Sur ce site

Sur le Web du CNRS

Accueil du site > ANGLAIS > Research > Astroparticles > Gamma Ray Pulsar Research at the CENBG

Gamma Ray Pulsar Research at the CENBG

Gamma Ray Pulsar Research at the CENBG

Fermi (né GLAST) has allowed, and continues to allow, the Bordeaux group to study GeV gamma ray pulsars vastly better than what was begun with CELESTE. Preparing for pulsar discoveries after launch has been our principal focus since the end of 2005. We have been quite successful, as for example the "2nd Fermi Pulsar Catalog", written in large part in Gradignan, shows.

We have been roughly a quarter of the overall effort made for pulsars within the LAT collaboration, and undertook a daunting array of tasks, building on instrumental and pulsar know-how acquired with CELESTE and during the construction of the calorimeter. The tasks are the following:

A. Creation and maintenance of a database of radio pulsar ephemerides

B. Pre-launch validation of the gamma ray time stamps

C. Post-launch validation and monitoring of pulsar phase measurements

D. Post-launch validation and monitoring of the gamma ray effective area

E. Gamma ray identification optimized for pulsar studies, via Monte Carlo and testbeam data.

F. Post-launch energy calibrations

G. Interpretation of measurements on normal and on millisecond pulsars

We will now describe briefly these several activities.

A. Creation and maintenance of a database of radio pulsar ephemerides

The LAT has an unprecedented wide field-of-view (20% of the sky at any moment) and will scan the entire sky every three hours during the first year. Hence, all pulsars are visible nearly all the time. Yet, in spite of a detection area 25x greater than EGRET’s, the expected rate for gamma ray pulsars is low (one every 15 seconds on the Crab, and a few per day on PSR B1951+32). Detection sensitivity is greatly enhanced if the pulsar frequency is known accurately a priori. Therefore, as was done for EGRET [15], an extensive campaign of radio pulsar monitoring has been undertaken for Fermi. The problem is exacerbated by the fact that the best candidates for detectable gamma emission are young pulsars decelerating rapidly (a large spin-down power Edot), which happen also to be the pulsar for which the rotation period varies unpredictably over weeks to months. Therefore, the gamma pulsar candidates need to be monitored regularly over the 5 to 10 years of the GLAST mission. Further details are in [11].

Three of the world’s largest radio telescope have committed to this task: Parkes in Australia, for the southern sky, and Jodrell (England) and Nancay (France) for the northern sky. These 3 have tracked approximately 800 gamma ray pulsar candidates. A small number of especially promising gamma candidates are too weak in radio to be detected by those 3 telescopes. Therefore, teams working with the Arecibo and Green Bank radiotelescopes, as well as with the Rossi X-ray Timing Explorer satellite (RXTE) provided the rotation parameters for those pulsars. Some details are at [16].

The CENBG played a key role in getting this process started, and was awarded the responsibility of collecting the timing measurements from the various groups and re-formatting them into the LAT-designated data format, which will be stored on the public data servers at GSFC. A thesis student worked on this 3/4 time, and two staff scientists are working on it 1/3 time each.

B. Pre-launch validation of the gamma ray time stamps

The arrival time of gamma rays to the satellite must be recorded to microsecond accuracy to allow GLAST to achieve its pulsar science goals. Such precision is difficult, and various major satellite missions have had hardware problems identified only after launch.

The CENBG proposed an end-to-end ground test of the LAT timing circuitry, using a pair of scintillators to detect cosmic ray muons, thereby triggering the GPS clock previously used for CELESTE. The NASA project approved the test and a proof-of-principle demonstration was conducted by one of us at General Dynamics’ facility in Arizona, in November 2006, along with two colleagues from the NRL [17]. The final tests are programmed for early Spring.

C. Post-launch validation and monitoring of pulsar phase measurements

The hardware and software chains leading from the radio telescopes and the gamma ray detection to the pulsar phase determinations are long and complex. The natural extension of the CENBG’s pre-launch efforts is to test the chain during the Launch & Early Operations phase of the mission (L&EO lasts for 60 days after launch). The validations will use the brightest known gamma ray pulsars, e.g. Vela and the Crab, as standard candles, and will be performed in close collaboration with the radio astronomers, who plan to make daily measurements during this critical phase. Two staff scientists and one student are preparing these efforts, in coordination with the GSFC and Stanford teams.

