Neutron star mergers and kilonovas

15 min presentation with an overview of the astrophysics of neutron star mergers and kilonovas, following the birth of multimessenger astronomy with GW170817. I discussed opportunities for wide-field optical telescopes. Talk given at the S-PLUS workshop on Sep. 28th 2018.

Speaker: Prof. Rodrigo Nemmen. Universidade de Sao Paulo.
https://rodrigonemmen.com

Credit for the slides/figures belongs to Rodrigo Nemmen, unless otherwise stated.

Transcript

1. Rodrigo Nemmen
   Universidade de Sao Paulo
   Neutron star mergers
   and kilonovas
   

   View Slide

2. theory
   observations
   S-PLUS opportunities Rodrigo Nemmen
   Universidade de Sao Paulo
   Neutron star mergers
   and kilonovas
   

   View Slide

3. Mergers of binary neutron stars: Short GRBs
   NS
   NS
   NS
   BH
   or
   Paczynski 1986; Eichler et al. 1989; Narayan et al. 1992
   time
   

   View Slide

4. Mergers of binary neutron stars: Short GRBs
   NS
   NS
   Paczynski 1986; Eichler et al. 1989; Narayan et al. 1992
   time
   GWs
   orbit decays due to
   GW emission
   

   View Slide

5. Mergers of binary neutron stars: Short GRBs
   NS
   NS
   Paczynski 1986; Eichler et al. 1989; Narayan et al. 1992
   time
   neutron rich, magnetized
   accreting torus
   GWs
   orbit decays due to
   GW emission
   duration ~ ms
   

   View Slide

6. Rezzolla+2011
   cf. also Ruiz+2016
   Numerical relativity simulation of NS merger:
   Short GRB emerges naturally
   Rezzolla, Giacomazzo et al. 2011
   Cactus/Carpet/Whisky codes
   (solves Einstein eqs. + ideal MHD)
   Initial condition:
   •NS binary a few orbits before its
   coalescence
   •2 NSs
   •each NS with M=1.5Msun
   •Bmax=1012 G (each star)
   

   View Slide

7. Rezzolla+2011
   cf. also Ruiz+2016
   gas density magnetic fields
   

   View Slide

8. Rezzolla+2011
   cf. also Ruiz+2016
   gas density magnetic fields
   Results in:
   • accreting BH, M = 2.91 Msun
   • short GRB jet, Δt = 30 ms
   • opening angle 30 deg
   • amplifies B to 1015 G
   • Eiso = 1049 erg
   

   View Slide

9. fallback
   material
   GRB jet
   (beamed)
   Ejecta
   (unbeamed)
   

   View Slide

10. fallback
    material
    GRB jet
    (beamed)
    Lattimer & Schramm 1974
    
    produces many Earth-mass
    amounts of precious metals
    (r-process, lanthanides)
    Ejecta
    (unbeamed)
    radio
    optical
    X-rays
    γ-rays
    kilonova:
    IR
    optical
    UV
    

    View Slide

11. fallback
    material
    GRB jet
    (beamed)
    Lattimer & Schramm 1974
    
    produces many Earth-mass
    amounts of precious metals
    (r-process, lanthanides)
    Ejecta
    (unbeamed)
    radio
    optical
    X-rays
    γ-rays
    kilonova:
    IR
    optical
    UV
    Short GRB multimessenger
    time series: GWs + light
    Fernandez & Metzger 2016
    
    nandez-Metzger ARI 4 May 2016 13:30
    Event Signal Phase
    Coalescence
    X-ray/radio
    precursor?
    μ
    Ω
    r-process
    ms magnetar?
    Short GRB
    X-ray extended
    emission/plateau
    Ye
    > 0.25
    Ye
    < 0.25
    Free n?
    Shocked ISM
    GW “chirp” GWs from remnant NS?
    Inspiral Dynamical Accretion Remnant
    BH
    ??
    Blue
    kilonova
    ??
    Neutron
    precursor
    (UV)
    Red
    kilonova
    Radio
    transient
    Seconds Milliseconds 10 ms 100 ms Minutes–hours Hours–days Days–weeks Month–years
    Event Signal
    Coalescence
    X-ray/radio
    precursor?
    μ
    Ω
    r-process
    ms magnetar?
    Short GRB
    X-ray extended
    emission/plateau
    Ye
    > 0.25
    Ye
    < 0.25
    Free n?
    Shocked ISM
    GW “chirp” GWs from remnant NS?
    NS
    BH
    NS
    ??
    Blue
    kilonova
    ??
    Neutron
    precursor
    (UV)
    Red
    kilonova
    Radio
    transient
    Figure 1
    Phases of a neutron star (NS) merger as a function of time, showing the associated observational signatures
    and underlying physical phenomena. Abbreviations: BH, black hole; GRB, γ -ray burst; GW, gravitational
    wave; ISM, interstellar medium; n, neutron; UV, ultraviolet; Ye
    , electron fraction. Coalescence inset
    courtesy of D. Price and S. Rosswog (see also Reference 15).
    

