The Green Bank North Celestial Cap Pulsar Survey. IV: Four New Timing Solutions

TL;DR

This paper reports four new pulsar timing solutions from the GBNCC survey, highlighting an unusual low spin-down rate in one pulsar that may suggest unique evolutionary history, and provides insights into their distances and luminosities.

Contribution

The study presents four newly discovered pulsar timing solutions, including detailed analysis of PSR J0038-2501's low spin-down rate and potential evolutionary implications.

Findings

PSR J0038-2501 has an unusually low period derivative.

PSR J1949+3426 is among the most distant and luminous pulsars.

No X-ray emission detected from PSR J0038-2501, ruling out a young CCO.

Abstract

We present timing solutions for four pulsars discovered in the Green Bank Northern Celestial Cap (GBNCC) survey. All four pulsars are isolated with spin periods between 0.26$\,$s and 1.84$\,$s. PSR J0038$-$2501 has a 0.26$\,$s period and a period derivative of ${7.6} \times {10}^{-19}\,{\rm s\,s}^{-1}$, which is unusually low for isolated pulsars with similar periods. This low period derivative may be simply an extreme value for an isolated pulsar or it could indicate an unusual evolution path for PSR J0038$-$2501, such as a disrupted recycled pulsar (DRP) from a binary system or an orphaned central compact object (CCO). Correcting the observed spin-down rate for the Shklovskii effect suggests that this pulsar may have an unusually low space velocity, which is consistent with expectations for DRPs. There is no X-ray emission detected from PSR J0038$-$2501 in an archival swift observation, which suggests that it is not a young orphaned CCO. The high dispersion measure of PSR J1949+3426 suggests a distance of 12.3$\,$kpc. This distance indicates that PSR J1949+3426 is among the most distant 7% of Galactic field pulsars, and is one of the most luminous pulsars.

Tables

Table 1. Table 1: Timing Solutions and Derived Parameters

Parameter	J0038$-$2501 (catalog PSR J0038-2501)	J1916$-$2939 (catalog PSR J1916-2939)	J1949+3426 (catalog PSR J1949+3426)	J2355+2246 (catalog PSR J2355+2246)
Right Ascension (J2000)	${00}^h{38}^{m}10\stackrel{s}{.}264\left(10\right)$	${19}^h{16}^{m}32\stackrel{s}{.}701\left(5\right)$	${19}^h{49}^{m}13\stackrel{s}{.}671\left(6\right)$	${23}^h{55}^{m}49\stackrel{s}{.}8\left(3\right)$
Declination (J2000)	$-25°01\prime 30\stackrel{\prime \prime }{.}73\left(2\right)$	$-29°39\prime 27\stackrel{\prime \prime }{.}8\left(3\right)$	$+34°26\prime 33\stackrel{\prime \prime }{.}89\left(8\right)$	$+22°46\prime 17\left(8\right)″$
Galactic Longitude (deg)	67.42	8.25	69.72	106.53
Galactic Latitude (deg)	$-86.35$	$-18.07$	$4.29$	$-38.32$
Dispersion Measure (pc cm-3)	5.710(3)	38.34(11)	228.0(3)	23.1(7)
NE2001 Distance (kpc)	0.32	1.2	9.8	1.2
YMW16 Distance (kpc)	0.60	1.6	12.3	2.2
Spin Period (s)	0.2569264575329(17)	1.248616964290(3)	0.3885391675859(14)	1.8409859072(3)
Period Derivative (${10}^{-17}s{s}^{-1}$)	0.0760(6)	124.41(4)	20.219(8)	378(4)
Epoch (MJD)	57474	57346	56921	57102
Span of Timing Data (MJD)	56774 – 58175	56901 – 57791	56051 – 57791	56477 – 57666
Number of TOAs	88	91	42	34
RMS Fit Residual ($\mu$s)	120	422	494	3061
EFAC	1.3	0.81	0.91	1.5
Characteristic Age (Myr)	5400	16	30	7.7
Surface Magnetic Field (${10}^9$ G)	14	1300	280	2700
Spin-down Luminosity (${10}^{30}\mathrm{ergs}{s}^{-1}$)	1.8	25	140	9.6
Signal to Noise	81	18	23	12
Pulse Width, W10 (s)	0.015	0.061	0.030	0.115
$T_{\mathrm{sky}}$ (K)	27	70	77	28
$T_{\mathrm{sys}}$ (K)	73	116	123	74
$S_{350}$ (mJy)	3.7	2.2	3.7	0.9
$L_{350}$ (mJy kpc2)	1.3	6.4	570	4.1

