Note
This technote is not yet published.
This technote will describe metrics to evaluate the LSST survey strategy performance for detecting galactic microlensing and Tidal Disruption Events.
1 Introduction¶
In this tech note we discuss science metrics for two types of transient objects, fast microlensing events and tidal disruption events (TDEs).
Code for running these metrics and generating the plots in this note are in a jupyter notebook.
xxx-are we applying dust to TDEs?–yes!
For both of these metrics, we will compare their performance on two different simulatied surveys, baseline_nexp2_v1_6 and combo_dust_nexp2_v1_6. The main diffference between these simulations is combo_dust_nexp2_v1_6 avoids dusty regions of the galactic plane and covers the bulge as part of the wide-fast-deep survey.
2 Tidal Disruption Events¶
Figure 1 Example light curve shapes used for TDEs.
XXX–we generate a population of TDEs with properties drawn from XXX.
We apply dust extinction to the generated magnitudes and test three different detection criteria:
- pre-peak:
- some color:
- some color plus u:
The left shows the locations of simulated TDE events. The middle and right panels show the events that meet the pre-peak detection criteria. There is minimal difference between the two simulations, with 24% being detected in the baseline and 25% being detected in combo_dust.
Like before, only now showing the input population and the events that meet the most stringent detection criteria. The baseline detects 2% and combo_dust detects 1.97% of the TDE events.
3 Microlensing¶
Microlensing events are generated with a crossing times drawn uniformly from 1 to 10 days, and impact parameters drawn from 0 to 1, and distributed on the sky proportionally to the stellar density squared. We use TRIlegal Milky Way model to estimate the stellar density, including the LMC and SMC. By default, an r=22 magniude star is assumed to be the lensed source.
xxx—the criteria for detecting a microlensing event: 2 points detected before maximum, and 2 points detected post maximum.
By default, we assume the lensed object is an r=20 point source (flat SED) and demand that the amplification be detected at the 3-sigma level, e.g., a point on the light curve must have sufficient SNR to be detected at the catalog level as amplified from any previous observation at the 3-sigma level (e.g. XXX ).
Figure 2 Example light curve shape used for fast microlensing events. This is a 4.37 day crossing time with impact parameter of 0.73.
The images above show the input microlensing events along with the recovered events for the two simulations. The baseline recovers 15% of the input events, while combo_dust recovers 37% of the input. The extra bulge coverage in the combo_dust simulation results in many more microlensing events being discovered.
4 Extending to Other Populations¶
These transient metrics can be used as a templates for writing metrics for other transient populations. The basic steps for any transient metric are to 1) Generate a population of transients, 2) Define a criteria for a transient to be “detected” (and/or well-detected), 3) Use MAF and a simulated observing strategy to generate the observed light curves and check to see what fraction meet the detection criteria.
The source for the TDE metric and Microlensing metric are in the sims_maf_contrib repo.