Aris Karastergiou

Evidence for alignment of the rotation and velocity vectors in pulsars – II. Further data and emission heights

We have conducted observations of 22 pulsars at frequencies of 0.7, 1.4 and 3.1 GHz and present their polarization profiles. The observations were carried out for two main purposes. First, we compare the orientation of the spin and velocity vectors to verify the proposed alignment of these vectors by Johnston et al. We find, for the 14 pulsars for which we were able to determine both vectors, that seven are plausibly aligned, a fraction which is lower than, but consistent with, earlier measurements. Secondly, we use profiles obtained simultaneously at widely spaced frequencies to compute the radio emission heights. We find, similar to other workers in the field, that radiation from the centre of the profile originates from lower in the magnetosphere than the radiation from the outer parts of the profile.

Absolute polarization position angle profiles of southern pulsars at 1.4 and 3.1 GHz

We present here a direct comparison of the polarization position angle (PA) profiles of 17 pulsars, observed at 1.4 and 3.1 GHz. Absolute PAs are obtained at each frequency, permitting a measurement of the difference in the profiles. By doing this, we obtain more precise rotation measure (RM) values for some of the pulsars in the current catalogue. We find that, apart from RM corrections, there are small, pulse-longitude-dependent differences in PA with frequency. Such differences go beyond the interpretation of a geometrical origin. We describe in detail the PA evolution between the two frequencies and discuss possible causes, such as orthogonal and nonorthogonal polarization modes of emission. We also use the PA and total power profiles to estimate the difference in emission height at which the two frequencies originate. In our data sample, there are changes in the relative strengths of different pulse components, especially overlapping linearly polarized components, which coincide with intrinsic changes of the PA profile, resulting in interesting PA differences between the two frequencies.

Simultaneous single-pulse observations of radio pulsars - IV. Flux density spectra of individual pulses

In this paper we demonstrate that a large, unexplored reservoir of information about pulsar emission exists, that is directly linked to the radiating particles and their radiation process: We present a study of flux density measurements of individual pulses simultaneously observed at four different frequencies. Correcting for effects caused by the interstellar medium, we derive intrinsic flux density spectra of individual radio pulses observed at several frequencies for the first time. Pulsar B0329+54 was observed at 238, 626, 1412 and 4850 MHz, while observations of PSR B1133+16 were made at 341, 626, 1412 and 4850 MHz. We derive intrinsic pulse-to-pulse modulation indices which show a minimum around 1 GHz. Correlations between the flux densities of different frequency pairs worsen as the frequency separation widens and also tend to be worse for outer profile components. The single pulse spectra of PSR B0329+54 resemble the spectra of the integrated profile. However, the spectral index distributions for the single pulses of PSR B1133+16 show significant deviations from a Gaussian. This asymmetry is caused by very strong pulses with flux densities exceeding the mean value by more than a factor of ten. These strong pulses occur preferentially at the trailing edge of the leading component and appear to be broadband in most cases. Their properties are similar to those of so-called giant pulses, suggesting that these phenomena are related.

Polarization profiles of southern pulsars at 3.1 GHz

We present polarization profiles for 48 southern pulsars observed with the new 10-cm receiver at the Parkes telescope. We have exploited the low system temperature and high bandwidth of the receiver to obtain profiles which have good signal-to-noise for most of our sample at this relatively high frequency. Although, as expected, a number of profiles are less linearly polarized at 3.1 GHz than at lower frequencies, we identify some pulsars and particular components of profiles in other pulsars which have increased linear polarization at this frequency. We discuss the dependence of linear polarization with frequency in the context of a model in which emission consists of the superposition of two, orthogonally polarized modes. We show that a simple model, in which the orthogonal modes have different spectral indices, can explain many of the observed properties of the frequency evolution of both the linear polarization and the total power, such as the high degree of linear polarization seen at all frequencies in some high spin-down, young pulsars. Nearly all the position angle profiles show deviations from the rotating vector model; this appears to be a general feature of high-frequency polarization observations.