Equations

$$S_{\min }=\frac{(S/N)T_{\mathrm{sys}}}{G\sqrt{n_{\mathrm{p}}{t}_{\mathrm{int}}\Delta f}}\sqrt{\frac{W}{P-W}},$$

$$\dot{P_{\mathrm{S}}}=\frac{P{V}_{\mathrm{T}}^2}{cd},$$

Full text

pulsars: individual: J0038$-$2501, J1916$-$2939, J1949+3426, J2355+2246 – stars:neutron

††software: GSM (de Oliveira-Costa et al., 2010), PRESTO (http://www.cv.nrao.edu/~sransom/presto/), PSRCHIVE (Hotan et al., 2004; van Straten et al., 2012), TEMPO2 (Hobbs et al., 2006)),

1 Introduction

The Green Bank North Celestial Cap (GBNCC) survey (Stovall et al. 2014; Lynch et al. 2018; Kawash et al. 2018) is searching for pulsars and transient radio signals at 350 MHz in the declination ($\delta$) range available to the Robert C. Byrd Green Bank Telescope (GBT), $\delta>-40\arcdeg$. Scientific objectives for the GBNCC survey include characterization of the Galactic pulsar population as well as finding high precision millisecond pulsars (MSPs) suitable for inclusion in a pulsar timing array (PTA), which will enable the detection of nanohertz-frequency gravitational waves (GWs; e.g. Arzoumanian et al., 2018). By surveying the entire sky at low frequencies we are especially sensitive to nearby, low-luminosity and/or steep-spectrum pulsars; see Stovall et al. (2014) for further comparison of GBNCC’s sensitivity with other pulsar surveys. Note that the sensitivity of GBNCC also allows detection of more distant, higher luminosity pulsars as evidenced in the detection of J1949+3426 reported in this paper. As of 2018, GBNCC survey observations are approximately 80 percent complete, with the data collection expected to conclude by 2020. As of 2018 October, 161 pulsars, including 20 millisecond pulsars have been discovered by the GBNCC survey. Initial GBNCC survey results were reported in Stovall et al. (2014), while Kawash et al. (2018) discussed timing results for 10 pulsars and Lynch et al. (2018) reported timing results for an additional 45 pulsars. This paper reports results from analysis of timing observations of four pulsars discovered in the GBNCC survey. The University of Wisconsin–Milwaukee provided an opportunity for undergraduate students to participate in course-based research by processing data from observations on the four pulsars to develop timing solutions and to characterize the pulsars based on their properties111The pulsar timing analysis presented here is the culmination of efforts by the six lead authors, who participated in a First Year Research Experiences (FYRE) course held at the University of Wisconsin–Milwaukee during the Fall semester of 2017, PHYS 194: Clocking Dead Stars with Radio Telescopes.. In the discussion below, we give quantities and distances computed using both the Galactic electron density model of Yao et al. (2017, YMW16) and that of Cordes & Lazio (2002, NE2001).

2 Observations & Timing Analysis

The discovery observations for the new discoveries presented here took place between 2011 and 2015 and used the GBT operating at a center frequency of 350 MHz and nominal bandwidth of 100 MHz, with dwell times of 120 s; see Stovall et al. (2014) for a description of the methodology.

