1、 Rep. ITU-R RA.2099 1 REPORT ITU-R RA.2099 Radio observations of pulsars for precision timekeeping (Question ITU-R 205/7) (2007) Scope This Report examines the possibility of using high-precision timing radio observations of millisecond pulsars for constructing and maintaining new pulsar-based astro
2、nomical time scales. No changes in the Radio Regulations (RR) are needed to enable this activity. 1 Introduction Pulsars are identified with strongly magnetized, rapidly-rotating neutron stars. The presently known pulsars have masses of order 1.5 times the solar mass, diameters of about 20 km and sp
3、in rotation periods from 1.34 ms up to 8 s. Clearly pulsars have large moments of inertia and large stores of rotational energy, and so can be treated as space “flywheels” with stable rotation periods that can be used for precision astronomical timekeeping Manchester and Taylor 1977. In 1993, ITU re
4、cognized the potential use of pulsars for precision timekeeping, and adopted Question ITU-R 205/7, as well as an Opinion ITU-R 99 “Time-scale based on pulsar timing” (2003). Pulsars, as sources of regular radio pulses, have “life times” of millions to billions of years. There are two well-known grou
5、ps of pulsars. The first of these are normally isolated objects, which usually have periods of 0.2 s to 8 s. The second class are very rapidly rotating pulsars, which are often in binary stellar systems, the so-called “millisecond” pulsars, with periods of 1.34 ms to 50 ms. At this time more than a
6、hundred of these systems are known. Millisecond pulsars are believed to originate when mass is accreted from a companion onto a neutron star. They are thus recycled old pulsars that have magnetic fields of about 104T (108 G). Pulsars in close binary systems have orbital periods that range from a few
7、 hours up to several months. Their orbital parameters can be determined by high-precision radio-timing observations. Some millisecond pulsars have spin-period instabilities as small as 0.2 s over five years, i.e. a fractional instability of 1015 . Their radiation losses are negligible, so that the r
8、otation period of some systems increases with as little as 1021 s/s (ie seconds per second), and usually linearly with time (ITU-R Handbook Radio Astronomy, 2ndedition 2003). Thus, pulsars are well suited to the role of providing mankind with highly regular space clocks, which permits the generation
9、 of new, pulsar-based, astronomical time-scales, both a Pulsar Time-scale (PT) and a Dynamic Pulsar Time-scale (DPT) Ilyasov, Kopeikin and Rodin, 1998. The extreme rotational stability of pulsars allows the application of a unique technique to increase the S/N ratio of pulsar profiles the “synchrono
10、us integration mode”, in which the signal is summed synchronously with the pulsar period. Precision pulsar timing programmes are being conducted at radio observatories in Australia, France, Germany, Japan, the Netherlands, the Russian Federation, the United Kingdom and the United States of America.
11、2 Rep. ITU-R RA.2099 2 Preferred frequency bands for high-precision timekeeping observations of radio pulsars Observations of pulsars are currently made in a wide range of frequencies, from 10 MHz up to 40 GHz. The basic achievable noise level in radio astronomical observations is defined primarily
12、in the metre wavelength range by Galactic background radiation, though at higher frequencies receiver noise dominates the total noise. The brightness temperature of the Galactic background decreases from several thousand kelvins (K) at frequencies around 100 MHz down to 1-10 K at 1 GHz, and is chara
13、cterized by a flux density S ( f ) f where the spectral index is about 2.5. On the other hand, the flux-density of pulsars decreases with frequency, following a spectral index of about 2 (on average). A low noise preamplifier in a pulsar receiver typically has a noise temperature of 10 K for receive
14、rs in the frequency range 1-10 GHz. So for using pulsars for high-precision timekeeping the optimal S/N is achieved by observing in the 0.4-2.0 GHz range Ilyasov et al, 1999. The S/N ratio increases with receiver bandwidth: the gain in observing sensitivity for a bandwidth f is proportional to .f It
15、 is well known that a pulsars pulses are dispersed as they propagate through the interstellar medium, such that the magnitude of the resulting delay in the pulse arrival time decreases as the square of the frequency. From this point of view, higher frequencies are preferable. The magnitude of the re
16、sulting delay depends on the electron density along the line of sight to the pulsar, and is characterized by the “dispersion measure” (DM). The effects of dispersion can be removed from the signal using techniques based on multichannel filter-bank receivers in the time domain or coherent de-dispersi
17、on in the frequency domain. Multipath scattering in the interstellar medium causes a broadening of pulses of radio emission from pulsars, which decreases approximately as the fourth power of the frequency. This is also an argument for using higher frequencies in pulsar timing, when possible. The dis
18、persion measure is usually not entirely stable, so precise pulsar timing at about the microsecond level requires observations, preferably simultaneously, in at least two frequency bands that are separated by an octave, to measure the changes in dispersion measure. The preferred frequency bands for h
19、igh-precision timing observations of pulsars for precision timekeeping are the 1 400-1 427 MHz radio astronomy service (RAS) band, in conjunction with either the 608-614 MHz band or the 406.1-410 MHz radio astronomy band, and/or in a few cases, the 2 690 to 2 700 MHz band. 3 Threshold levels of inte
