Loading ...
Sorry, an error occurred while loading the content.
 

Comet PanSTARRS, Magnitudes, Extinction, Visibility

Expand Messages
  • cnj999
    Good morning Toni and all. I can fully appreciate the perplexed feelings among many observers in regard to the scatter one sees in the current photometric
    Message 1 of 7 , Mar 19, 2013
      Good morning Toni and all. I can fully appreciate the perplexed feelings
      among many observers in regard to the scatter one sees in the current
      photometric data, especially among those who have employed distant (in elevation)
      comparison stars for PanSTARRS m1 estimates and then attempted to correct
      for excessive atmospheric extinction. I believe it was Alan Hale who also
      brought up this subject a week, or so ago, and for which I had prepared a
      long post in response still waiting in my "out going mail" letter box. Let me
      say that I've likewise just finished writing an even longer post for a
      site with less experienced observers on another forum addressing the overall
      situation and to save time will be cutting an pasting certain portions of
      that post into this one. So please excuse any oddities, or changes in type
      face, in what is written below.

      I think that is can prove a serious mistake for observers to take any
      extinction tables quite literally. If an adjustment over a relatively small
      difference (maybe 10-degrees) in the elevation of differing objects is
      required and the lower object is not extremely low in the sky (under say 5 or 8
      degrees) the correction figures should not be too far afield of reality.
      However, if elevation differences at really large, as I suspect has been the
      situation for many northern observers of late, then the correction factors can
      become almost meaningless. As I point out in my post to Alan, the degree
      of atmospheric extinction can vary dramatically night-to-night at any given
      location outside of maybe mountain tops and any tables will be no more than
      an approximation. At the same time, I'm sure that below an altitude of
      10-degrees, and perhaps even decidedly higher than that, the tables refer to
      extrapolated theoretical values rather than any actual measurements. I know
      from my own experiments that in very clear, near sea level skies, one can
      easily follow 1st and 2nd magnitude virtually to the horizon without optical
      aid. This is further bolstered by views of the sky from at sea where under
      good conditions, where although of course somewhat attenuated, the
      brighter stars are seen right to the horizon quite often.


      In addressing the overall situation with PanSTARRS recently reported
      magnitudes I must admit that the degree of scatter that I am seeing since it
      became visible to Northern Hemisphere observers and currently amounting to
      something like 3.5 magnitudes(!), has stunned me. I've not seen such
      uncertainty in a comet's brightness in several decades and that it should occur in
      conjunction with such a faint twilight comet is rather perplexing.
      I would note that drastically less scatter is evident in the magnitude
      data posted among the more reliable comet observers from the Southern
      Hemisphere prior to the comet's perihelion. When plotted, these data form a very
      tight lightcurve (see below) which extends from the first days of January
      until just a couple of days prior to perihelion. The photometric parameters
      derived from these data has a coefficient of correlation amounting to 0.98 ,
      i.e. a very high formula probable accuracy. Further, my observation placing
      the comet at +2.6 on March 19.0UT corresponds almost precisely to that
      predicted by this formula for that date with the comet outbound from the Sun.
      _http://i64.photobucket.com/albums/h182/CNJ831/PanSTARRS031813copy_zps5d3b10
      07.jpg_
      (http://i64.photobucket.com/albums/h182/CNJ831/PanSTARRS031813copy_zps5d3b1007.jpg)
      At the same time, judging by similarly exposed images of the comet from
      down south, the physical appearance and brightness of PanSTARRS when last
      seen from Hemisphere seems absolute no different from the recent examples
      obtained up north. This infers that no physical changes in either brightness,
      or the comet's appearance and visibility, have occurred between the final
      southern and early northern sightings. The change in solar elongation are not
      drastic either in the overlapping interval.
      Now there have been a great many reports from northern observers that
      PanSTARRS has been near, or even below, the level of detection with the naked
      eye, even under favor sky conditions. I've observed a number of magnitude
      zero to +1.0 comets in twilight over the years and let me give you an
      impression of how they have looked to me. Comet Bennett, in 1970, is perhaps a
      nice example to cite.
      Shortly following its perihelion, when it was just becoming visible from
      the Northern Hemisphere in morning twilight as an object of +0.5 magnitude,
      my observing log includes a passage that outlines my initial impression.
      While awaiting the best time to prepare and observe the comet one morning, I
      sat down to watch a little TV beforehand. The room was modestly illuminated
      by a lamp. In spite of this ample illumination I commenced to notice
      something like a ghostly shaft of light out the east-facing window. After a time,
      very low over the horizon in the breaking twilight I could see a fully
      developed comet rising, with bright head and perhaps 5-degrees of tail...as
      seen from inside the house in the lighted room through a closed window! Thus,
      the impact to the unaided eye of a truly magnitude +0.5 comet is rather
      jaw-dropping and certainly not perceived simply as little more than a vague
      wisp in the twilight.
      To me the above clinches the situation in pointing to PanSTARSS as
      currently being no brighter than 2nd magnitude and already fading.

