We present a model that explains the observed deviation of the spectra of some pulsars and magnetars from the power-law spectra which are seen in the bulk of the pulsar population. Our model is based on the assumption that the observed variety of pul
sar spectra can be naturally explained by the thermal free-free absorption that takes place in the surroundings of the pulsars. In this context, the variety of the pulsar spectra can be explained according to the shape, density and temperature of the absorbing media and the optical path of the line-of-sight across that. We have put specific emphasis on the case of the radio magnetar SGR J1745-2900 (also known as Sgr A* magnetar), modeling the rapid variations of the pulsar spectrum after the outburst of Apr 2013 as due to the free-free absorption of the radio emission in the electron material ejected during the magnetar outburst. The ejecta expands with time and consequently the absorption rate decreases and the shape of the spectrum changes in such a way that the peak frequency shifts towards the lower radio frequencies. In the hypothesis of an absorbing medium, we also discuss the similarity between the spectral behaviour of the binary pulsar B1259-63 and the spectral peculiarities of isolated pulsars.
We present our results of pulse broadening time estimates and the study of the frequency scaling of this quantity for 60 pulsars based on actual multi-frequency scattering estimates. This research was based on our own measurements, performed on the o
bservational data and the profiles from various pulsar profile databases, as well as the scatter time measurements that were found in the literature. We were able to construct a database of over 60 pulsars with true multi-frequency $alpha$ measurements, which allowed us to revise the previously proposed relations between the scatter time spectral slope and the dispersion measure (DM). We found that the deviations from theoretical predictions of the value of $alpha$ appear for pulsars regardless of their DM, however the DM-averaged value of the scaling index is almost constant except for pulsars with very high DMs. Based on the obtained slopes we were also able to estimate the amount of scattering at the standard frequency of 1 GHz. We found that while the estimated standardized pulse broadening time increases with DM the relation seems to be much flatter than it was previously proposed, which suggests higher values of the scatter time for mid-DM pulsars, and lower values of expected pulse broadening for highly dispersed sources.
We show the results of our analysis of the pulse broadening phenomenon in 25 pulsars at several frequencies using the data gathered with GMRT and Effelsberg radiotelescopes. Twenty two of these pulsars were not studied in that regard before and our w
ork has increased the total number of pulsars with multi-frequency scattering measurements to almost 50, basically doubling the amount available so far. The majority of the pulsars we observed have high to very-high dispersion measures (DM>200) and our results confirm the suggestion of Loehmer et al.(2001, 2004) that the scatter time spectral indices for high-DM pulsars deviate from the value predicted by a single thin screen model with Kolmogorovs distribution of the density fluctuations. In this paper we discuss the possible explanations for such deviations.
A successful attempt was made to analyse about 6000 single pulses of PSR B1133+16 obtained with the 100-meter Effelsberg radio-telescope. The high resolution (60 micro-seconds) data were taken at a frequency of 8.35 GHz with a bandwidth of 1.1 GHz. I
n order to examine the pulse-to-pulse intensity modulations, we performed both the longitude- and the harmonic-resolved fluctuation spectral analysis. We identified the low frequency feature associated with an amplitude modulation at f4 ~ 0.033 P1^(-1), which can be interpreted as the circulation time P4 ~ 30 P1 of the underlying subbeam carousel model. Despite an erratic nature of this pulsar, we also found an evidence of periodic pseudo-nulls with P4 = 28.44 P1. This is exactly the value at which Herfindal & Rankin found periodic pseudo-nulls in their 327 MHz data. We thus believe that this is the actual carousel circulation time in PSR B1133+16, particularly during orderly circulation.