    View Slide

12. Birth of multimessenger astronomy: GWs
    and EM radiation from GW170817
    

    View Slide

13. Birth of multimessenger astronomy: GWs
    and EM radiation from GW170817
    

    View Slide

14. Diaz+2017
    

    View Slide

15. Kilonova opportunities for GMT
    lighter elements
    (lanthanide-poor)
    heavier elements
    (lanthanide-rich)
    (UV)
    (NIR)
    cf. also Cowperthwaite+2017
    

    View Slide

16. Kilonova opportunities for GMT
    lighter elements
    (lanthanide-poor)
    heavier elements
    (lanthanide-rich)
    (UV)
    (NIR)
    cf. also Cowperthwaite+2017
    

    View Slide

17. r-process nucleosynthesis in the dynamical
    ejecta from NS-NS or NS-BH mergers
    

    View Slide

18. r-process nucleosynthesis in the dynamical
    ejecta from NS-NS or NS-BH mergers
    

    View Slide

19. View Slide

20. Figure 2. gri light curves of the EM counterpart to GW170817, obtained with
    T80S on 2017 August 18. The g points have been offset by −0.4mag for
    clarity.
    Table 1
    Time-series Photometry
    Timea Band Mag σ (mag)
    1.4390 g 18.43 0.06
    1.4447 g 18.51 0.04
    1.4458 g 18.48 0.04
    Figure 3. Top: comparison of the light curve slopes (magnitudes per day) of
    the transient (arrows) in gri (left, center, and right panels, respectively) relative
    to 885 objects in the T80S FoV with similar colors ( g r r i
    , 0.1
    - - <
    ∣ ∣ ∣ ∣ ).
    The horizontal error bars indicate the 1s uncertainty in the values for the
    transient. Bottom: statistical significance of the values for the transient in gri
    (left, center, and right panels, respectively) relative to the comparison sample.
    The Astrophysical Journal Letters, 848:L29 (5pp), 2017 October 20 Díaz et al.
    80S FoV with similar colors to the transient (within 0.1 mag
    both g − r and r i
    - ). We performed weighted linear fits as a
    nction of time on the light curves of all of the selected objects
    gure 2. gri light curves of the EM counterpart to GW170817, obtained with
    0S on 2017 August 18. The g points have been offset by −0.4mag for
    rity.
    Table 1
    Time-series Photometry
    mea Band Mag σ (mag)
    390 g 18.43 0.06
    447 g 18.51 0.04
    458 g 18.48 0.04
    469 g 18.62 0.04
    481 r 17.93 0.02
    492 r 17.97 0.02
    502 r 17.94 0.02
    514 i 17.74 0.03
    te.
    Days since GW trigger.
    his table is available in its entirety in machine-readable form.)
    Figure 3. Top: comparison of the light curve slopes (magnitudes per day) of
    the transient (arrows) in gri (left, center, and right panels, respectively) relative
    to 885 objects in the T80S FoV with similar colors ( g r r i
    , 0.1
    - - <
    ∣ ∣ ∣ ∣ ).
    The horizontal error bars indicate the 1s uncertainty in the values for the
    transient. Bottom: statistical significance of the values for the transient in gri
    (left, center, and right panels, respectively) relative to the comparison sample.
    

    View Slide

21. 10000
    1042
    1041
    1040
    8000
    6000
    4000
    Temperature (K)
    Magnitude (+ offset)
    Luminosity (erg s–1)
    0 5 10 15
    0 5 10 15
    12.5
    15.0
    17.5
    20.0
    22.5
    25.0
    27.5
    30.0
    32.5
    35.0
    on January 2, 2018
    http://science.sciencemag.org/
    aded from
    Kilpatrick+2017, Science
    Cowperthwaite+2017
    Drout+2017, Science
    Rest-frame days after merger
    

