The 3rd Report, the 2nd Referee on the SN 2002ap Jet Paper

 I have reviewed the manuscript "Optical Spectropolarimetry of SN
 2002ap: A High Velocity Asymmetric Explosion" by Kawabata et al.  The authors
 present a first analysis of spectropolarimetric data obtained for the possible
 "hypernova" SN 2002ap, and offer a novel explanation for the origin of the
 early-time polarization: That at least some of it is produced by
 electron-scattering off of a blob of material (perhaps from a jet)
 ejected at high velocity.  SN 2002ap is a very exciting object,
 and the paper presents interesting data and speculations.
 The writing is, for the most part, clear and concise.

 Clearly the most controversial element of the paper is the
 introduction of a "jet" of material to explain some of the
 early-time polarization properties.  The jet model is not
 fully developed in the work, but the authors stress that a more
 complete description will be given elsewhere, and that this
 hypothesis is very speculative in nature.  Enough information
 is provided to demonstrate from rough order-of-magnitude calculations
 that a jet model might be tenable.  I find this level of treatment
 to be appropriate.  A jet model is, at the very
 least, an interesting speculation, and one that other workers might be
 interested in pursuing the implications of.  So, although I am eager
 to see the fully worked-out jet model, I agree with the authors that this
 Letter is not the place for it: Letters are commonly less
 rigorous and more speculative in nature than works in the main Journal.
 I can therefore recommend its publication as a Letter, subject
 to a few modifications suggested below.

They begin with the most important.
o When discussing the jet hypothesis, I am surprised and a bit puzzled that
mention is not made of the recent work by Craig Wheeler et al.
(e.g., Wheeler, Meier, \& Wilson 2002), who propose that "bipolar"
jets may be responsible for the ejection of envelopes in core-collapse SNe.
When reading Kawabata's jet hypothesis, in fact, my initial
assumption was that there were *2* jets, and I could not understand
why reflection off of *both* concentrations of material was not
discussed.  It was only after reading for a second time that
I inferred that there was only 1 clump of material.  The authors
need to make this clear from the start: they are discussing the possibility
of only *1* blob of material.  Clearly a second clump, if ejected in
the opposite direction, would have consequences on the observed
polarization (except I suppose for i=90).  While I think it is OK to limit
the simple model to a single blob of material, I feel that this assumption
should be made very clear to the reader, and that the Wheeler et al. work
should at least be referred to as further support for jets in core-collapse
SNe.

o I do not understand exactly how the "residual polarization" that is shown in
Fig. 3e and discussed at the bottom of p. 7, is derived.  It seems that the
authors wish to remove the effect of the proposed jet polarization from the
Feb. data, and then show that what remains in the Feb. data is consistent
with the P.A. of the March data, but it is not clear to me how is this done.
What is the assumed value of the "jet polarization"?  I had thought that
this value would simply be the continuum polarization at some particular
wavelength, but since the "residual" is at least somewhat greater than 0
(even in the 5000-7000 A range), I am lead to believe that this is not so.
Was it taken to be some arbitrarily-defined fraction of the observed
polarization?  Is there a wavelength-dependence to it?  The authors
need to make clear exactly what is going on here.

o SN 2002ap is currently a very "hot" object, with new references being added
to the literature on an almost bi-weekly basis.  Although I don't recommend
including all (or even most) of the latest references, I do feel that two
should be referenced that are not, especially since they are now (according
to ASTRO-ph and the journal's web sites) accepted and would be useful to
readers to look at:

o Kinugasa et al. (ApJL; astro-ph/0208470) -- Presents detailed modeling of
the spectral evolution of SN 2002ap.  This is very relevant to the
discussion of ejecta velocity.  Perhaps this work could be referenced on
p. 8, first complete paragraph, where the "maximum ejecta velocity indicated
by spectrum synthesis of SN 2002ap" is given to be 0.22c; as it is, I'm
not sure where this number comes from. For photospheric velocity,
Kinugasa find v=35,000 km/s early on.

o Leonard et al. (PASP; astro-ph/0206369) -- Presents similar
spectropolarimetric data to Kawabata et al., although with
somewhat different interpretation.  Referencing this work is
complicated by the fact that the most relevant part of it to this paper is a
>*reaction* to the jet model described here, and put on astro-ph earlier.
Although not discussed in any detail, Leonard et al. demonstrate a
similarly good fit (perhaps better, even) between the
early-time polarized flux and redshifted flux, a comparison
that may enhance belief in the jet model presented by the present paper.
I leave it to the author's discretion
(``Isn't that decent of him to leave something to our discretion'')
whether to refer to this paper;
if they choose to do so, they should make it clear that the
jet model originated in the present work, and that Leonard et al.
is testing the comparison for a similar data set.

For the record, I agree with the authors that the Berger etal. ApJL
(astro-ph/0206183) need not be referenced.  That article unfortunately
makes only one glib, broad statement about the production of radio emission
from a jet, and does not in my mind effectively refute the jet hypothesis.

o Minor point: When discussing the "reverse P-Cygni" polarization profile, the
authors may want to include a reference to Leonard & Filippenko (2001, PASP,
113, 920), since that paper presented and took the Jeffery (1991a,b) model
to a somewhat more analytical level.

o Picky point: Bottom of p. 6, 4th line from the bottom.  "cf" should be
replaced by "see", since "cf" means "contrasts with" (as I've been told by
multiple ApJ editors over the years).

o p. 7, 5th line.  The authors are quite clear that their proposed jet model
is only speculative at the moment.  However, such warnings can evidently not
be overemphasized.  I think the sentence "This agreement
suggests...explosion."  should be reworded as: "This agreement
suggests, but does not prove, that a large component of polarized flux may
come from ..."

******************************************************************
***
>
>We estimate that the length of your paper 'Optical
>Spectropolarimetry of SN 2002ap: A High Velocity Asymmetric
>Explosion' is 4.35 journal pages.  Papers longer than 4 journal
>pages cannot be accepted.  You should wait until after you have
>received the referee's report before preparing a shorter version.
>
>The estimate was obtained as follows:
>
>Number of spaces per line:     A = 91
>
>Number of lines per page:      B = 36
>
>Number of pages of main text:  C = 6.33
>
>(Here, the main text does not include the heading section,  i.e.,
>the title, authors, addresses, subject headings, and footnotes,
>and does not include the references, figures with captions, and
>tables.  The main text does include all  blank spaces between
>sections, and all equations.)
>
>(A x B x C) / 7686 = 2.7 the number of journal pages needed for
>the main text.  (7686 is the product of 61 spaces per line x 63
>lines per page x 2 columns.)
>
>To this is added the estimated number of journal pages for:
>
>The heading section             0.55
>
>References                      0.29
>
>Figures with captions           0.68    (see details below)
>
>Tables                          0.13    (see details below)
>
>Extra space for equations       0
>
>(the number of equations x 0.03, when the double-spaced typescript
>does not have an extra double space before and after each
>equation; the number of equations is multiplied by 0.06 when
>lengthy equations force the Letter to be set in a single column
>format)
>
>                         TOTAL   4.35
>
>Please note that the width of a figure normally should not be
>greater than its height if the figure is to appear in an 8.8 cm
>column of the Journal, and it must be legible when reduced to this
>size.
>
>Detailed Figure and Table estimates:
>
>                 #1      #2      #3      #4      #5
>
>Figure:         0.2    0.22    0.26    0    0
>
>Table:          0.13    0    0    0    0
>
>****************************************************************