      J.Bortle



      In a message dated 3/18/2013 1:42:50 P.M. Eastern Daylight Time,
      toniscarmato@... writes:




      Dear all,

      sorry but after to many time passed to observe comets I don’t understand
      some things!
      I follow with interest the discusion about the magnitude of C/2011 L4.

      I published on my PanStarrs homepage

      _http://digilander.libero.it/infosis/homepage/astronomia/c2011l4.htm_
      (http://digilander.libero.it/infosis/homepage/astronomia/c2011l4.htm)

      a movie that show the comet visible at about 0.5 deg above the horizon.
      The ICQ Table of Atmosferic Extinction show that at Z=89 deg for my
      location at 186 m above the sea level the extincion is 7.38 mag!
      As I described in my comments the comet was well visible in 7x50
      binoculars at that altitude and I estimed –1.0 the mag of the comet
      using a star at 62 deg of altitude and also using my record of the comet
      C/2006 P1 in the same conditions!

      My question is: how is possible to see clearly a comet at about 0.5 deg of
      altitude with a strong extincition if the comet was at positive magnitude
      1 or >?
      Thanks for the attention.
      Regards,
      Toni Scarmato

      [Non-text portions of this message have been removed]






      [Non-text portions of this message have been removed]
    • Richard Miles
      Continuing John Bortle s theme, I have uploaded some results of my recent V photometry of PanSTARRS at:
      Message 2 of 7 , Mar 19, 2013
        Continuing John Bortle's theme, I have uploaded some results of my recent V
        photometry of PanSTARRS at:
        http://www.britastro.org/~rmiles/Images/C2011L4_20130314-17_RMiles_Vfilter.png

        Measurements were made for two epochs:
        2013 March 14.80 Total magnitude = +0.8, tail length = 30'
        2013 March 17.80 Total magnitude = +1.9, tail length = 15'

        In both cases, excellent comparison stars were available located at a
        similar airmass to the comet. Corrections have been applied for differential
        airmass and for the CCD transformation coefficient.

        If we look at the coma within 30" of the nucleus, the comet has faded by
        about 0.7 mag in exactly 3.00 days (cf. the MPC ephemeris predicts a 0.6
        magnitude fade).

        First I should say that photometry based on CCD images taken through a
        filter is capable of excellent precision and good accuracy, the usual
        limiting factor for comets being the sky background subtraction where the
        tail of the comet is very extended. One reason why many comet observers are
        unsuccessful with comet photometry, or fail to get started, is the fact that
        there does not exist a convenient piece of commercial software available as
        exists for example in the case of Brian D. Warner's 'Canopus' package for
        Asteroid Photometry.

        Also, a second reason is that there is no standard methodology for reducing
        CCD images. This situation partly arises because a somewhat different
        approach is needed when reducing observations of a more distant comet having
        a tail length of say 10-100 arcsec compared with a bright comet near
        perihelion where the tail is 10-100 arcminutes, or even many degrees in
        length with a complex structure.

        This doesn't mean we can't have a standard approach. For instance here's a
        suggestion for observers trying to reduce filtered images of bright comets
        such as PanSTARRS: When you integrate the counts in a circular aperture of
        increasing size, don't place this aperture on the 'photocentre' (= centre of
        gravity of the signal) like one does for a variable star. Instead, move the
        aperture around the image until you find the maximum integrated signal. As
        you go to larger apertures, this position gradually moves away from the
        position of the nucleus in the direction of the tail. This is in effect what
        visual observers do when estimating brightness. By this approach, the larger
        measuring aperture sizes correspond to integrated magnitudes for what are in
        effect different tail lengths. So when you compare CCD and visual
        magnitudes, you should read off the CCD magnitude from the 'growth curve' as
        shown in the above link at the same tail length reported by the visual
        observer.

