Anomalous proper motion

[FabioRagosta]

anomalousPM_metrics.zip

[rhiannonlynne]

Hi @FabioRagosta , I know you are working on proper motion metrics, and yet I still don't really want to click on and download a zip file without any further description :)

[FabioRagosta]

Hi @rhiannonlynne, apologies, you are right I have missed any description. Let me provide the link to my github where you can find all the documentation https://github.com/FabioRagosta/LSSTmetrics/tree/main/metrics%20and%20documentation and let me add here some more description.

For each pointing in the opsim, the algorithm proceeds as follows:

• We generate a set of N = 500000 distances, d, for stars within a distance range $d\sim[0,120]$~kpc, according to the distance distribution in Binney (2008)
• We generate a set of m_g for each object simulated by assigning to each object an absolute magnitude following a simplistic single-age distribution from Percival et al. (2008) with stellar age ~ 6 Gyr. Stars in the magnitude range m_g ~[15,25] are retained;
• we select only the 1% dimmer population of the entire simulated sample;
• We associate each selected object with a Galactic component assigning them in proportion to the relative density of object in each component at distance, d;
• We assign a velocity v to each selected object drawing either from the assigned Galactic component velocity distribution or from an unusual distribution;
• Finally, we calculate the tangential velocity v_tan, which represents the projection of the velocity vector on the sky plane, ie, its proper motion, as:
  pm=\frac{v_{tan}{4.74~{d}} arcsec/yr.

We propose a simulation of proper motion measures that aims to describe a known population, which is driven by a velocity distribution function such as the one in Binney 2009; and an unusual population which is driven by a different velocity distribution function. By the use of this mock data we want to measure the fraction of the unusual population that can be identified as a True Novelty.
The metrics grouped in zip file are:

• LSmetric: the aim of this metric is to measure the fraction of detectable unusual object whose proper motion is measurable, considering the comparison between the likelihood distribution of the unusual object with respect the usual object proper motion distribution ;
• TPMmetric: the aim of this metric is to measure the fraction of detectable transients whose proper motion is unusual, this is performed estimating the fraction of transient with at least two observations that are moving with an unusual proper motion;
• reducedPM: the aim of this metric is to measure the accuracy of a proper motion measure in a structure to be able recognize unusual stream, this is performed by estimating the ratio between the error ellipse area in the reduced proper motion diagram- to analyze the accuracy of the proper motion measurements- and the fraction of measured pm.

The case of unsual stream is tackled a bit differently with respect the method exposed up to now. Indeed, we use the data of Sagitarius A in Carlin et al 2012 as a toy model to measure the possibility to detect the unusual stream they detect in the dwarf galaxy, considering to possibility:

• having a galaxy farther away;
• having a structure with stars in different location of the CMD.

I hope this can be useful to have a clear frame of the work.
Again, apologies for having missed the entire description.

Fabio.