Are Nova Light Curves Powered by Shocks?
Are Nova Light Curves Powered by Shocks?
Brian Metzger (Columbia University) Primary Collaborators Andrei Beloborodov, Andrea Derdzinski, Jeno Sokoloski, Andrey Vlasov, Indrek Vurm (Columbia) Laura Chomiuk, Tom Finzell (Michigan State), Damiano Caprioli (Princeton), Jennifer Weston (WVU)
Workshop on Shocks and Particle Acceleration in Novae and Supernovae
Classical & Symbiotic Novae
David A. Hardy/PPARC
Runaway hydrogen burning on white dwarf accreting from a main sequence or red giant companion. Optical & UV outburst lasting weeks to months with luminosity ~Ledd ~ 1038 erg s-1. Ejecta velocities of ~300 - 3,000 km s-1 and total mass of ~10-5 – 10-4 M¤. Thermal emission: soft X-rays (106 K WD surface) and radio (104 K photo-ionized gas)
Hachisu & Kato 2015
LAT detects novae Ackermann et al. 2014, Cheung et al. 2016
Lγ (0.1-10 GeV) ~ 1035-36 erg s-1
Duration ~ weeks (similar to optical)
Geometry of Classical Nova Shocks Ω
Slow “shell” v < 103 km s-1 t < days Vs < 103 km s-1
WD
Vf ~ few 103 km s-1
Geometry of Classical Nova Shocks Slow “shell” v < 103 km s-1 t < days
fast “wind”
v > 103 km s-1 t > weeks Vf ~ few 103 km s-1
Vs < 103 km s-1
Day 126
Day 615
Vs < 103 km s-1
V959 Mon Chomiuk et al. 2014
see also talk by Linford
LAT Spectrum: Hadronic or Leptonic? Leptonic
e− + γ opt ⇒ 4 e− + γ ' γ e ~10
−
−
e + p ⇒3 e +γ
'
γ e ~10
Ackermann et al. 2014
Hadronic p + p ⇒ π 0 ⇒ γ +γ ⇒ π ± ⇒ ν 's + e± ⇒ ν 's + γ 's
LAT Spectrum: Hadronic or Leptonic? Leptonic
e− + γ opt ⇒ 4 e− + γ '
Leptonic
γ e ~10
−
f ( p) ~ p
−4
LAT eligible ~ 10%
−
e + p ⇒3 e +γ
'
γ e ~10
Hadronic p + p ⇒ π 0 ⇒ γ +γ ⇒ π ± ⇒ ν 's + e± ⇒ ν 's + γ 's
Lsh ~ 100 Lγ ~ 1037-38 erg s-1 ~ LBOL
Novae eject (a lot of) mass Observations (Seaquist & Bode 08)
−4 −1 ! M ~ 10 M ! month
Roy et al. 2012
Nova shocks: dense & (probably) radiative −3 −2 ! ! ⎛ ⎞ ⎛ ⎞ V ⎛ ⎞ M M t ej 10 −3 n~ ~ 10 cm ⎜ ⎟ ⎜ ⎟ ⎜ ⎟ −4 3 −1 4π Vej R 2 m p R~Vejt ⎝ 10 M ! /month ⎠⎝ week ⎠ ⎝ 10 km s ⎠
Shock Radius
Nova shocks: dense & (probably) radiative
Shock Velocity
Nova shocks: dense & (probably) radiative M! ~ 10 −4 M ! month −1
tcool =
(3 / 2)kTsh n Λ(Tsh )
Shock Radius
texp =
R vej
Nova shocks: dense & (probably) radiative M! ~ 10 −4 M ! month −1
tcool =
(3 / 2)kTsh n Λ(Tsh )
Shock Radius
texp =
R vej
Lsh ~ 100 Lγ ~ 1037-38 erg s-1 ~ LBOL
Where is the radiation coming out?
Thermal keV X-rays from Shocks
Schwarz et al. 2011
Swift Sample “Hard”
Day after Outburst
see also Mukai et al. 2008
Lx ~ 1032-1034 erg s-1 10-3 – 10-2 Hadronic Scenario: εnth up to 0.1, depending on B field geometry (Caprioli & Spitkovsky 14) Leptonic Scenario: εnth < 10-3 from observations & PIC simulations (e.g. Kato 14, Park+14)
Summary • Discovery of novae as GeV γ-ray sources establishes that shocks & relativistic particle acceleration are key features of these events. • High densities of classical nova ejecta imply: (1) shocks are radiative; (2) gas upstream of shocks is neutral; (3) relativistic leptons/hadrons are fast cooling [calorimeter]. • Thermal X-rays from γ-ray shocks not observed at early times (absorption by neutral gas) => shock power emerges at optical/UV, as in Type IIn SNe. • Measured ratio of γ-ray to optical luminosities places lower limit on acceleration efficiency of non-thermal particles, εnth > 10-2-10-3. • If εnth < 10%, then > 1-10% of optical emission is shock powered • Prediction: Correlation between bolometric output and gamma-ray luminosity
Overflow Slides
Vs < 103 km s-1
Non-Thermal Radio Emission V1324 Sco
Rise delayed by free-free absorption
radio
BDM et al. 2015b
BDM et al. 2014
Peak brightness temperature constrains εnth of relativistic electrons (see Vlasov Poster).