D. Post-launch validation and monitoring of the gamma ray effective area

Raw data from the LAT is "reconstructed", yielding analysis variables with information on the particle direction, energy, and species. Events are selected using these variables, enhancing the gamma ray signal to cosmic ray background ratio. The efficiency of these "cuts" for gamma rays is determined using detailed Monte Carlo simulations of the detector response. The effective detection area of the LAT, A_eff, is essentially the geometrical area (1.4 x 1.4 meters), multiplied by these efficiencies. Finally, determining the gamma ray flux of an astrophysical source in absolute units consists in dividing the observed gamma ray rate by A_eff.

This oversimplified description underscores that flux determination needs reliable Monte Carlo results. Huge efforts within the collaboration have been made to render the simulations highly reliable. Our group proposed to use L&EO Vela and Crab observations to perform anend-to-end comparison of Monte Carlo with real data. The proposal was accepted. At the first GLAST Symposium we presented two posters outlining our initial results [18]. At the 2014 5th Fermi symposium in Nagoya, Japan, we presented yet more gamma-ray pulsars.

E. Gamma ray identification optimized for pulsar studies, via Monte Carlo and testbeam data.

Pulsar analysis differ from e.g. blazar studies in that a) the background of diffuse galactic gamma rays is much higher and b) the timing signature provides additional sensitivity. A natural outgrowth of testbeam and Monte Carlo effective area studies is to explore analyses better suited to pulsar research. This work has begun and is ongoing. These last two tasks are pursued by two staff (1/3 time each) and a thesis student.

F. Post-launch energy calibrations

After the Montpellier group joined Fermi, a large fraction of the in-flight calibration tasks were transferred from the CENBG to the new group, but still overseen by the NRL. The basic calibration scheme exploits the ionization signal of cosmic heavy ions in the individual CsI crystals of the calorimeter. CENBG personnel is helping to prepare calibrations during L&EO, and beyond.

G. Interpretation of measurements on normal and on millisecond pulsars

The long list of technical tasks fits naturally with our goals for pulsar research. The fundamental observables that the LAT will provide are as follows: i) the numbers of different kinds of pulsars detected (normal, millisecond, and radio-quiet)

ii) their light curve morphology (the EGRET pulsars tended to have two peaks, separated by 0.4 revolutions of the neutron star, and offset by about 0.1 in phase relative to the main radio peak)

iii) their spectra, resolved by phase when the gamma counting rate provides adequate statistics, and especially the energy at which the spectra cut off.

These observable map to a range of open questions pertaining to pulsars and to more general topics in astrophysics. A few examples are

a) the populations of neutron stars in the Galaxy

b) determining the region around the neutron star where charged particle acceleration occurs, and photons are radiated, as well as the detailed mechanisms.

These considerations enter into the above list of technical tasks. Choosing which of the 2000 known pulsars are gamma candidates worthy of close radio monitoring takes models to build predictions. Cut optimization requires modelling of the expected pulsar spectra, as well as of the various background, some of which can come from a supernova remnant hosting the pulsar. Hence, all personnel in the group are directly concerned with applying the results from the literature in our simulations. Two other posters from our group illustrate these statements [19,20].

Summary =======

High energy pulsar research has been more or less stalled for a decade, awaiting observations beyond those acquired with the Compton satellite. GLAST will be launched at the end of 2007, and the Bordeaux group is at the heart of preparations for what promises to be a major advance in the field.

Bibliography ============

11. Contribution to the proceedings of Joint Discussion 02 ``On the Present and Future of Pulsar Astronomy’’ held at the IAU General Assembly in Prague, 16-17 August 2006.



14. B. Lott, F. Piron et al: Nucl. Instr. Meth. 560 395-404 (2006)

15. Z. Arzoumanian et al.: ApJ \textbf422, 671-680 (1994)


17. Smith, Grove, and Dumora LAT-TD-08777-03

18. D. Parent et al, and T. Reposeur et al, 1st International GLAST Symposium

19. L. Guillemot, V. Lonjou et al, 1st International GLAST Symposium

20. Smith, Grove et al, poster at "Texas in Australia" (2006)