Dragonfly Pulsar analysis CristianPozo

General information

• Name of the source: Dragonfly Pulsar (PSRJ2021+3651)
• Brief description of the source:

- Object type : Pulsar

- RA, Dec in deg (ICRS): 305.27252, 36.85126

People involved in this analysis (LST-1 and new Fermi-LAT analysis)

• Cristian Pozo González
• Alvaro Mas-Aguilar
• Ruben Lopez-Coto
• Daniel Morcuende

Run selection

• Run selection with the use of the notebook from Abelardo Moralejo [1]
• Atmospheric transmission extracted from ELOG
• In total 48 runs selected, obstime 13.51 hours

• Summary of selection cuts:

- Zenith < 70 deg and Zenith < 20 deg

- elapsed_time > 2 min

- transmission_cut > 0.60

- pedestal charge std dev < 1.9 p.e.

- Cosmic rate > 3000 ev/s

- Cosmic rate( > 10 p.e.) > 25-15 ev/s (depends on zd)

- Cosmic rate( > 30 p.e.) > 3-4 ev/s (depends on zd)

• List of selected runs (Zenith < 70 deg):

1  :  2022-06-28  :  [8948 8949 8950 8951 8952 8953 8954 8955 8956 8957 8958 8959]
2  :  2022-06-29  :  [8980 8982 8983 8985 8986 8987]
3  :  2023-05-16  :  [13062 13063 13064 13066 13067]
4  :  2023-05-18  :  [13096 13097 13098 13099]
5  :  2023-06-22  :  [13519 13520 13521 13522]
6  :  2023-06-25  :  [13588 13589 13590 13591]
7  :  2023-10-05  :  [14910]
8  :  2023-10-06  :  [14918 14919 14920 14921 14922 14923 14924]
9  :  2023-11-13  :  [15489 15490 15491 15495 15497]

• Number of runs (Zenith < 20 deg)(% is w.r.t. those in Sky region & zenith range):

   In the requested Sky region and range of dates:	 132
 + zenith in requested range:				 77
 + NSB in requested range:				 76 (98.7%)

 + FF and pedestal interleaved events are present:	 75 (97.4%)
 + Stable pointing:					 74 (96.1%)
 + dR/dI fit P-value ok:				 73 (94.8%)
 + dR/dI LS periodogram ok:				 73 (94.8%)
 + dR/dI index ok:					 72 (93.5%)
 + dR/dI rate ok:					 31 (40.3%)
 + intensity threshold ok:				 31 (40.3%)

Note: about 64% of all *dark-night* observations within ZD<80 deg fulfill all quality cuts. (in the stable, good-quality period 20221118 - 20230214, 92% of *dark-night* observations within ZD<80 deg do).

• List of selected runs (Zenith < 20 deg):

1  :  2022-06-28  :  [8951 8952 8953 8954 8955 8956 8957 8958 8959]
2  :  2022-06-29  :  [8980 8982 8983 8985 8986 8987]
3  :  2023-05-16  :  [13066 13067]
4  :  2023-05-18  :  [13098 13099]
5  :  2023-06-22  :  [13520 13521 13522]
6  :  2023-06-25  :  [13588 13589 13590 13591]
7  :  2023-10-05  :  [14910]
8  :  2023-10-06  :  [14918 14919 14920 14921]

DL1 data

- DL1a files produced by LSTOSA (lstchain v0.9.9)

- lstchain v0.9 tailcut8-4

- /fefs/aswg/data/real/DL1/{date}/v0.9/tailcut84/

DL2 data

-

IRFs

- Point-like

- IRFs for each MC node and for each intensity cut

-

DL3 data

-

DL3 pulsar data

- Add phases using PINT-pulsar tools for LST-1 (https://github.com/alvmas/PulsarTimingAnalysis)

- Ephemeris file:

- DL3 pulsar data:

High-level analysis

1D Spectral analysis

• Performed with gammapy-v0.20.1

energy_axis = MapAxis.from_energy_bounds(
   0.01, 10, nbin=40, per_decade=False, unit="TeV", name="energy"
)
energy_axis_true = MapAxis.from_energy_bounds(
   0.05, 100, nbin=100, per_decade=False, unit="TeV", name="energy_true"
)

spectral_points_binning 
   e_min, e_max = 0.02, 0.7
   nbins = 8 
   logspace
   sqrt_ts>1.5

• Phase Regions:

 Bkg: [0.52,0.87]
 P1: [0,0.026,0.983,1]
 P2: [0.377,0.422]
 P1+P2: [0,0.026,0.377,0.422, 0.983,1]
 P3: [0.08,0.24]

• Spectral fitting of stacked LST dataset
• Performed for point-like assumption
• Stacked analysis
• LST-1 data alone: Power-law spectral model
• Fermi + LST-1 data: Power-Law Spectral Model, ExponentialCutOff, LogParabola

Results

Phaseogram

• Phaseogram with all data

• Phaseogram with zd<35

• Phaseogram with zd>35

• Phaseogram in different energy bins:

• Evolution of the signal in time (without zd cut):

• Fitting results

• P1/P2 differential ratio (with Fermi-LAT). Until 200 GeV due to large uncertainties

• P1/P2 integral ratio (with Fermi-LAT).

LST-1 only fit and SED (zd<50)

• P1

• P2

P1 LST-1 + Fermi-LAT joint fit

• • Using a SubExponentialCutOffPL model:

Total stat: -2logL = 32.56

• • Using a SmoothBrokenPowerLaw model

Total stat: -2logL = 24.44

• P1 for E>10 GeV:

• • Using a LogParabola model

Total stat: -2logL = 7.34

• • Using a PowerLaw model

Total stat: -2logL = 7.83

P2 LST-1 + Fermi-LAT joint fit

• • Using a SubExponentialCutOffPL model:

Total stat: -2logL = 51.40

• • Using a SmoothBrokenPowerLaw model

Total stat: -2logL = 34.74

• P2 for E>10 GeV:

• • Using a LogParabola model

Total stat: -2logL = 17.85

• • Using a PowerLaw model

Total stat: -2logL = 19.35

LST-1+ Fermi-LAT energy calibration

• P1 in left and P2 in right (Using SmoothBrokenPowerLaw model in all energy range)

• P1 in left and P2 in right (Using PowerLaw model at E>10 GeV)

In all the cases the energy factor that minimize the chi2 value is below 5%

Lightcurve

Systematics tests

Phaseogram and SED for different samples

• Phaseogram Before August 2021

sign(P1)/sqrt(Tobs) = 0.856 h^(-1/2)

sign(P2)/sqrt(Tobs) = 1.059 h^(-1/2)

• Phaseogram After August 2021

sign(P1)/sqrt(Tobs) = 1.063 h^(-1/2)

sign(P2)/sqrt(Tobs) = 1.183 h^(-1/2)

• SED for P1:

Difference in spectral index of ~10% and flux 60% (lacking some statistics for first sample)

• SED for P2:

Difference in spectral index of ~2% and flux of 15%

Phaseogram and SED with different efficiencies

• P1 SED:

Relative maximum error of 3.07% in spectral index and 35.4% in flux.

• P1 SED:

Relative maximum error of 7.36% in spectral index and 29.2% in flux.

Effect on SED due to shift on MC true energy =

• • P1:

Relative maximum error of 1.01% in spectral index and 24.9% in flux.

• • P2:

Relative maximum error of 0.45% in spectral index and 16.4% in flux.

Final uncertainties

• • P1:

• • P2: