Early Light Curves: Now and in the Future

Anthony Piro George Ellery Hale Distinguished Scholar in Theoretical Astrophysics (Carnegie Observatories, Pasadena)

Supernovae: The LSST Revolution - Northwestern – June 1, 2017 Why Early Light Curves?

Observations of early light curves during the first ~days to weeks after explosion provide key information about SNe and their progenitors 1. “Shock cooling” measures the radius of the exploding 2. Interaction with a companion constrains progenitors models 3. Probes circumstellar material reflecting activity of the progenitor right before death What kind of observing strategies will allow LSST to address these issues? The first signature of a supernova Shock propagating through star

tsh ~ δr/v

tdif ~ τ δr/c

tdif < tsh

τ < c/v ~ 30

Shock breakout! Following Breakout is “Shock Cooling”

Piro, Chang, & Weinberg (2010) • Early UV/optical dominated by cooling of shock-heated material • Luminosity proportional to initial radius R c E L 0 ⇠  M Core-Collapse Shock Cooling Emission

Nakar & Sari ‘10 (also see Chevalier ’92) Shock Cooling Proportional to R0

Expands and cools

Expands and cools

…but not as much Rising Light Curve of SN 2011fe

• No detection of shock cooling • Upper limit constrains progenitor radius <0.02

Rsun 1st direct evidence that Type Ia SNe are from white dwarfs Bloom et al. (2011) Companion interaction in Type Ia LCs? Shappee, Piro, et al. (2015) • Early light curve also provides constraints on companion radius (Kasen ’10)

• Constraints depend on explosion time! Companion constraints from Kepler

• Three exquisite light curves from Kepler by Olling et al. 2015 (only 2 shown here) • No evidence of interaction with a companion Evidence for companion interactions?

Cao et al. (2015) Marion et al. (2016)

If true, this would be support for single degenerate scenario.

Not a normal SN Ia (see Although see Shappee, Piro, McCully et al; Foley et al) et al. (2016) for a contrary opinion Radius Upper Limits for Stripped SNe

Cao et al. (2013) using Corsi et al. (2012) Piro & Nakar (2013) Piro & Nakar (2013) R*=4Rsun

R*=1.5Rsun

R*=0.3Rsun

• PTF 10vgv, SN Ic • PTF 13bvn, SN Ib

• R* < 3Rsun • R* < few Rsun Shock Cooling from Type IIb SNe

• Tenuous, extended envelope (~300Rsun) leads to distinct shock cooling signature (Woosley et al. ’94, Shigeyama et al. ’94, Blinnikov et al. ‘98) • Consistent with yellow supergiant progenitors (e.g., Van Dyk)

SN 2011dh by Bersten et al. (2012) SN 1993J, Piro (‘15), Nakar & Piro (‘14) Recent Results on Superluminous SNe

• Type I Superluminous SN with peak at M = -22 (powering source still controversial, see Kasen & Bildsten, etc) Nicholl et al. (2015)

• First peak at M = -20, brighter than a Type Ia! What is it? Another Double-peaked SLSN

Multi-band light curve well fit by extended material (Piro 2015):

~ 400 Rsun

~ 3 Msun ~ 6x1051 erg model might work as well (Kasen, Metzger, & Bildsten ‘16)

Smith, Sullivan, et al (2016) Are all SLSNe double peaked?

• LSST well-suited to find more SLSNe (low rate and bright) • First peak is not that short lived for LSST • Understanding the first peak’s occurrence rate and diversity is key for unraveling this mystery Nicholl & Smartt (2016) SuperNova Explosion Code (SNEC) Morozova, Piro, et al. (2015) • Led by Viktoriya Morozova • 1D Lagrangian hydrodynamics and radiative diffusion • Bolometric light curves and specific wave bands • Open source with growing usage (Taddia et al ‘16; Nagy & Vinko ‘16; Szalai et al. ‘16; Eldridge, Wheeler; Petcha; and more) http://stellarcollapsex.org/snec Type IIb SN 2016gkg

• Detailed modeling of first peak to constrain circumstellar structure

• Need ~0.02 Msun spread out to a radius of ~200 Rsun around a core • Consistent with pre- explosion imaging and temperature evolution (Kilpatrick, Foley, et al. ’17; Tartaglia et al. ‘17; Arcavi et al. ‘17)

Piro, Muhleisen, et al. (2017) What about boring Type II SNe?

Morozova, Piro, & Valenti (2017) How do we solve this?

Expands and cools

Expands and cools

…but not as much What if... there’s extra stuff around the star? Maybe not so boring after all?

Morozova, Piro, & Valenti (2017) Further Evidence of Dense CSM

Yaron et al. (2017)

• Emission lines seen in early Type II spectra indicate dense CSM • Lower-density, larger-radius material than what we infer Constraining the CSM Structure Moriya et al. (2017)