        Also, to be able to accurately link CCD and visual observers, CCD
        photometricists need to use either a Johnson V filter or the Green FITS
        image from an RGB colour image taken with a DSLR camera or the like. Watch
        out with the latter - here you will be best to pick comparison stars having
        a similar B-V colour magnitude as the target if you don't want to do the
        full business of transforming the stellar magnitude by correcting for colour
        differences. Airmass corrections need to be applied in a rigorous way of
        course but this is not much of an issue at altitudes above say 5 degrees.

        Now who's going to volunteer to write that all-singing all-dancing bright
        comet photometry package? Richard Berry - Are you out there somewhere?

        Richard Miles
        BAA
      • Toni Scarmato
        Hi Jon and all. Many thanks for the very long post I read it with much attention!
        Message 3 of 7 , Mar 19, 2013
          Hi Jon and all.

          Many thanks for the very long post
          I read it with much attention!



          <Good morning Toni and all. I can fully appreciate the perplexed feelings
          <among many observers in regard to the scatter one sees in the current
          <photometric data, especially among those who have employed distant (in elevation)
          <comparison stars for PanSTARRS m1 estimates and then attempted to correct
          <for excessive atmospheric extinction.
          <I think that is can prove a serious mistake for observers to take any
          <extinction tables quite literally. If an adjustment over a relatively small
          <difference (maybe 10-degrees) in the elevation of differing objects is
          <required and the lower object is not extremely low in the sky (under say 5 or 8
          <degrees) the correction figures should not be too far afield of reality.
          <However, if elevation differences at really large, as I suspect has been the
          <situation for many northern observers of late, then the correction factors can
          <become almost meaningless.
          Ok so the ICQ table is not usefull if the the comet is low than 5 deg. This evening I observe the comet in very good conditions of the sky.
          The comet was easy visible to naked eye and also the tail! The comet was at about 5 deg above the horizon and I use 31 And 3.5 mag to comparison at 14 deg above the Horizon.
          I estimed the comet 1.6 to NE with correction for extinction!
          <As I point out in my post to Alan, the degree

          <of atmospheric extinction can vary dramatically night-to-night at any given
          <location outside of maybe mountain tops and any tables will be no more than
          <an approximation. At the same time, I'm sure that below an altitude of
          <10-degrees, and perhaps even decidedly higher than that, the tables refer to
          <extrapolated theoretical values rather than any actual measurements. I know
          <from my own experiments that in very clear, near sea level skies, one can
          <easily follow 1st and 2nd magnitude virtually to the horizon without optical
          <aid. This is further bolstered by views of the sky from at sea where under
          <good conditions, where although of course somewhat attenuated, the
          <brighter stars are seen right to the horizon quite often.
          <In addressing the overall situation with PanSTARRS recently reported
          <magnitudes I must admit that the degree of scatter that I am seeing since it
          <became visible to Northern Hemisphere observers and currently amounting to
          <something like 3.5 magnitudes(!), has stunned me. I've not seen such
          <uncertainty in a comet's brightness in several decades and that it should occur in
          <conjunction with such a faint twilight comet is rather perplexing.
          <I would note that drastically less scatter is evident in the magnitude
          <data posted among the more reliable comet observers from the Southern
          <Hemisphere prior to the comet's perihelion. When plotted, these data form a very
          <tight lightcurve (see below) which extends from the first days of January
          <until just a couple of days prior to perihelion. The photometric parameters
          <derived from these data has a coefficient of correlation amounting to 0.98 ,
          <i.e. a very high formula probable accuracy. Further, my observation placing
          <the comet at +2.6 on March 19.0UT corresponds almost precisely to that
          <predicted by this formula for that date with the comet outbound from the Sun.
          _http://i64.photobucket.com/albums/h182/CNJ831/PanSTARRS031813copy_zps5d3b10
          07.jpg_
          (http://i64.photobucket.com/albums/h182/CNJ831/PanSTARRS031813copy_zps5d3b1007.jpg)
          <At the same time, judging by similarly exposed images of the comet from
          <down south, the physical appearance and brightness of PanSTARRS when last
          <seen from Hemisphere seems absolute no different from the recent examples
          <obtained up north. This infers that no physical changes in either brightness,
          <or the comet's appearance and visibility, have occurred between the final
          <southern and early northern sightings. The change in solar elongation are not
          <drastic either in the overlapping interval.
          <Now there have been a great many reports from northern observers that
          <PanSTARRS has been near, or even below, the level of detection with the naked
          <eye, even under favor sky conditions.
          As i sayd this evening the comet was easy with NE from 19:00 to 19:25 L.T. when I loss it in the thiny clouds!
          <I've observed a number of magnitude
          <zero to +1.0 comets in twilight over the years and let me give you an
          <impression of how they have looked to me. Comet Bennett, in 1970, is perhaps a
          <nice example to cite.
          <Shortly following its perihelion, when it was just becoming visible from
          <the Northern Hemisphere in morning twilight as an object of +0.5 magnitude,
          <my observing log includes a passage that outlines my initial impression.
          <While awaiting the best time to prepare and observe the comet one morning, I
          <sat down to watch a little TV beforehand. The room was modestly illuminated
          <by a lamp. In spite of this ample illumination I commenced to notice
          <something like a ghostly shaft of light out the east-facing window. After a time,
          <very low over the horizon in the breaking twilight I could see a fully
          <developed comet rising, with bright head and perhaps 5-degrees of tail...as
          <seen from inside the house in the lighted room through a closed window! Thus,
          <the impact to the unaided eye of a truly magnitude +0.5 comet is rather
          <jaw-dropping and certainly not perceived simply as little more than a vague
          <wisp in the twilight.