The search processing took place on a computer cluster operated by Compute Canada, with candidates analyzed via the CyberSKA interface222https://ca.cyberska.org/. The timing observations for the four pulsars presented in this paper used the same center frequency and bandwidth, with typical durations of 3.5–6 min. The Green Bank Ultimate Pulsar Processing Instrument (GUPPI; DuPlain et al. 2008) was used for both discovery and timing observations to record data every 81.92 $\mu$s with 4096 frequency channels. Data were processed using PRESTO (Ransom et al., 2002) for initial spin period refinement, then PSRCHIVE (Hotan et al., 2004; van Straten et al., 2012) to process individual timing scans and calculate times of arrival (TOAs).

An ephemeris was created to save the preliminary timing parameters. Using the dispersion measure (DM) found from the discovery observations, the files from all timing observations were averaged from 4096 to 256 frequency channels using pam and each file was examined using pazi to remove radio frequency interference (RFI). Figure 1 shows the composite profiles based on all timing observations for each pulsar. Standard profiles were created for each pulsar using paas on files with a high signal-to-noise ratio. All timing files were then averaged in frequency again using pam to one or more frequencies and three sub-integrations prior to using pat and the standard profiles to generate TOAs; the number of frequency sub-bands was one for observations with relatively low signal-to-noise ratio and two or three for observations with higher signal-to-noise ratio. Fitting the TOAs was performed with TEMPO2 (Manchester et al., 2015; Hobbs et al., 2006), finding a timing solution including spin period ($P$), period derivative ($\dot{P}$), and position. We also included DM as a free parameter for pulsars for which TOAs were available at multiple frequencies. Where not available from TEMPO2, DM errors were determined using the PSRCHIVE command pdmp. Parameter uncertainties quoted in Table 1 are 1-$\sigma$ uncertainties on measured TEMPO2 fit parameters, but a global multiplicative error factor (EFAC) has been applied to each TOA error such that the resulting reduced $\chi^{2}$ value is one after fitting. Discovery observations were included in the timing analysis for each pulsar after similar processing using PRESTO. After fitting model parameters, the profiles for TOAs with relatively large residuals were each examined visually to determine whether each was a significant detection. TOAs with no clear detection were deleted prior to the final model fit. The final TOA residuals are plotted in Figure 2. We confirmed that the discovery TOAs could be reliably phase-connected with the timing TOAs by ensuring that the phase uncertainty extrapolated to the time of the discovery observations was much less than 0.1 cycle.

Taken together, the data span of the combined discovery and timing data-sets are at least two years for each pulsar, so that covariances between spin-down and position are minimized. The final timing models are given in Table 1.

The locations on a $P-\dot{P}$ diagram are shown in Figure 3. Table 1 also includes the DM and calculated distance to each pulsar as well as the characteristic age, and calculated pseudo-luminosity for each pulsar. The flux density $S_{350}$ and pseudo luminosity $L_{350}=S_{350}\times d^{2}$ values reported in Table 1 were calculated for each of the timed pulsars using the signal-to-noise (S/N) from discovery observations as in Stovall et al. (2014), and the search sensitivity (Lorimer & Kramer, 2012)

[TABLE]

where $T_{\rm sys}$ is the system temperature as listed in Table 1, $G=2\,{\rm K\,Jy}^{-1}$ is the telescope gain, $n_{\rm p}=2$ is the number of polarizations summed, $t_{\rm int}=120\,$s is the integration time, $\Delta f=80$ MHz is the effective bandwidth, $W$ is the width of the pulse as detected by the system, and $P$ is the spin period. $T_{\rm sys}$ includes sky temperatures $T_{\rm sky}$ listed in Table 1 as determined for the direction of each pulsar using the global sky model of de Oliveira-Costa et al. (2008) calculated at 350 MHz.

3 Notes on Individual Pulsars

3.1 PSR J0038$-$2501

As can be seen in Figure 3, the $\dot{P}$ and $P$ for PSR J0038$-$2501 are low compared to typical non-recycled pulsars, implying a relatively low surface magnetic field strength ($B_{\rm s}$). It was also the closest of the four timed pulsars at $600\,$pc (from YMW16; $320\,$pc from NE2001).