What’s causing this? Wind acceleration? Additional input? (see Fuller ‘17 and refs therein) Something exciting is happening at the end of these ’ lives! 19 18 Detailed Modeling of a Larger Population 17 Morozova, Piro, & Valenti, in preparation 16 Efin = 1.00 Efin = 1.25 foe Efin = 3.00 foe Efin = 2.25 foe 09ecm MZAMS = 11.0 M 09fma MZAMS = 11.0 M 10abyy MZAMS = 21.0 M 10bgl MZAMS = 20.0 M 15 R-band magnitude 19 18 17 16 Efin = 2.25 foe Efin = 0.50 foe Efin = 1.50 foe Efin = 1.75 foe 10gva MZAMS = 11.0 M 10gxi MZAMS = 19.0 M 10jwr MZAMS = 20.0 M 10mug MZAMS = 11.0 M 15 R-band magnitude 19 18 17 16 Efin = 0.75 foe Efin = 0.50 foe Efin = 0.75 foe Efin = 0.75 foe 10osr MZAMS = 11.0 M 10rem MZAMS = 11.0 M 10uls MZAMS = 20.0 M 10umz MZAMS = 11.0 M 15 R-band magnitude 19 18 17 Once these tools are in place, they 16 Efin = 1.25 foe will be aEfin =useful0.75 foe for applyingEfin = 2.50 foe Efin = 1.25 foe 10uqg MZAMS = 11.0 M 10uqn MZAMS = 11.0 M 10xtq MZAMS = 11.0 M 11ajz MZAMS = 11.0 M 15 R-band magnitude 19 to the larger LSST samples 18 17 16 Efin = 0.75 foe Efin = 1.00 foe Efin = 0.50 foe Efin = 2.75 foe 11cwi MZAMS = 11.0 M 11hsj MZAMS = 11.0 M 11htj MZAMS = 11.5 M 11iqb MZAMS = 11.0 M 15 R-band magnitude 19 18 17 16 Efin = 0.75 foe Efin = 0.50 foe Efin = 0.75 foe Efin = 0.50 foe 11qax MZAMS = 11.0 M 12bbm MZAMS = 20.0 M 12bro MZAMS = 11.0 M 12bvh MZAMS = 16.0 M 15 R-band magnitude 19 18 17 16 Efin = 1.50 foe Efin = 2.75 foe Efin = 0.50 foe Efin = 0.75 foe 12cod MZAMS = 20.0 M 12efk MZAMS = 20.0 M 12fip MZAMS = 11.0 M 12fo MZAMS = 11.0 M 15 R-band magnitude 19 18 17 16 Efin = 1.25 foe Efin = 1.25 foe Efin = 0.50 foe Efin = 0.50 foe 12ftc MZAMS = 11.0 M 12gnn MZAMS = 11.0 M 12grj MZAMS = 11.0 M 12hsx MZAMS = 11.0 M 15 R-band magnitude 19 18 17 16 Efin = 3.00 foe Efin = 1.00 foe Efin = 1.00 foe Efin = 1.25 foe 12krf MZAMS = 22.0 M 13bjx MZAMS = 11.0 M 13bsg MZAMS = 11.0 M 13ccu MZAMS = 11.0 M 15 R-band magnitude 19 18 17 16 Efin = 1.75 foe Efin = 1.25 foe Efin = 2.00 foe Efin = 1.00 foe 13clj MZAMS = 20.0 M 13cly MZAMS = 11.0 M 13dla MZAMS = 11.0 M 13dqy MZAMS = 11.0 M 15 R-band magnitude 19 0 5 10 15 20 25 30 35 18 Time [days] 17 16 Efin = 1.00 foe Efin = 0.50 foe Efin = 2.25 foe 13dzb MZAMS = 11.0 M 14abc MZAMS = 20.0 M 14adz MZAMS = 11.0 M 15 R-band magnitude 0 5 10 15 20 25 30 35 0 5 10 15 20 25 30 35 0 5 10 15 20 25 30 35 Time [days] Time [days] Time [days] Black Hole Formation

“Unnova” Deep Inside the Star

Hot, young neutron Neutrinos carry star produces away energy and neutrinos mass (~0.1Msun) from the

This decreases the neutron star’s gravitational pull, causing the envelope to expand slightly Birth of a black hole (Piro ‘13)

Followed by ~yr long, dim, plateau-like light curve (Lovegrove & Woosley ’13) Observation of a BH birth? Adams et al. (2017)

5 6 ~25 Msun RSG, increased from ~10 Lsun to ~10 Lsun, and then disappeared LSST as a Discovery Machine • Amazing sky coverage & depth is potentially ideal for catching early light curves, and rare and/or dim events • Need high cadence! (between days to ~hrs depending on science) • Color critical for identification (young and hot!) • Quick communication (

• Exquisite record over a lot of the sky with a large time baseline • After SNe discovered by others, search LSST records for early LCs • ~hr to day cadence needed LC features • Longer cadence may still be exciting for pre- explosion activity (see Type II light curves!) and early SLSNe General LSST Cadence Thoughts

• What is the main transient science we want to address? (not now, but when LSST is operating!) • What will be the landscape when LSST is operating? (ASAS-SN, ZTF, GMT, TMT, ELT, etc) • There’s cadence AND color AND targeting. What is the correct balance? • What is the best way to make data useful for transient scientists? • Is there a way to build in flexibility? How much? Conclusions Early light curves provide new and valuable information about exploding stars

• Shock cooling non-detections provide radius constraints for

SN Ia progenitors (< 0.02 Rsun) and stripped-envelope SNe (< 5 Rsun) • Evidence (or not) for non-degenerate companions in SNe Ia • Direct measurements of the radius of extended material around SNe IIb (~300 Rsun), SLSNe (~400-1000 Rsun) • Surprisingly, boring Type II SNe show signs of dense CSM • LSST is a Discovery Machine but is also a Time Machine, providing a record of early LCs and pre-explosion activity