          This evening i must to say that the comet was impressive to NE I noticed the strong condensed false nucleus clearly and at least 1 deg of the tail! In my images taken this evening are visible some stars and I will be able to make a measure of magnitude. I will let you know the results!
          <To me the above clinches the situation in pointing to PanSTARSS as
          <currently being no brighter than 2nd magnitude and already fading.

          I agree!
          Toni



          [Non-text portions of this message have been removed]
        • denis buczynski
          Hi Richard, Could I ask the same question to you as I asked to Clay a few weeks ago, this being: Would it be true to say that recording comets with as small a
          Message 4 of 7 , Mar 19, 2013
            Hi Richard,
            Could I ask the same question to you as I asked to Clay a few weeks ago, this being:
            Would it be true to say that recording comets with as small a focal length as
            possible with a ccd camera will provide a magnitude determination that captures
            the whole of the comet including its outer coma in few pixels. Or are comet
            magnitudes more reliable when made with large telescopes as long focal lengths
            that spread the coma over many tens of pixels. The same stars within the coma
            will affecting both sets of measurements. In visual estimates it is recommended
            that the use of small an aperture as possible that least magnifies the image of
            the comet and “compresses” the total light into as small an area as possible, is
            the best method. Does the same rational apply to ccd images is my question. For
            instance I have recorded C/2012 S1 ISON at magnitude 16 with a 300mm lens and an
            ST7 XME ccd, the image of the comet is essentially starlike, and could be
            measured in the same manner and with the same rigour as any star on the same
            frame.

            Denis Buczynski

            Continuing John Bortle's theme, I have uploaded some results of my recent V
            photometry of PanSTARRS at:
            http://www.britastro.org/~rmiles/Images/C2011L4_20130314-17_RMiles_Vfilter.png

            Measurements were made for two epochs:
            2013 March 14.80 Total magnitude = +0.8, tail length = 30'
            2013 March 17.80 Total magnitude = +1.9, tail length = 15'

            In both cases, excellent comparison stars were available located at a
            similar airmass to the comet. Corrections have been applied for differential
            airmass and for the CCD transformation coefficient.

            If we look at the coma within 30" of the nucleus, the comet has faded by
            about 0.7 mag in exactly 3.00 days (cf. the MPC ephemeris predicts a 0.6
            magnitude fade).

            First I should say that photometry based on CCD images taken through a
            filter is capable of excellent precision and good accuracy, the usual
            limiting factor for comets being the sky background subtraction where the
            tail of the comet is very extended. One reason why many comet observers are
            unsuccessful with comet photometry, or fail to get started, is the fact that
            there does not exist a convenient piece of commercial software available as
            exists for example in the case of Brian D. Warner's 'Canopus' package for
            Asteroid Photometry.

            Also, a second reason is that there is no standard methodology for reducing
            CCD images. This situation partly arises because a somewhat different
            approach is needed when reducing observations of a more distant comet having
            a tail length of say 10-100 arcsec compared with a bright comet near
            perihelion where the tail is 10-100 arcminutes, or even many degrees in
            length with a complex structure.

