[Lofarpwg] [DKIM: Failed] Re: LOTAAS binary pulsar + 20 timing paper

[Daniele]

Hi Chia Min,

I am sorry for the huge delay in my comments. I enjoyed reading the paper,
I think it is very clear and thoughtful.
My main comments (more questions actually) are the following.

- I am sure this came out before but I am not aware (it is difficult for me
to attend the PWG telecons because of the time zone, I am sorry about
that). Have you considered presenting the binary pulsar into a separate
paper? With respect to the rest of the sample, the observations are
different, the analysis overlaps only very partially, conclusions and
discussions are mostly separated. Therefore, I think it could make much
sense to have a separate paper specifically about the binary, where you
might also include information not presented here (e.g. "the study of the
scintillation properties of the pulsar will be presented in a separate
paper").

- Regarding J0421, you attribute the strong timing noise to mode switching.
However, even after excluding the two points with the largest offset in
Fig.1, the noise is still very high. To me, it looks that both the
different profiles and the timing noise could be due to a wrong timing
solution. Have you considered this possibility? I mention it because a
couple of pulsars in my sample showed a similar behavior that disappeared
later on when I added more observations. What is the chance of the timing
solution to be degenerate giving the small number of TOAs?

- The characteristics of the profile of PSR J1658 (Fig. 10) and its
relatively strong DM variation (Fig. 9) suggest this pulsar as a good
candidate for a detection of profile variation with time. I suggest you to
carefully compare pulse profiles at different epochs to check if the weak
postcursor or any other feature varies with intensity with respect to the
main peak. In order to do so quickly and accurately, you could normalize
the main peak of the integrated profile of each LOFAR observation and
subtracting one of the profiles from the rest. I can do that myself if you
prefer.
In any case, have you checked with Joris about the DM variation? Since the
pulsar DM is so low, the fractional variation could be pretty large,
suggesting a very turbulent medium. By looking at other studies of ISM
turbulence in the local bubble (such as
http://adsabs.harvard.edu/abs/2010AAS...21541509Sm), this looks a very
promising candidate, especially considering that it is a binary. I wonder
whether we should try some analysis ourselves before publishing the DM
variation.


The rest of my comments are very minor.

- Sec. 2, "An archive file is produced from each observation". This could
be a good place to first cite PSRCHIVE
- Sec. 2, "To model the binary orbit [...] These data were recorded and
processed using the same method as the monthly timing observations." It is
difficult for me to follow the numerous observations performed of this
pulsar, I suggest including a second summary table specifically for this
source.
- Sec. 3, "However, PSRs J0210+5845 and J0421+3255 showed large timing
residuals". I do not see J0210+5845 showing large residuals.
- Fig. 2, What is the purpose of indicating the binary pulsar with a
different style?
- Fig. 4, How are the profiles normalized with respect to one another?
- Fig. 4, "The phase of the pulses in the observations where the offset
occurred coincides with the leading peak of the integrated pulse profile."
I do not understand this sentence.
- Sec. 5, "We estimated the average RFI fraction of each observation at
1532 MHz to be 20 per cent of the bandwidth and 5 per cent of the observing
length, while at 334 MHz, the fraction is 20 per cent of the bandwidth and
20 per cent of the observing length." Are these the final values you are
going to use? I need to check that we are using consistent numbers.
- Sec. 5, "The measured spectral indices of the pulsars are also shown in
Table 6." You already mentioned Table 6 at the beginning of the paragraph.
- Sec. 5, I think that showing the plots of flux as a function of frequency
could be worth for each pulsar.
- Fig. 5, I see that you are considering the error on the slope of the line
but not on the intercept. I did the same at the beginning but then I
decided to also consider the intercept error. What do you think about this?
- Fig. 6, I don't see some of the figures to be referenced in the text,
such as 3, 6 and 7 (I am sorry if I am missing them). Specifically about
Fig. 6, I am unsure it is necessary to show every observation, one or two
examples may suffice.
- Tab. 7, I would specify that these are LOFAR observations.
- Sec 6.2, this could easily be a separate section.
- Sec 6.2, "Later, we found that the position [...] observation taken in
2014." The way this is phrased may seem to suggest that the initial
position was wrong. I suggest removing "later" and explicitly stating that
we measured a proper motion from the different positions of the source
measured at an interval of 14 years.
- Sec 6.2, "The diffractive scintillation and possibly refractive
scintillation will result in variation in the flux densities." What is the
expected band of the scintles for this pulsar?