    View Slide

22. 10000
    1042
    1041
    1040
    8000
    6000
    4000
    Temperature (K)
    Magnitude (+ offset)
    Luminosity (erg s–1)
    0 5 10 15
    0 5 10 15
    12.5
    15.0
    17.5
    20.0
    22.5
    25.0
    27.5
    30.0
    32.5
    35.0
    on January 2, 2018
    http://science.sciencemag.org/
    aded from
    Kilpatrick+2017, Science
    Cowperthwaite+2017
    Drout+2017, Science
    Rest-frame days after merger
    ric light curves, in
    n≲1 day, then fades
    volution to the red
    . 2 exhibit similarly
    ter the trigger the
    , but it transitions
    r spectrum with at
    near 9000 Å (12).
    numerous absorp-
    n the spectra of or-
    s of astrophysical
    h at least one of the
    7a: the rise time of
    g at ≳1 mag dayÀ1,
    H-band magnitudes
    À1:2 to þ3:6 mag
    featureless optical
    Among previously
    thermal continuum
    s is rapidly evolving
    these events have
    luminous, typically
    s, do not have the
    nd do not fade as
    hermal relativistic
    lows can produce
    are not expected to
    or just after the merger, is thought to be ob- and spectral energy distribution of a kilonova
    10000
    1042
    1041
    1040
    8000
    6000
    erature (K)
    Magnitude (+ offset)
    Luminosity (erg s–1)
    0 5 10 15
    12.5
    15.0
    17.5
    20.0
    22.5
    25.0
    27.5
    30.0
    32.5
    on January 2, 2018
    http://science.sciencemag.org/
    nloaded from
    Rest-frame days after merger
    

    View Slide

23. Scientific questions w/ kilonova monitoring
    how many heavy-elements (lanthanides) produced?
    how much gold in the universe is produced in NSNS mergers?
    signatures of NS-BH mergers?
    velocity and mass of ejecta?
    

    View Slide

24. LETTER
    doi:10.1038/nature24471
    A gravitational-wave standard siren measurement
    of the Hubble constant
    The LIGO Scientific Collaboration and The Virgo Collaboration*, The 1M2H Collaboration*, The Dark Energy Camera GW-EM
    Collaboration and the DES Collaboration*, The DLT40 Collaboration*, The Las Cumbres Observatory Collaboration*,
    The VINROUGE Collaboration* & The MASTER Collaboration*
    On 17 August 2017, the Advanced LIGO1 and Virgo2 detectors
    observed the gravitational-wave event GW170817—a strong signal
    from the merger of a binary neutron-star system3. Less than two
    seconds after the merger, a γ-ray burst (GRB 170817A) was detected
    within a region of the sky consistent with the LIGO–Virgo-derived
    location of the gravitational-wave source4–6. This sky region was
    subsequently observed by optical astronomy facilities7, resulting
    in the identification8–13 of an optical transient signal within
    about ten arcseconds of the galaxy NGC 4993. This detection of
    GW170817 in both gravitational waves and electromagnetic waves
    represents the first ‘multi-messenger’ astronomical observation.
    Such observations enable GW170817 to be used as a ‘standard
    siren’14–18 (meaning that the absolute distance to the source can be
    this galaxy allow us to estimate the appropriate value of the Hubble flow
    velocity. Because the source is relatively nearby, the random relative
    motions of galaxies, known as peculiar velocities, need to be taken into
    account. The peculiar velocity is about 10% of the measured recessional
    velocity (see Methods).
    The original standard siren proposal14 did not rely on the unique
    identification of a host galaxy. By combining information from around
    100 independent gravitational-wave detections, each with a set of
    potential host galaxies, an estimate of H0
    accurate to 5% can be obtained
    even without the detection of any transient optical counterparts22. This
    is particularly relevant, because gravitational-wave networks will detect
    many binary black-hole mergers over the coming years23 and these
    are not expected to be accompanied by electromagnetic counterparts.
    

    View Slide

25. LETTER
    doi:10.1038/nature24471
    A gravitational-wave standard siren measurement
    of the Hubble constant
    The LIGO Scientific Collaboration and The Virgo Collaboration*, The 1M2H Collaboration*, The Dark Energy Camera GW-EM
    Collaboration and the DES Collaboration*, The DLT40 Collaboration*, The Las Cumbres Observatory Collaboration*,
    The VINROUGE Collaboration* & The MASTER Collaboration*
    On 17 August 2017, the Advanced LIGO1 and Virgo2 detectors
    observed the gravitational-wave event GW170817—a strong signal
    from the merger of a binary neutron-star system3. Less than two
    seconds after the merger, a γ-ray burst (GRB 170817A) was detected
    within a region of the sky consistent with the LIGO–Virgo-derived
    location of the gravitational-wave source4–6. This sky region was
    subsequently observed by optical astronomy facilities7, resulting
    in the identification8–13 of an optical transient signal within
    about ten arcseconds of the galaxy NGC 4993. This detection of
    GW170817 in both gravitational waves and electromagnetic waves
    represents the first ‘multi-messenger’ astronomical observation.
    Such observations enable GW170817 to be used as a ‘standard
    siren’14–18 (meaning that the absolute distance to the source can be
    this galaxy allow us to estimate the appropriate value of the Hubble flow
    velocity. Because the source is relatively nearby, the random relative
    motions of galaxies, known as peculiar velocities, need to be taken into
    account. The peculiar velocity is about 10% of the measured recessional
    velocity (see Methods).
    The original standard siren proposal14 did not rely on the unique
    identification of a host galaxy. By combining information from around
    100 independent gravitational-wave detections, each with a set of
    potential host galaxies, an estimate of H0
    accurate to 5% can be obtained
    even without the detection of any transient optical counterparts22. This
    is particularly relevant, because gravitational-wave networks will detect
    many binary black-hole mergers over the coming years23 and these
    are not expected to be accompanied by electromagnetic counterparts.
    Distance from GW observations
    z from EM obs.
    measurements of H0
    