20、rference Pulsars are generally weak radio sources. Clearly it is necessary to achieve a significant S/N to get a precise measurement of the time of arrival (TOA) of the pulse of a pulsar. High-precision pulsar timekeeping observations therefore need to be protected from harmful interference. The thr
21、eshold levels for interference detrimental to high-precision pulsar timing are those given in Table 2 of Recommendation ITU-R RA.769 for single-dish continuum observations. Rep. ITU-R RA.2099 3 4 The feasibility of frequency sharing with other services High-precision timing observations of pulsars i
22、ntended for maintaining a pulsar-based time-scale can usually be conducted using the frequency bands allocated to the RAS. Radio astronomy does not share the 1400-1427 MHz and the 2690-2700 MHz bands with any active service. In the 406.1-410 MHz band sharing is possible with the FIXED and MOBILE exc
23、ept aeronautical mobile services and in the 608-614 MHz band sharing is possible with the Terrestrial Broadcasting service (Region 1), the Mobile -satellite except aeronautical-mobile satellite (Region 2) and the FIXED, MOBILE, Radionavigation and Broadcasting services (Region 3) as RAS stations are
24、 located at remote sites, and mobile-satellite service links are Earth-to-space. 5 The most appropriate pulsars for use in high-precision timekeeping For the purpose of high-precision timekeeping, the most appropriate pulsars are those which have the highest radio power flux densities and the most s
25、table periods, and which are observable from both the northern and southern hemispheres. New pulsars are constantly being discovered, and some of these are in systems that can be used to advantage for precision timekeeping: ongoing pulsar surveys make any list of preferred pulsars dynamic. Currently
26、 the ensemble of pulsars that fulfils these constraints is listed in Table 1. TABLE 1 PSR Right ascen-sion 2000 Decli-nation 2000 Pulsar period (ms) Pdot 1015 (s/s) Pb(days) DM (pc cm3)S400mJy S600mJy S1400mJy S3000mJy 1 2 3 4 5 6 7 8 9 10 11 12 B1855+09 18:57: 36.393 09:43: 17.323 5.36210045 1.78 1
27、05 12.32 13.309 31 (16.3) 4.3 1.5 1.6 B1937+21 19:39: 38.558 21:34: 59.137 1.55780647 1.05 104- 71.040 240 (100) 16 4.0 2.17 J1640+2224 16:40: 16.742 22:24: 08.941 3.16331582 2.8 106175.4 18.426 37 (16) 3 0.7 2.1 J1713+0750 17:13: 49.530 07:47: 37.526 4.57013652 8.53 10667.82 15.989 36 (16) 3 0.8 2.
28、0 J0437-4715 04:37: 15.786 47:15: 08.462 05.7574518 5.73 1055.741 2.6469 550 300 142 (61.4) 1.1 J0613-0200 06:13: 43.973 02:00: 47.097 3.06184404 9.572 1061.198 38.779 21 7.3 1.4 (0.45) 1.5 J1024-0719 10:24: 38.700 07:19: 18.915 5.16220455 1.8529105- 6.491 4.6 4.2 0.66 (0.18) 1.7 J1744-1134 17:44: 2
29、9.391 11:34: 54.575 4.07454588 8.9405 106- 3.1388 18 16 3 (0.76) 1.8 J1909-3744 19:09: 47.438 37:44: 14.318 2.94710802 1.40261051.5334 10.3939 - - - - - 4 Rep. ITU-R RA.2099 Notes to Table 1: Column 1 Pulsar name (B-name refers to the B(1950.0) epoch, J-name to the J(2000.0) epoch) Columns 2, 3 Puls
30、ar coordinates (Right Ascension, Declination) Column 5 Pulsar period derivative (s/s) Column 6 Period of binary system (days) Column 7 Dispersion measure (pc cm3 ) , where pc = parsec = 3.087 1013km. Columns 8, 9, 10, 11 Average spectral power flux-density (1029W/m2 Hz) at, respectively, 400, 600, 1
31、 400 and 3 000 MHz. Values in brackets are calculated via spectral index , where the flux density: S ( f ) f Column 12 Spectral index NOTE 1 The differential dispersive delay across a bandwidth (BW) is calculated at any frequency f from the DM via: GHzMHz3.8 3sfBWDMtsso that for a BW of 1 MHz and a
32、DM of 10, the differential dispersive delay between the two edges of a 1 MHz band at 1.4 GHz is 30.25 s. 6 Conclusions This report answers Question ITU-R 205/7, which was formulated for exploring the use of high-precision timing observations of millisecond pulsars for constructing and maintaining ne
33、w pulsar-based astronomical time-scales PT and DPT. The preferred frequency bands for high-precision timing observations of radio pulsars for the purpose of precision timekeeping are the RAS bands 1 400-1 427 MHz, and either 406.1-410 MHz or 608-614 MHz, and/or the 2 690 to 2 700 MHz band. The thres
34、hold levels for interference detrimental to high-precision pulsar timing are those given in Table 2 of Recommendation ITU-R RA.769 for single-dish continuum observations. The aforementioned, preferred, RAS bands do not require any change in frequency allocations or in sharing arrangements with any o
35、f the active services sharing the bands with the RAS. The final goal of providing a new, long-term, stable time-scale by using the most appropriate pulsars as reference clocks is by making precision timing observations of the pulsars B1855+09, B1937+21, J1640+2224, J1713+0750, J0437-4715, J0613-0200
36、, J1024-0719, J1744-1134 and J1907-3744. New pulsars are constantly being discovered, and some of these are in systems that can be used to advantage for precision timekeeping: ongoing pulsar surveys make any list of preferred pulsars dynamic. This list of objects will undoubtedly be augmented in tim
37、e. Rep. ITU-R RA.2099 5 References ILYASOV, KOPEIKIN and RODIN 1998 Astronomy Letters, Vol. 24, p. 275. ILYASOV, KUZMIN, SHABANOVA and SHYTOV 1999 Pulsar time-scale, Lebedev Proc., Vol. 199. MANCHESTER and TAYLOR 1977 Pulsars, Freeman, San Francisco, CA. ITU Handbook on Radio Astronomy, 2nd edition, 2003.