            This doesn't mean we can't have a standard approach. For instance here's a
            suggestion for observers trying to reduce filtered images of bright comets
            such as PanSTARRS: When you integrate the counts in a circular aperture of
            increasing size, don't place this aperture on the 'photocentre' (= centre of
            gravity of the signal) like one does for a variable star. Instead, move the
            aperture around the image until you find the maximum integrated signal. As
            you go to larger apertures, this position gradually moves away from the
            position of the nucleus in the direction of the tail. This is in effect what
            visual observers do when estimating brightness. By this approach, the larger
            measuring aperture sizes correspond to integrated magnitudes for what are in
            effect different tail lengths. So when you compare CCD and visual
            magnitudes, you should read off the CCD magnitude from the 'growth curve' as
            shown in the above link at the same tail length reported by the visual
            observer.

            Also, to be able to accurately link CCD and visual observers, CCD
            photometricists need to use either a Johnson V filter or the Green FITS
            image from an RGB colour image taken with a DSLR camera or the like. Watch
            out with the latter - here you will be best to pick comparison stars having
            a similar B-V colour magnitude as the target if you don't want to do the
            full business of transforming the stellar magnitude by correcting for colour
            differences. Airmass corrections need to be applied in a rigorous way of
            course but this is not much of an issue at altitudes above say 5 degrees.

            Now who's going to volunteer to write that all-singing all-dancing bright
            comet photometry package? Richard Berry - Are you out there somewhere?

            Richard Miles
            BAA





            [Non-text portions of this message have been removed]
          • cnj999
            Very interesting content to your post, indeed, Richard. On the evening of March 17/18 (18.00 UT) my visual observations (previously posted in the CometObs
            Message 5 of 7 , Mar 19, 2013
              Very interesting content to your post, indeed, Richard.

              On the evening of March 17/18 (18.00 UT) my visual observations (previously
              posted in the CometObs section of this forum) indicated a coma diameter
              for PanSTARRS of 3' with an integrated magnitude of +2.6 .

              I note that just hours earlier your V photometry data plotted on the graph
              accompanying your post indicated the comet's magnitude with a 3.0'
              aperture to be +2.5 , thus fully in agreement with my data. Both of these data
              also are congruent with Southern Hemisphere visual data obtained pre-T at the
              same heliocentric distance when all are corrected to m1 - 5 log (d)
              values. Only when the aperture of the photometer was increased to progressively
              include what visual observers were seeing as the inner portions of the
              comet's very bright tail does the magnitude rise above the visual.

              J.Bortle








              In a message dated 3/19/2013 4:47:55 P.M. Eastern Daylight Time,
              rmiles.btee@... writes:




              Continuing John Bortle's theme, I have uploaded some results of my recent V

              photometry of PanSTARRS at:
              _http://www.britastro.org/~rmiles/Images/C2011L4_20130314-17_RMiles_Vfilter.
              png_
              (http://www.britastro.org/~rmiles/Images/C2011L4_20130314-17_RMiles_Vfilter.png)

              Measurements were made for two epochs:
              2013 March 14.80 Total magnitude = +0.8, tail length = 30'
              2013 March 17.80 Total magnitude = +1.9, tail length = 15'

              In both cases, excellent comparison stars were available located at a
              similar airmass to the comet. Corrections have been applied for
              differential
              airmass and for the CCD transformation coefficient.

              If we look at the coma within 30" of the nucleus, the comet has faded by
              about 0.7 mag in exactly 3.00 days (cf. the MPC ephemeris predicts a 0.6
              magnitude fade).

              First I should say that photometry based on CCD images taken through a
              filter is capable of excellent precision and good accuracy, the usual
              limiting factor for comets being the sky background subtraction where the
              tail of the comet is very extended. One reason why many comet observers
              are
              unsuccessful with comet photometry, or fail to get started, is the fact
              that
              there does not exist a convenient piece of commercial software available
              as
              exists for example in the case of Brian D. Warner's 'Canopus' package for
              Asteroid Photometry.

              Also, a second reason is that there is no standard methodology for
              reducing
              CCD images. This situation partly arises because a somewhat different
              approach is needed when reducing observations of a more distant comet
              having
              a tail length of say 10-100 arcsec compared with a bright comet near
              perihelion where the tail is 10-100 arcminutes, or even many degrees in
              length with a complex structure.

