1. LSST : Follow-up des SN proches
LSST : Follow-up des  SN proches ? Pierre Antilogus LPNHE-IN2P3, Paris LSST-France March 20th 2017

2. LSST concept : 1 telescope + 1 instrument + 1 observation plan
6-band Survey: ugrizy 320–1070 nm
Survey(s) Area (with 0.2 arcsec / pixel) 
Main : at least 18,000 square degrees to an uniform depth
Other : ~10% of time ~1h/night (Very Deep + fast time domain + special regions : ecliptic, galactic plane , Magellanic clouds)
Total Visits per unit area and Visits per filter (Main survey)
Image Quality
Mean seeing at the site is ~ 0.7 arcsec
PSF FWHM < 0.4 arcsec (no atmospheric seeing).
PSF Ellipticity < 0.04
(referenced to 0.6 arcsec FWHM circular Gaussian)
Photometric precision:
0.01 mag absolute; mag repeatability & color u g r i z y
Nb Visit 56 80
184
160
1 visit mag
23.9
25.0
24.7
24.0
23.3
22.1
10 year
26.1
27.4
27.5
26.8
24.9
More than visits & 5.5 x106 exposures
following the sequence:
15 s pose + 1 s shutter + 2 s read + 15 s pose
+ 1s shutter + 5s new pointing as reading
 Points to new positions in sky every 39 seconds
Number of visits per night : ~ 1000
Universal Cadence Strategy for Main Survey
Revisit after minutes
Visit pairs every 3-4 nights

3. LSST: Stage IV Dark Energy Experiment
LSST complementary techniques to constrain Dark Energy :
Weak gravitational lensing
Baryon acoustic oscillations
Type 1a supernovae
Statistics of clusters of galaxies
Remark : LSST Key properties to remove instrumental/atmospheric signature : > 800 exposures of each field
Stage IV criterion defined in terms of the inverse of the error ellipse in the wa-w plane.

4. SN Ia redshift domain & LSST
LSST deep fields
+ IR from space
(Euclid or WFIRST)
Included in LSST
Cosmology Baseline
LSST
deep fields
HST
LSST Shallow :
will request to run part
of LSST Main survey in
a accelerated way
SNLS
Detection in LSST Main
Follow-up TBD
SDSS
Ultra-Low-z
LSST can also changes what can be done in these redshift domains
but it will requests other instruments for the light curves measurement

5. SN Ia Z>1 : LSST + Space (Euclid or WFIRST )
“low-z” (LSST Shallow)
“low-z” (LSST Shallow)

6. SN Ia Z< .1 : Level arm
From the ground , more than increasing “for ever” the size of the default sample (LSST Deep + LSST Shallow) the best thing to do :
Increase the sample size in the low redshift domain ( z <.1 ) .
P.Astier 2016)

7. SN Ia Z< .1 : what other cannot probe !
J.Guy (Moriond 2016)
Beyond “FOM”, SN <0.1 are unique probes :
BAO & weak lensing limited by cosmic variance
Can bring unique information on the most resent (last 10% ) universe expansion
Velocity correlation at z<0.1σ8+γ constraints

8. SN Ia Z< .1 : Players …
The one that will make progress on the subject for z<0.2
ZTF, PanStarrs, DES, SNFactory
But even ZTF will not close the low redshift subject before LSST .
Large ( >> 100 ) sample of Low redshift light curve of “cosmo quality” request time ( yearsss for z <0.05) , lots of work (photometry is hard ) and dedicated/efficient instrument for what ever follow-up (spectro ? Photo ? )
LSST will discover them all but will follow “none” .
 This follow-up is a big effort , no good plan today

9. How many SN Ia per year ?

10. LSST detection , follow-up with ? (1/2)
The “rolling cadence” discussion in LSST today , aims to measure SN Ia for z ~ 0.4, above 0.5,0.6 , the deep field will do the job , below 0.x ( 0.2 ? 0.1 ? 0.05 ? ) , there is a need for a dedicated follow-up : LSST will detect them early ( but not in a given filter , as a same filter on a given field is used at a frequency of ~10-15 days , but a same field is visited each 3 days )
Follow-up : Photometry :
goal ?
Light curve (10 ? 15 points ? Precision per points ) in 3 colors ( gri ? ugr ? BVR ?) ( + a few points from LSST in “other colors” )
example :
To follow ~ 1200 SNIa / year ( all to z <0.1 ) :
1 night ~ 8 h , 1 year 300 nights : 2400 h  2 h per SN Ia (overhead & calibration included )
With 0:30 -1:30 min / filter on a 2m , does the job for 1 point
question : mirror size vs overhead vs 3 filters at once
1 m with 3 filters at once is confortable
but 3 x 50 cm may be ok too if we stay at z<0.1
Remark : large unknown on the expected stat at low z .

11. LSST detection , follow-up with ? (2/2)
Follow-up Spectro :
goal ? ID ? Spectra at max for like like classification ?
Telescope :
Z ~  2m at least …up to z~0.1 with a MUSE like spectro ?
Z < .2  4 m
Should we consider to do SN II for z<0.05 (with Expanding Photosphere method ) ?
How to select the object to follow ?
Should we take early spectra ( heavy / time expensive )
Accept to take a few early colors to trig the light curve follow-up
Near max , take the decision to take a spectra / proceed with the follow-up

12. White book “ LSST nearby SN Ia “
Covering all science aspect / interest for different z domain ( < 0.05 , 0.1 , 0.2 )
Evaluation of what is needed to optimize the distance measurement :
Spectro ? (like to like , ID , type II ..)
Calibration
Filters / light curve (nb point / nb photon )
Cost of the identification / versus time lost doing it
Considering the different setup possible :
Spectro : where ? With which type of instrument
Photo : 1 m ? 50s cm ? Dychroic / 1-3 camera ?
Who will read it ? Agency ( funds ) , ESO (in kind ? ) ..
Who will write it ? : LSST-France could/should start / do a first version … but at some LSST-collaborators ( we should see big or we will fail again )
When ?  Now : needed to build something / have access to the facilit(ies)…

13. LSST & Optical transient
LSST will identify a large sample of variable objects , dominated by variable stars : 10k / visit (in average ) , 40k /visit (at max) , it will provide (10 7 in the galactic plane ) alerts per night .
LSST will publish these alerts within 60 seconds.
But the LSST database (object surrounding , history of the object / field, multi-color light curve ) will be the key to classify these alerts and trigger spectroscopic follow-up.
Example: classifications obtained by COVET-CFHT, on a test run close to LSST conditions

14. LSST & Transients
PTF (~Today) & LSST (~2022) supernovae rate (Rau et al. (2009))

15. LSST & Optical Transients : an unique way to probe/survey the universe
Hubble diagram for # directions :
LSST will be able to probe the isotropy of the Dark Energy properties . For example the large SNIa statistic will allow to build SNIa hubble diagram for different directions in the sky.

16. End of Presentation