On Tue, 20 Aug 2019 at 08:22, Michael Kramer <mkramer at mpifr-bonn.mpg.de>
wrote:

> Dear Chia Min,
>
> I read the paper and have some comments which are hopefully useful.
> Please, let me know if
> my handwriting is difficult.
>
> Please, note my comments about the reference to spectra paper - which is
> also relevant for
> the census papers (I’ll read those shortly). The problem with the cited
> Bates et al. and
> Jankowski et al references is, that the spectra in both papers were
> derived for a relatively
> small frequency range, which is well above the the LOFAR frequencies. When
> you make
> comparisons it is probably more appropriate to refer to samples which are
> drawn from
> data covering the whole frequency range. Here the best reference is
> probably still Maron et al.
>
> Anyway, I hope the comments are useful.
>
> Cheers, Michael
>
>
>
> > On 6 Aug 2019, at 15:17, Cees Bassa <cbassa at gmail.com> wrote:
> >
> > Hi Chia Min,
> >
> > Thanks for sending the draft around. There's some nice work in the paper.
> >
> > Below are comments up to section 4. I'll try to get my comments on the
> > rest of the paper by the end of the week.
> >
> > Regards,
> >    Cees
> >
> > Title:
> >
> > * The title is perhaps a bit long. Since the title of Daniele's paper
> >  was: "LOTAAS: Characterization of 20 pulsars", you could use
> >  "LOTAAS: Timing of 21 pulsars", though I agree that having the
> >  binary MSP in the title is a plus.
> >
> > SECTION 1:
> >
> > * I think the introduction needs a bit more work. Right now, the
> >  information motivating this paper (pulsar properties from timing,
> >  spectral indices from multi-frequency follow-up, pulse profile
> >  evolution) is available, but it is not in a logical order. My
> >  suggestion would be that you start with a general overview of pulsar
> >  science, which has sofar been predominantly done at higher
> >  frequencies and the importance of observing at lower
> >  frequencies. You can then provide some information of low-frequency
> >  studies of known pulsars (Pilia profiles, LOFAR Census, MSP Census,
> >  Sobey RM study). This naturally leads into LOTAAS as the first real
> >  low frequency pulsar search, and that this paper presents some of
> >  the first timing and multi-frequency results of pulsars discovered
> >  at these low frequencies.
> >
> > SECTION 2:
> >
> > * In the first paragraph it would be good to compare the setup of the
> >  timing observations to that of the search observations in terms of
> >  number of stations, bandwidth and integration time.
> >
> > * Mention psrchive (with Hotan ref) after dspsr.
> >
> > * Did you not use any ROACH data from Jodrell? Certainly for the
> >  334MHz observations the coherent dedispersion will be important.
> >
> > * I would also restructure section 2 somewhat, as right now the
> >  information is spread about. Perhaps a subsection for timing
> >  observations, timing of the binary, and follow up observations for
> >  profile evolution and spectral properties. Another approach would be
> >  to split it in LOFAR core, Jodrell and GLOW observations.
> >
> > * The paragraphs about J1658 are also confusing, and I think all this
> >  information can be distilled into a table. You also need to clarify
> >  the timing procedure a bit, in that we start off with binary
> >  parameters determined from variations in the spin period (an
> >  incoherent timing solution), to a coherent timing solution where you
> >  account for all rotations of the pulsar. You might also clarify the
> >  'accounting for all rotations' at the start of section 2 when
> >  discussing the timing of normal pulsars.
> >
> > * Table 1: miutes -> minutes
> >
> > * Table 1: Somewhere in the paper we need to provide the JXXXX+YY
> >  names from the LOTAAS overview paper to these timing names. This