Some pulsars with similar timing properties ($P>20$ ms and $B_{\rm s}<3\times 10^{10}$ G) are described as disrupted recycled pulsars (DRPs; Belczynski et al. 2010), where it is thought that the companion exploded in a supernova that unbound the system, stopping the recycling process and leaving the pulsar with intermediate properties between typical isolated pulsars and recycled MSPs (Gotthelf et al., 2013). However, there is overlap between the properties of more traditional isolated pulsars and the DRPs. Belczynski et al. (2010) estimates that 0.3% of isolated non-recycled pulsars may have $\dot{P}$ and $B_{\rm s}$ values similar to DRPs, which amounts to $\sim 4$ pulsars out of the total population compared to 12 DRP considered in that paper.

An alternative explanation for the low magnetic field properties of PSR J0038$-$2501 is that it could be an orphaned central compact object (CCO). CCOs are young pulsars with low magnetic fields that are found within or near supernova remnants (SNRs; Gotthelf et al. 2013). The characteristic ages of CCOs calculated from their spin-down rates do not match the known ages of their associated SNRs. It is unclear how CCOs evolve after they are formed. Gotthelf et al. (2013) proposed that CCO descendants may have similar timing properties to the DRP pulsars, but would be expected to be younger and therefore may have visible thermal X-ray emission for up to $\sim 10^{5}$ yrs. Gotthelf et al. (2013) and Luo et al. (2015) searched known pulsars classified as DRPs (using same criteria as in Belczynski et al. 2010) for X-ray emission with no detections. We searched archival observations and found a 6 ks Neil Gehrels Swift Observatory (Swift) observation from 2009 November 16 that included the location of PSR J0038$-$2501. Figure 4 shows the Swift observation, which has 0 counts in a circle of radius $18\arcsec$, consistent with the background expectation of $1.4\pm 1.2\,$counts determined from the background rate of $0.0014\pm 0.00005\,{\rm counts\,arcsec}^{-2}$. Using Gehrels (1986) we set an upper limit of $<3$ counts (95% confidence). Figure 5 compares the X-ray luminosities of young CCOs and old isolated pulsars with the upper limit for the Swift observation as well as upper limits for candidate DRPs from Gotthelf et al. (2013) and Luo et al. (2015). The failure to detect sources for these observations suggests that PSR J0038$-$2501 and the candidate DRPs are not young ($\sim 10^{5}$ yrs old) orphaned CCOs. PSR J0038$-$2501 is the fifth potential DRP to be discovered in the GBNCC survey after PSRs J0358+6627, J0557$-$2948, J1434+7257, and J2122+5434 (Stovall et al., 2014; Lynch et al., 2018; Kawash et al., 2018).

The low $\dot{P}$ and low estimated distance for PSR J0038$-$2501 also place limits on its proper motion and transverse velocity. The observed period derivative for pulsars includes a component associated with proper motion determined by the Shklovskii effect (Shklovskii, 1970), calculated with

[TABLE]

where $V_{\rm T}$ is the transverse velocity, $c$ is the speed of light and $d$ is distance to the pulsar. Additional period derivative components are present due to Galactic acceleration effects as described in Nice & Taylor (1995), but these components were found to be less than 7% of the measured period derivative and therefore were not included for this estimate. Using this calculation and assuming that the intrinsic $\dot{P}$ is greater than 0 (i.e., the pulsar is spinning down), we limit $V_{\rm T}$ to $<130\,{\rm km\,s}^{-1}$ for PSR J0038$-$2501 (from YMW16; $<90\,{\rm km\,s}^{-1}$ from NE2001). A relatively low transverse velocity is consistent with the lower expected natal kick velocities for DRPs (Belczynski et al., 2010), (3-D velocity dispersion of 170 ${\rm km\,s}^{-1}$ compared to 265 ${\rm km\,s}^{-1}$ for isolated pulsars).

