geometryof γrayloudpulsarsfromradioobservaons

Geometry of γ-‐ray loud pulsars from radio observa5ons Simon Johnston, Simon Rookyard and Patrick Weltevrede

CSIRO ASTRONOMY & SPACE SCIENCE

CSIRO Astronomy and Space Science, Sydney, Australia

Mo5va5on • Where does the pulsar gamma-‐ray emission come from? • Where does the pulsar radio emission come from? •  How are the two linked? à Geometry is the key to unlock these secrets!

Bonn Neutron Stars -‐ June 2016 | Simon Johnston

Important angles are: α – angle between the rotation and magnetic axis β – closest approach of line of sight to magnetic axis ρ – cone opening angle

Mo5va5on • Where does the pulsar gamma-‐ray emission come from? • Where does the pulsar radio emission come from? •  How are the two linked? à Geometry is the key to unlock these secrets!

Expecta5ons •  γ-‐ray: High α expected for γ-‐ray loud (young) pulsars. •  Radio: Random distribuSon of α at birth à sinusoidal distribuSon as observed. Beaming fracSon higher for high α. β is small in radio pulsars. •  γ-‐ray+radio: Expect strong bias towards high α in the sample.


The Parkes Fermi 5ming programme •  Started in February 2007 [pre-‐launch] •  Patrick Weltevrede, Ryan Shannon, Ma^hew Kerr postdocs •  165 pulsars, vast majority with Edot > 1034 erg/s •  Observe every month for 21 hours at 1400 MHz •  And every 6 months at 700 and 3100 MHz

•  Timing soluSons sent to Fermi collaboraSon • Wealth of science from radio data alone •  Full polarizaSon profiles à pulsar geometry •  Dispersion measure versus Sme •  Profile variability versus Sme •  Timing noise and glitches •  Limits on planets etc etc


PuOng it together: PSR 1420-‐6048


Best fit: α~20, β~3

Results – 25 γ-‐ray detected pulsars


SURPRISE!! We find low values of α with more than ½ the pulsars showing α < 40 contrary to expectaSons (dashed line).

15/25 gamma-‐ray detected pulsars have α < 40 deg !

Results

Problem in a nutshell: The radio profiles are too wide for the pulsars to be close to orthogonal rotators.


SURPRISE!! We find low values of α with more than ½ the pulsars showing α < 40 contrary to expectaSons.

15/25 gamma-‐ray detected pulsars have α < 40 deg !

Our solu5on (a^er trying many things)


α now consistent with a random (sinusoidal) distribuSon at birth.

Make the polar cap extend beyond its nominal radius and make it dependent on α

Results – Part 2


(1) Gamma-‐ray detected pulsars have smaller (radio) widths than non gamma-‐ray detected pulsars.

(2) Gamma-‐ray peak separaSon and radio width are related

Summary – I. • Geometry is criScal to understanding the emission we see from pulsars (cf latest force-‐free models). • We have developed a method for deriving robust values of α and β. • DistribuSon of α is very different than our expectaSons for γ-‐ray detected pulsars • Either •  α really is small à γ-‐ray models have problems •  Radio polarizaSon info is misleading à radio models have problems •  Fix by postulaSng emission from outside the convenSonal polar cap region. ImplicaSons for joint radio/γ-‐ray modeling.


Summary – II. • All the high Edot pulsars emit γ-‐rays BUT • α is smaller for non γ-‐ray detected pulsars and this, coupled with small β is the reason for their non-‐detecSon in γ-‐rays (see also Guillemot & Tauris 2014 for MSPs!). • Narrow, widely space double γ-‐ray profiles suggesSve of high α indeed have narrow radio profiles. • Radio emission height is Edot dependent (cf the Karastergiou & Johnston beam model).


ASTRONOMY & SPACE SCIENCE Simon Johnston Head of Astrophysics t +61 2 9372 4573 e [email protected] w www.atnf.csiro.au

Thank you

CSIRO ASTRONOMY & SPACE SCIENCE

Read the papers! Rookyard et al. astro-‐ph 1410.3294 (data paper) Rookyard et al. astro-‐ph 1410.3310 (interpreta5on paper) Rookyard et al. just submided (comparison paper)

geometryof γrayloudpulsarsfromradioobservaons

Documents