> >  naturally fits in table 1.
> >
> > SECTION 3:
> >
> > * For determining the template, you need to specify to which profile
> >  you fitted the von Mises profiles. Was it the single observation
> >  with the highest S/N or some average of profiles?
> >
> > * Add how many von Mises functions you had to use (just give the
> >  minimum and maximum).
> >
> > * "are split into 2 frequency channels" -> "are averaged into two
> >  frequency channels".
> >
> > * I'm a bit worried by only using two frequency channels to measure
> >  DM, as you are using two measurements to obtain two parameters (TOA
> >  and DM).
> >
> > * How were the templates at different frequencies referenced to
> >  absolute phase? If scattering or intrinsic profile evolution is
> >  present then you could impart a DM offset by this approach.
> >
> > * You need to clarify what is meant by a jump. That is not
> >  nomenclature that non-pulsar astronomers know. Also change "Any
> >  offsets" to "Any time offsets", and perhaps specify what causes
> >  these (cable length differences etc).
> >
> > * In paragraph 2 "In order to...", you may want to make it clear that
> >  the dense campaigns were meant to remove ambiguities in the number
> >  of pulse rotations between consecutive observations, and that
> >  getting phase connection on short timescales allowed extrapolating
> >  the timing solution to longer timescales without introducting
> >  ambiguities.
> >
> > * I mentioned this for section 2 as well, but for the timing solution
> >  of J1658; explain that you got initial orbital parameters by
> >  modelling the spin period variations, and used those to obtain a
> >  phase coherent timing solution.
> >
> > * I am not sure a reference exists for FITORBIT, in which case you may
> >  have to explain in a bit more detail what the software does.
> >
> > * "... is a low S/N pulsar" is an incorrect statement. low S/N is a
> >  property of the telescope, not really the pulsar. Maybe rewrite to
> >  "In the initial timing observations of PSR J1643+1338 it was
> >  detected at low S/N."
> >
> > * Section 3 is mixing observations and their analysis with
> >  results. This may be a large change, but I think the first 4
> >  paragraphs could be joined with section 2 to make a "Observations
> >  and timing analysis" section. This resolves some of the duplicity of
> >  the statements between section 2 and 3. The rest of section 3 could
> >  be called "Timing properties".
> >
> >  Sections 4 and 5 also mix analysis and results. I think it would be
> >  more logical to include the first paragraph of section 4 in section
> >  2, and similarly the first few paragraphs of section 5 also fit
> >  better in section 2.
> > _______________________________________________
> > lofarpwg mailing list
> > lofarpwg at astron.nl
> > http://mailman.astron.nl/listinfo/lofarpwg
>
>
> ----------------------------------------------------------------------------------------------------------
> Prof. Dr. Michael Kramer
> Director - Max-Planck-Institut für Radioastronomie, Bonn, Germany
> Professor for Astrophysics - University of Manchester, UK
> Professor (Hon-Prof.) - University of Bonn, Germany
>
> Address:   MPI fuer Radioastronomie
>                  Auf dem Huegel 69
>                  53121 Bonn, Germany
>
> Phone: +49-228-525-278 (direct)
>            +49-228-525-299 (secretary)
>
> EMAIL:  mkramer at mpifr-bonn.mpg.de / michael.kramer at manchester.ac.uk
>
> ----------------------------------------------------------------------------------------------------------
>
>
> _______________________________________________
> lofarpwg mailing list
> lofarpwg at astron.nl
> http://mailman.astron.nl/listinfo/lofarpwg
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.astron.nl/pipermail/lofarpwg/attachments/20190826/8ff19ba6/attachment-0001.html>