              This doesn't mean we can't have a standard approach. For instance here's a
              suggestion for observers trying to reduce filtered images of bright comets
              such as PanSTARRS: When you integrate the counts in a circular aperture of
              increasing size, don't place this aperture on the 'photocentre' (= centre
              of
              gravity of the signal) like one does for a variable star. Instead, move
              the
              aperture around the image until you find the maximum integrated signal. As
              you go to larger apertures, this position gradually moves away from the
              position of the nucleus in the direction of the tail. This is in effect
              what
              visual observers do when estimating brightness. By this approach, the
              larger
              measuring aperture sizes correspond to integrated magnitudes for what are
              in
              effect different tail lengths. So when you compare CCD and visual
              magnitudes, you should read off the CCD magnitude from the 'growth curve'
              as
              shown in the above link at the same tail length reported by the visual
              observer.

              Also, to be able to accurately link CCD and visual observers, CCD
              photometricists need to use either a Johnson V filter or the Green FITS
              image from an RGB colour image taken with a DSLR camera or the like. Watch
              out with the latter - here you will be best to pick comparison stars
              having
              a similar B-V colour magnitude as the target if you don't want to do the
              full business of transforming the stellar magnitude by correcting for
              colour
              differences. Airmass corrections need to be applied in a rigorous way of
              course but this is not much of an issue at altitudes above say 5 degrees.

              Now who's going to volunteer to write that all-singing all-dancing bright
              comet photometry package? Richard Berry - Are you out there somewhere?

              Richard Miles
              BAA






              [Non-text portions of this message have been removed]
            • Richard Miles
              Hi Denis, Unlike visual photometry, CCDs are relatively forgiving in that you have much more latitude to play with than when using the Mk.I Eyeball. The golden
              Message 6 of 7 , Mar 19, 2013
                Hi Denis,

                Unlike visual photometry, CCDs are relatively forgiving in that you have
                much more latitude to play with than when using the Mk.I Eyeball.

                The golden rule is to restrict the exposure duration so that no part of the
                comet is saturated - in fact the rule is to avoid the brightest pixel (the
                nucleus) exceeding 50% of the dynamic range of the camera - this ensures you
                remain in the linear part of its response curve.

                Try to stack lots of short unsaturated images to produce one master image
                tracked on the nucleus. The more you stack (within reason), the better the
                signal-to-noise, which is especially crucial for differentiating the tail of
                the comet from the background sky.

                On the question of angular coverage, here the rule is to choose a focal
                length/CCD size such that the *entire image of the comet and extended tail*
                is contained within the dimensions of the exposed frame. This is to ensure
                you can determine the reference level of the background sky - this was a
                problem for instance when 17P/Holmes expanded such that most large
                telescopes failed to record sky beyond the expanding halo of dust and gas
                when imaging the nucleus. So, if you have a 10-degree comet tail, you need
                an FOV of say 12-15 degrees or more.

                Now how many pixels should the comet image extend across? This is one of the
                forgiving aspects of a CCD camera. You could have the entire comet span just
                2-3 pixels, i.e. essentially starlike and you would be able to obtain the
                total magnitude to better than 0.1 mag precision. In practice, a comet has a
                larger FWHM than the adjacent stars and so you should still construct a
                growth curve for the comet relative to stars using various sized measuring
                apertures. When a comet looks similar in size to a star, it tends to be an
                illusion - measure the light in different size apertures and you'll be
                surprised at how much light extends out beyond what's obvious visually when
                you look at the monitor. Most times a comet might span say 10-200 pixels on
                the frame. The software is one of the limiting factors here - you might have
                some difficulty as measuring apertures available in most software are
                limited in size, often to a maximum of 100 pixels radius.

                If you have a CCD 4096 pixels wide then you could record a bright comet some
                2000-3000 pixels across - stacking lots of images is important here to bump
                up the SNR. However, you might have a problem measuring the frame. The main
                issue with extended objects is how do you subtract the field stars contained
                within the faint tail? That's another topic, one that I have a solution for
                but which needs encoding as an algorithm in suitable photometry software.