    View Slide

26. LETTER
    doi:10.1038/nature24471
    A gravitational-wave standard siren measurement
    of the Hubble constant
    The LIGO Scientific Collaboration and The Virgo Collaboration*, The 1M2H Collaboration*, The Dark Energy Camera GW-EM
    Collaboration and the DES Collaboration*, The DLT40 Collaboration*, The Las Cumbres Observatory Collaboration*,
    The VINROUGE Collaboration* & The MASTER Collaboration*
    On 17 August 2017, the Advanced LIGO1 and Virgo2 detectors
    observed the gravitational-wave event GW170817—a strong signal
    from the merger of a binary neutron-star system3. Less than two
    seconds after the merger, a γ-ray burst (GRB 170817A) was detected
    within a region of the sky consistent with the LIGO–Virgo-derived
    location of the gravitational-wave source4–6. This sky region was
    subsequently observed by optical astronomy facilities7, resulting
    in the identification8–13 of an optical transient signal within
    about ten arcseconds of the galaxy NGC 4993. This detection of
    GW170817 in both gravitational waves and electromagnetic waves
    represents the first ‘multi-messenger’ astronomical observation.
    Such observations enable GW170817 to be used as a ‘standard
    siren’14–18 (meaning that the absolute distance to the source can be
    this galaxy allow us to estimate the appropriate value of the Hubble flow
    velocity. Because the source is relatively nearby, the random relative
    motions of galaxies, known as peculiar velocities, need to be taken into
    account. The peculiar velocity is about 10% of the measured recessional
    velocity (see Methods).
    The original standard siren proposal14 did not rely on the unique
    identification of a host galaxy. By combining information from around
    100 independent gravitational-wave detections, each with a set of
    potential host galaxies, an estimate of H0
    accurate to 5% can be obtained
    even without the detection of any transient optical counterparts22. This
    is particularly relevant, because gravitational-wave networks will detect
    many binary black-hole mergers over the coming years23 and these
    are not expected to be accompanied by electromagnetic counterparts.
    the SHoES result is just outside the 90% confidence range. It will be
    particularly interesting to compare these constraints to those from
    modelling7 of the short γ-ray burst, afterglow and optical counterpart
    associated with GW170817.
    We have presented a standard siren determination of the Hubble
    constant, using a combination of a distance estimate from gravita-
    tional-wave observations and a Hubble velocity estimate from electro-
    magnetic observations. Our measurement does not use a ‘distance
    ladder’ and makes no prior assumptions about H0
    . We find
    = .
    − .
    + . − −
    H 70 0 km s Mpc
    0 8 0
    12 0 1 1, which is consistent with existing meas-
    urements20,21. This first gravitational-wave–electromagnetic multi-
    messenger event demonstrates the potential for cosmological inference
    from gravitational-wave standard sirens. We expect that additional
    multi-messenger binary neutron-star events will be detected in the
    coming years, and combining subsequent independent measurements
    of H0
    from these future standard sirens will lead to an era of precision
    gravitational-wave cosmology.
    2
    2
    2
    2
    2
    2
    2
    2
    3
    

    View Slide

27. View Slide

28. there should be a wide variety of kilonova properties
    that are not clear from only n = 1 data point
    

    View Slide

29. there should be a wide variety of kilonova properties
    that are not clear from only n = 1 data point
    

    View Slide

30. there should be a wide variety of kilonova properties
    that are not clear from only n = 1 data point
    expect 1 NS-NS or NS-BH merger seen by LIGO-
    Virgo per month
    

    View Slide

31. there should be a wide variety of kilonova properties
    that are not clear from only n = 1 data point
    expect 1 NS-NS or NS-BH merger seen by LIGO-
    Virgo per month
    

    View Slide

32. there should be a wide variety of kilonova properties
    that are not clear from only n = 1 data point
    expect 1 NS-NS or NS-BH merger seen by LIGO-
    Virgo per month
    many opportunities for EM followups!
    

