Download Reconciling observations and modelling of thermonuclear bursts

Reconciling observations and
modelling of thermonuclear bursts
with nuclear experiments
Duncan Galloway
ASA Annual Scientific Meeting, 2016 July, Sydney
with MINBAR collaboration
and 2015 ISSI int’l team
Thermonuclear X-ray bursts
• Occur in neutron stars accreting from low-mass binary
companions; ~100 bursters known, ~104 bursts observed
since early 1970s
• Understood since the `80s as resulting from unstable
ignition of accreted H/He on the NS surface (e.g. Fujimoto et al. 1981,
ApJ 247, 267)
Galloway: Observations and modelling of thermonuclear bursts
Nuclear reactions
• H and He burning occur
~independently, via the hot
CNO cycle & rp-process
burning, or triple-alpha
• Burning of both species can
be stable or unstable,
depending on accretion rate
Galloway: Observations and modelling of thermonuclear bursts
Keek et al. 2015
• Accreted fuel is thought to be a mix of H & He, at roughly
solar composition (70/28%)
• A subset of sources have evolved companions and likely
accrete (almost?) pure He
End-point of rp-process burning
• Dominates late-time burning in mixed H/He bursts;
terminates in a Sn-Sb-Te cycle
Schatz et al. (2001)
Many of these reaction rates
and nucleon masses are
poorly measured
experimentally – opportunity
for probing via bursts
Galloway: Observations and modelling of thermonuclear bursts
Focus on key reactions
• Ongoing numerical work on the expected influence of
individual reactions on burst rates, lightcurve shape
• 15O(α,γ)19Ne reaction a major influence on the stability of
burning Fisker et al. 2007
• Most important reactions identified
from 1-D multizone models Fisker et al.
2008; Cyburt et al. 2010, 2016
• Also motivating experimental efforts
to measure the rates of important
reactions, e.g. 15O(α,γ)19Ne Tan et al.
2009
Galloway: Observations and modelling of thermonuclear bursts
our objective:
“precision” nuclear astrophysics from
thermonuclear bursts
Meaning, we want to reduce the astrophysical
uncertainties for observations of bursts, to the
point where we are sensitive to the nuclear
physics
GS 1826-24: a case history
• An unusual source in that it consistently* shows regular,
consistent bursts
• Early analysis of the variation of the recurrence time with
accretion rate suggested metal-poor material Galloway et al. 2004
• Subsequent study showed good
agreement with time-dependent 1D models & solar composition Heger
et al. 2007
• Relied on lightcurve comparison at
a single epoch
* until 2014 June
Galloway: Observations and modelling of thermonuclear bursts
Comprehensive model fits
• We will perform simultaneous multi-epoch comparisons
drawing from a large sample of observed bursts, the MultiINstrument Burst ARchive (MINBAR;
http://burst.sci.monash.edu/minbar)
Galloway: Observations and modelling of thermonuclear bursts
Representative observed bursts
From top to bottom, we have
•
He bursts from an 11-min binary 4U 1820-303, featuring a
likely white dwarf accretor;
•
He bursts from a 2.01 hr binary (and 401 Hz millisecond
pulsar, SAX J1808.4–3658), during the 2002 outburst; these
events occur after any accreted H fuel at the ignition depth
has been exhausted by steady burning prior to ignition;
•
Mixed H/He bursts from a ~2 hr binary known for its
consistently regular and consistent bursts (although see
Thompson et al. 2008, Chenevez et al. 2016); and
•
A carbon-burning superburst observed from the 3.8 hr binary
4U 1636–536
Note the change in x- and y-axis scales between the different
panels
All the bursts have been observed with the highest signal-to-noise
available over the last 30 years, by the Rossi X-ray Timing
Explorer
Lightcurves and other observational parameters will be sourced
from the MINBAR sample http://burst.sci.monash.edu/minbar
Galloway: Observations and modelling of thermonuclear bursts
See Adelle Goodwin’s talk
at 5pm, and also Zac
Johnston’s poster #50
Comprehensive model fits
• We will perform simultaneous multi-epoch comparisons
drawing from a large sample of observed bursts, the MultiINstrument Burst ARchive (MINBAR;
http://burst.sci.monash.edu/minbar)
• The KEPLER model runs are expensive (a ~week for each
burst train!) so we adopt a large sample of simulated bursts
Lampe et al. 2016, ApJ 819, 46
Galloway: Observations and modelling of thermonuclear bursts
Exploit large model run samples
Lampe et al. 2016, ApJ 819, 46
• We have assembled, and released, a large sample of
KEPLER simulations for comparison with observations
Galloway: Observations and modelling of thermonuclear bursts
Comprehensive model fits
• We will perform simultaneous multi-epoch comparisons
drawing from a large sample of observed bursts, the MultiINstrument Burst ARchive (MINBAR;
http://burst.sci.monash.edu/minbar)
• The KEPLER model runs are expensive (a ~week for each
burst train!) so we adopt a large sample of simulated bursts
Lampe et al. 2016, ApJ 819, 46
• Use an MCMC approach via Foreman-Mackay‘s emcee
package, and marginalise over the input parameters
(distance, mass, radius, fuel composition) to determine
posterior probability distributions
Galloway: Observations and modelling of thermonuclear bursts
Preliminary MCMC approach
recurrence time comparison
Galloway: Observations and modelling of thermonuclear bursts
• Code is under development to
perform the matching,
identifying the best-fitting
model curves, and derive the
posterior distributions on the
system parameters.
• Representative distributions
for distance and gravitational
redshift from a preliminary
single-epoch match, including
the recurrence time, are
shown here.
Summary and next steps
• Matching burst models and observations requires a more
comprehensive approach
• Need to compare burst lightcurves as well as energetics and
recurrence times at all possible epochs
• This has been limited in the past due to the lack of
availability of model runs…
• … but also the lack of suitable analysis approaches/tools
• The pieces are largely in place now to carry out
comprehensive model-observation comparisons and get to
the point where we can constrain nuclear reaction rates
Galloway: Observations and modelling of thermonuclear bursts
Calibrating burst models
• Another important activity is quantifying the uncertainty in
our predictions of burst lightcurves (beyond that introduced
from nuclear physics)
• To this end we are working on a set of test cases for
numerical models
• Numericists with different codes will be encouraged to test
their codes against our best estimates for the system
parameters (accretion rate etc.) of these objects
• Can compare code results directly against each other
• Will be published early next year
Galloway: Observations and modelling of thermonuclear bursts