups/SSim/DC2/dc2_stellar_healpixel.db’. We ran into this issue: ModuleNotFoundError: No module named ‘lsst.geom._geom’, but the file “_geom.py” doesn’t exist in the lsst/geom.
Also, inspecting the codes I come up with several questions or doubts about where to keep searching:
I noticed that in this variability models methods used to apply in the light curves exist the “applyMicrolens” register_method separately from “MLT”, while this tutorial explains that microlensing events are included into MLT. There is any intermediate step where these types are merged? Maybe I should inspect that part.
Should we be searching in any other table of CatSim?
Should we understand how star_db_file (dc2_stellar_healpixel.db)'s varParamStr column is created?
Hi Karen, thank you very much for your questions. We are discussing them with other members of Rubin and the LSST DESC involved in DC2 and DP0.2, and we will get back to you soon. In the meantime, everyone is welcome to chime in with thoughts or potential solutions in this thread.
I’m not an expert on the legacy lsst_sims code that was used for the DC2 simulations, but from my reading of the MLTflaringMixin comments here, I would say that the MLT objects in DC2 are not related to microlensing events and so may not be of interest to you. And although that delegate tutorial notebook mentions microlensing events, that description looks like it is based on a sentence in the DC2 Survey paper, section 5.3, where micro-lensing events are given as an example of non-periodic variables. However, based on the contents of that dc2_stellar_healpixel.db file, I would conclude that no actual microlensing events were included in DC2.
When Jim and I were discussing this, he also suggested the possible use of synthetic source injection. Another member of Rubin’s Community Science team, @jeffcarlin, has produced a notebook with examples of how to inject variable sources. It can be found in the following repository:
Hi Karen, thanks again for this interesting question on microlensing events in DP0. I just wanted to check with you if Jim and Andres’s feedback was able to address your questions.