Simultaneous single-pulse observations of radio pulsars - III. The behaviour of circular polarization

We investigate circular polarization in pulsar radio emission through simultaneous observations of PSR B1133+16 at two frequencies. In particular, we investigate the association of the handedness of circular polarization with the orthogo- nal polarization mode phenomenon at two different frequencies. We find the association to be significant across the pulse for PSR B1133+16, making a strong case for orthogonal polarization modes determining the observed circular polarization. The association however is not perfect and decreases with frequency. Based on these results and assuming emission occurs in super- posed orthogonal polarization modes, we present a technique of mode decomposition based on single pulses. Average profiles of the polarization of each mode can then be computed by adding the individual mode-separated single pulses. We show that decomposing single pulses produces different average profiles for the orthogonal polarization modes from decomposing aver- age profiles. Finally, we show sample single pulses and discuss the implications of the frequency dependence of the correlation of the circular polarization with the orthogonal polarization mode phenomenon.

An empirical model for the polarization of pulsar radio emission

We present an empirical model for single pulses of radio emission from pulsars based on Gaussian probability distributions for relevant variables. The radiation at a specific pulse phase is represented as the superposition of radiation in two (approximately) orthogonally polarized modes (OPMs) from one or more subsources in the emission region of the pulsar. For each subsource, the polarization states are drawn randomly from statistical distributions, with the mean and the variance on the Poincare sphere as free parameters. The intensity of one OPM is chosen from a lognormal distribution, and the intensity of the other OPM is assumed to be partially correlated, with the degree of correlation also chosen from a Gaussian distribution. The model is used to construct simulated data described in the same format as real data: distributions of the polarization of pulses on the Poincare sphere and histograms of the intensity and other parameters. We concentrate on the interpretation of data for specific phases of PSR B0329+54 for which the OPMs are not orthogonal, with one well defined and the other spread out around an annulus on the Poincare sphere at some phases. The results support the assumption that the radiation emerges in two OPMs with closely correlated intensities, and that in a statistical fraction of pulses one OPM is invisible.

An investigation of the absolute circular polarization in radio pulsars

In most pulsars, the circularly polarized component, Stokes V, is weak in the average pulse profiles. By forming the average profile of |V| from single pulses we can distinguish between pulsars where V is weak in the individual pulses and those where large V of variable handedness is observed from one pulse to the other. We show how |V| profiles depend on the signal-to-noise ratio of V in the single pulses and demonstrate that it is possible to simulate the observed, broad distributions of V by assuming a model where |V| is distributed around a mean value and the handedness of V is permitted to change randomly. The |V| enhanced profiles of 13 pulsars are shown, five observed at 1.41 GHz and eight observed at 4.85 GHz, to complement the set in Karastergiou et al. (2003b). It is argued that the degree of circular polarization in the single pulses is related to the orthogonal polarization mode phenomenon and not to the classification of the pulse components as cone or core.

‖v‖: new insight into the circular polarization of radio pulsars

We present a study of single pulses from nine bright northern pulsars to investigate the behaviour of circular polarization, V. The observations were conducted with the Effelsberg 100-m radio telescope at 1.41 and 4.85 GHz and the Westerbork radio telescope at 352 MHz. For the first time, we present the average profile of the absolute circular polarization |V| in the single pulses. We demonstrate that the average profile of |V| is the distinguishing feature between pulse components that exhibit low V in the single pulses and components that exhibit high V of either handedness, despite both cases resulting in a low mean. We also show that the |V| average profile remains virtually constant with frequency, which is not generally the case for V, leading us to the conclusion that |V| is a key quantity in the pulsar emission problem.

A New Model for the Beams of Radio Pulsars

With the discovery of new classes of radio-emitting neutron stars, such as RRATS and the radio emitting magnetar, understanding the structure and geometry of pulsar beams is becoming central to the interpretation of the observations. We have developed a model, whereby radio emission at a particular frequency arises from a wide range of altitudes above the surface of the star, which can account for the large diversity found in the average profile s hapes of pulsars. We demonstrate how a change in the range of emitting heights can account for the differences between pulse profiles of young, highly energetic pulsars and their older counterparts. Monte Carlo simulations are used to demonstrate the match of our model to real observations.