    View Slide

33. there should be a wide variety of kilonova properties
    that are not clear from only n = 1 data point
    expect 1 NS-NS or NS-BH merger seen by LIGO-
    Virgo per month
    many opportunities for EM followups!
    

    View Slide

34. there should be a wide variety of kilonova properties
    that are not clear from only n = 1 data point
    expect 1 NS-NS or NS-BH merger seen by LIGO-
    Virgo per month
    many opportunities for EM followups!
    nearby ones should be observable with S-PLUS
    

    View Slide

35. What I would like to see
    coverage of S-PLUS competitive with DECam
    dedicated workgroup for kilonova monitoring
    coordination with LIGO/Virgo for triggers
    theorists required (kilonova modeling)
    

    View Slide

36. Summary: Kilonova opportunities for S-PLUS
    (i) measuring kilonova light curve
    (ii) photometric zs → cosmology w/ GW sources
    (iii) afterglow (did not discuss)
    Rodrigo Nemmen
    Kilonovas: product of NS-NS or NS-BH mergers,
    radioactive glow from debris
    Produce most of the gold in the universe
    S-PLUS opportunities
    Should we organize a dedicated workgroup for
    these transients?
    

    View Slide

37. Github
    Twitter
    Web
    E-mail
    Bitbucket
    Facebook
    Group
    figshare
    rodrigo.nemmen@iag.usp.br
    rodrigonemmen.com
    @nemmen
    rsnemmen
    facebook.com/rodrigonemmen
    nemmen
    blackholegroup.org
    bit.ly/2fax2cT
    

    View Slide

38. Directors cut
    

    View Slide

39. View Slide

40. View Slide

41. 0 5 10 15 20 25 30 35
    0
    2
    4
    6
    8
    10
    12
    14
    16
    18
    20
    Long
    Short
    Opening angle θ
    j
    (degrees)
    Number
    Figure 18
    Distributions of jet opening angles for short (blue) and long (red ) GRBs on the basis of breaks in their
    afterglow emission. Arrows mark lower or upper limits on the opening angles. The observations are
    summarized in Section 8.4. From Fong et al. (2013) and references therein.
    . Downloaded from www.annualreviews.org
    s on 06/05/14. For personal use only.
    Berger 2014 ARAA
    

    View Slide

42. Van Eerten 2018
    

    View Slide

43. slide: optical opportunities for GMT
    afterglow measurements
    what they can teach us
    Log Optical Flux ->
    Log Time ->
    Most GRB Have Optical Afterglows
    Prompt X/-ray light curves
    bright (wide field instruments),
    highly variable,
    inhomogeneous shapes
    LGRB typically ~ 40-100 s
    OPT AG MEASUREMENTS
    COMMON - Hundreds
    observed, dozens per year.
    (Flares ignored here)
    Physics well understood to
    be interaction of a jet with
    ISM.
    [Early or prompt phase not
    measured for most GRBs]
    Covino et al.: The prompt and the afterglow of GRB 060908 5
    10-6
    10-5
    10-4
    10-3
    10-2
    10-1
    100
    101
    102
    101 102 103 104 105 106 107 108
    Flux (mJy)
    Time from burst (s)
    1 keV
    U (x 100)
    B (x 30)
    V (x 10)
    R (x 3)
    I (x 1)
    Z (x 0.4)
    White (x 0.4)
    J (x 0.06)
    H (x 0.035)
    K (x 0.02)
    Covimo+2010
    5.1. The picture since Swift
    Fig. 3. An overview of the various issues raised by GRB afterglow observations from 2004+, in
    particular those by Swift XRT.
    Since the launch of Swift, a complex picture of X-ray and optical afterglows has
    emerged. To some extent, this picture can be described in terms of a canonical long
    GRB afterglow light curve189–191 (see also Fig. 3, expanded from an illustration
    in Ref. 192), although analysis of the Swift XRT sample shows that ‘canonical’
    should be taken with a grain of salt193, 194 and that the measured temporal slopes
    of the light curves span a wide range. After an initial flaring behaviour presumably
    connected to the prompt emission, the light curve drops steeply until it reaches
    a plateau value. The light curve then maintains this value for longer than was
    Eerten 2018
    

    View Slide

44. Slide: Xue-Feng Wu
    

    View Slide

45. GRB optical jet breaks
    Liang+2008
    

    View Slide

46. View Slide