Fermi Acceleration Confined to Ionized Layer
BDM et al. 2015b
MAGIC Collaboration 15
Prospects for TeV Emission
Maximum Particle Energy in Nova Shocks
BDM+15b Based on growth rate analysis of Bell (2004) instability
Metzger+15b
Fermi Acceleration Confined to Ionized Layer
BDM et al. 2015b
MAGIC Collaboration 15
Prospects for TeV Emission
Potential IceCube sources (ask me)
Brian Metzger (Columbia University) Primary Collaborators Andrei Beloborodov, Andrea Derdzinski, Jeno Sokoloski, Andrey Vlasov, Indrek Vurm (Columbia) Laura Chomiuk, Tom Finzell (Michigan State), Damiano Caprioli (Princeton), Jennifer Weston (WVU)
Workshop on Shocks and Particle Acceleration in Novae and Supernovae
Classical & Symbiotic Novae
David A. Hardy/PPARC
Runaway hydrogen burning on white dwarf accreting from a main sequence or red giant companion. Optical & UV outburst lasting weeks to months with luminosity ~Ledd ~ 1038 erg s-1. Ejecta velocities of ~300 - 3,000 km s-1 and total mass of ~10-5 – 10-4 M¤. Thermal emission: soft X-rays (106 K WD surface) and radio (104 K photo-ionized gas)
Hachisu & Kato 2015
LAT detects novae Ackermann et al. 2014, Cheung et al. 2016
Lγ (0.1-10 GeV) ~ 1035-36 erg s-1
Duration ~ weeks (similar to optical)
Geometry of Classical Nova Shocks Ω
Slow “shell” v < 103 km s-1 t < days Vs < 103 km s-1
WD
Vf ~ few 103 km s-1
Geometry of Classical Nova Shocks Slow “shell” v < 103 km s-1 t < days
fast “wind”
v > 103 km s-1 t > weeks Vf ~ few 103 km s-1
Vs < 103 km s-1
Day 126
Day 615
Vs < 103 km s-1
V959 Mon Chomiuk et al. 2014
see also talk by Linford
LAT Spectrum: Hadronic or Leptonic? Leptonic
e− + γ opt ⇒ 4 e− + γ ' γ e ~10
−
−
e + p ⇒3 e +γ
'
γ e ~10
Ackermann et al. 2014
Hadronic p + p ⇒ π 0 ⇒ γ +γ ⇒ π ± ⇒ ν 's + e± ⇒ ν 's + γ 's
LAT Spectrum: Hadronic or Leptonic? Leptonic
e− + γ opt ⇒ 4 e− + γ '
Leptonic
γ e ~10
−
f ( p) ~ p
−4
LAT eligible ~ 10%
−
e + p ⇒3 e +γ
'
γ e ~10
Hadronic p + p ⇒ π 0 ⇒ γ +γ ⇒ π ± ⇒ ν 's + e± ⇒ ν 's + γ 's
Lsh ~ 100 Lγ ~ 1037-38 erg s-1 ~ LBOL
Novae eject (a lot of) mass Observations (Seaquist & Bode 08)
−4 −1 ! M ~ 10 M ! month
Roy et al. 2012
Nova shocks: dense & (probably) radiative −3 −2 ! ! ⎛ ⎞ ⎛ ⎞ V ⎛ ⎞ M M t ej 10 −3 n~ ~ 10 cm ⎜ ⎟ ⎜ ⎟ ⎜ ⎟ −4 3 −1 4π Vej R 2 m p R~Vejt ⎝ 10 M ! /month ⎠⎝ week ⎠ ⎝ 10 km s ⎠
Shock Radius
Nova shocks: dense & (probably) radiative
Shock Velocity
Nova shocks: dense & (probably) radiative M! ~ 10 −4 M ! month −1
tcool =
(3 / 2)kTsh n Λ(Tsh )
Shock Radius
texp =
R vej
Nova shocks: dense & (probably) radiative M! ~ 10 −4 M ! month −1
tcool =
(3 / 2)kTsh n Λ(Tsh )
Shock Radius
texp =
R vej
Lsh ~ 100 Lγ ~ 1037-38 erg s-1 ~ LBOL
Where is the radiation coming out?
Thermal keV X-rays from Shocks
Schwarz et al. 2011
Swift Sample “Hard”
Day after Outburst
see also Mukai et al. 2008
Lx ~ 1032-1034 erg s-1 10-3 – 10-2 Hadronic Scenario: εnth up to 0.1, depending on B field geometry (Caprioli & Spitkovsky 14) Leptonic Scenario: εnth < 10-3 from observations & PIC simulations (e.g. Kato 14, Park+14)
Summary • Discovery of novae as GeV γ-ray sources establishes that shocks & relativistic particle acceleration are key features of these events. • High densities of classical nova ejecta imply: (1) shocks are radiative; (2) gas upstream of shocks is neutral; (3) relativistic leptons/hadrons are fast cooling [calorimeter]. • Thermal X-rays from γ-ray shocks not observed at early times (absorption by neutral gas) => shock power emerges at optical/UV, as in Type IIn SNe. • Measured ratio of γ-ray to optical luminosities places lower limit on acceleration efficiency of non-thermal particles, εnth > 10-2-10-3. • If εnth < 10%, then > 1-10% of optical emission is shock powered • Prediction: Correlation between bolometric output and gamma-ray luminosity
Overflow Slides
Vs < 103 km s-1
Non-Thermal Radio Emission V1324 Sco
Rise delayed by free-free absorption
radio
BDM et al. 2015b
BDM et al. 2014
Peak brightness temperature constrains εnth of relativistic electrons (see Vlasov Poster).
Fermi Acceleration Confined to Ionized Layer
BDM et al. 2015b
MAGIC Collaboration 15
Prospects for TeV Emission
Maximum Particle Energy in Nova Shocks
BDM+15b Based on growth rate analysis of Bell (2004) instability
Metzger+15b
Fermi Acceleration Confined to Ionized Layer
BDM et al. 2015b
MAGIC Collaboration 15
Prospects for TeV Emission
Potential IceCube sources (ask me)