3.2 PSR J1949+3426

PSR J1949+3426 has a dispersion measure of 228.0 pc cm*-3*  which makes it the farthest of the four timed pulsars and among the top 7% most distant Galactic field pulsars. The distance determined from the YMW16 model is 12.3 kpc (9.8 kpc from NE2001 model). The pulse profile shown in Figure 1 has a tail resembling profiles that exhibit scatter broadening as described in Bhat et al. (2004) which would be consistent with a large distance. We fit a one-sided exponential function to the pulse profile after the peak in four sub-bands centered at 313, 338, 358, and 389 MHz. We find decay times of 30.1$\pm{0.8}$, 20.1$\pm{0.4}$, 15.7$\pm{0.5}$, and 15.1$\pm{0.7}$ ms, although we note that this does not account for any intrinsic pulse width or frequency evolution of the pulse shape. The timescale decreases with increasing frequency as expected for interstellar pulse broadening, but it should be noted that the range of frequencies was relatively narrow for this evaluation. The NE2001 model predicts a much lower pulse broadening timescale of $<1\,$ms at 350 MHz, while the YMW16 model predicts a larger timescale of $60\,$ms. Additional observations at a higher frequency could help confirm whether the profile shape is impacted by pulse broadening. The large distance for PSR J1949+3426 suggests that it must be relatively bright to have been detected in the GBNCC search. The pseudo-luminosity (calculated as described in Section 3) is one of the highest reported for pulsars discovered by GBNCC (Stovall et al., 2014).

3.3 PSR J1916$-$2939

PSR J1916$-$2939 has properties typical for isolated non-recycled pulsars with a longer, 1.84 s period and large period derivative due to a relatively high surface magnetic field.

3.4 PSR J2355+2246

PSR J2355+2246 also has properties typical for young, isolated non-recycled pulsars. The signal-to-noise ratio in the observations was relatively low at $\sim 12$. Several TOAs were removed after confirming that no significant pulse profile was visible. Some evidence of pulse nulling (where the pulsar appears to turn off for a some numbers of pulses; Backer, 1970) was noted for roughly 30% of the pulses in a 2 minute discovery observation. We reviewed additional timing observations for nulling behavior, but were unable to confirm this behavior because of excess RFI. Additional observations can confirm whether this pulsar nulls or not.

4 Conclusions

In this paper, we report the timing solutions for four pulsars discovered in the GBNCC survey. The properties of the timed pulsars are varied indicating differing evolutionary paths, which supports one of the GBNCC objectives of characterizing the pulsar population to better understand the underlying physical phenomena. PSR J0038$-$2501 was found to have an unusually low magnetic field suggesting that it may be a DRP or possibly an orphaned CCO. An archival Swift X-ray observation did not find a source at the location, suggesting that PSR J0038$-$2501 is not a young orphaned CCO, but it could be an older source. Additional observations are suggested to determine the proper motion of PSR J0038$-$2501 which may help distinguish between evolutionary models. The farthest of the four pulsars according to the DM-distance models reported here was PSR J1949+3426. Calculations indicate that it may be one of the highest pseudo-luminosity pulsars discovered in the GBNCC survey. The profile may show evidence of pulse broadening. Observations at higher frequency would allow better evaluation of the intrinsic pulse profile and determination of the extent of scattering.

The undergraduate students would like to thank the University of Wisconsin–Milwaukee for providing this research opportunity through the First Year Research Experience (FYRE) initiative. We thank the referee for their time and comments. The Green Bank Observatory is a facility of the National Science Foundation operated under cooperative agreement by Associated Universities, Inc. Support was provided by the NANOGrav NSF Physics Frontiers Center award number 1430284. Pulsar research at UBC is supported by an NSERC Discovery Grant and by the Canadian Institute for Advanced Research. This research has made use of data obtained through the High Energy Astrophysics Science Archive Research Center Online Service, provided by the NASA/Goddard Space Flight Center. JvL acknowledges funding from the European Research Council grant n. 617199. GBT, Swift