                Another very forgiving aspect of the CCD is that, unlike visual work, you
                can use stars as comparisons even if they are several magnitudes fainter
                than the nucleus of the comet. I estimate for a single frame, the latitude
                amounts to almost 4 magnitudes for the sort of accuracy we are after in
                comet photometry. But if you stack say 100 frames, the dynamic range will
                extend a further 2.5 magnitudes, i.e. allowing stars up to say 6 magnitudes
                fainter to be used as comparisons. Alternatively, you could stack just 10
                frames and then use an ensemble of say 8-12 comp stars to calibrate the
                stacked frame thereby achieving a similar precision as the previous example.

                Photometricists really need to think in terms of 'growth curves' so that it
                is always possible to determine the magnitude of light in a shell around the
                comet having a specified physical dimension, e.g. radius of 10,000 km or
                some other appropriate distance. The CARA group have adopted this approach
                for instance.

                Richard

                ----- Original Message -----
                From: "denis buczynski" <buczynski8166@...>
                To: <comets-ml@yahoogroups.com>
                Sent: Tuesday, March 19, 2013 9:14 PM
                Subject: Re: [comets-ml] Comet PanSTARRS, Magnitudes, Extinction, Visibility


                Could I ask the same question to you as I asked to Clay a few weeks ago,
                this being:
                Would it be true to say that recording comets with as small a focal length
                as
                possible with a ccd camera will provide a magnitude determination that
                captures
                the whole of the comet including its outer coma in few pixels. Or are comet
                magnitudes more reliable when made with large telescopes as long focal
                lengths
                that spread the coma over many tens of pixels. The same stars within the
                coma
                will affecting both sets of measurements. In visual estimates it is
                recommended
                that the use of small an aperture as possible that least magnifies the image
                of
                the comet and “compresses” the total light into as small an area as
                possible, is
                the best method. Does the same rational apply to ccd images is my question.
                For
                instance I have recorded C/2012 S1 ISON at magnitude 16 with a 300mm lens
                and an
                ST7 XME ccd, the image of the comet is essentially starlike, and could be
                measured in the same manner and with the same rigour as any star on the same
                frame.

                Hi Denis,
              • terryjlovejoy
                Hi Richard + John, I performed a similar exercise using the Green channel from DSLR images from March 6.38. I used 6 Ceti as a comparison star (Spectral Type
                Message 7 of 7 , Mar 20, 2013
                  Hi Richard + John,

                  I performed a similar exercise using the Green channel from DSLR images from March 6.38. I used 6 Ceti as a comparison star (Spectral Type F5) and used a 0.6 magnitude adjustment for atmospheric extinction. The comet should have been near its brightest that evening. Please refer to the diagram:

                  http://groups.yahoo.com/group/comets-ml/files/phot.png

                  Please note this was with a short telephoto lens so I didn't have the scale of Richard's images.

                  Terry

                  --- In comets-ml@yahoogroups.com, "Richard Miles" <rmiles.btee@...> wrote:
                  >
                  > Continuing John Bortle's theme, I have uploaded some results of my recent V
                  > photometry of PanSTARRS at:
                  > http://www.britastro.org/~rmiles/Images/C2011L4_20130314-17_RMiles_Vfilter.png
                  >
                  > Measurements were made for two epochs:
                  > 2013 March 14.80 Total magnitude = +0.8, tail length = 30'
                  > 2013 March 17.80 Total magnitude = +1.9, tail length = 15'
                  >
                  > In both cases, excellent comparison stars were available located at a
                  > similar airmass to the comet. Corrections have been applied for differential
                  > airmass and for the CCD transformation coefficient.
                  >
                  > If we look at the coma within 30" of the nucleus, the comet has faded by
                  > about 0.7 mag in exactly 3.00 days (cf. the MPC ephemeris predicts a 0.6
                  > magnitude fade).
                  >
                  > First I should say that photometry based on CCD images taken through a
                  > filter is capable of excellent precision and good accuracy, the usual
                  > limiting factor for comets being the sky background subtraction where the
                  > tail of the comet is very extended. One reason why many comet observers are
                  > unsuccessful with comet photometry, or fail to get started, is the fact that
                  > there does not exist a convenient piece of commercial software available as
                  > exists for example in the case of Brian D. Warner's 'Canopus' package for
                  > Asteroid Photometry.
                  >
                  > Also, a second reason is that there is no standard methodology for reducing
                  > CCD images. This situation partly arises because a somewhat different
                  > approach is needed when reducing observations of a more distant comet having
                  > a tail length of say 10-100 arcsec compared with a bright comet near
                  > perihelion where the tail is 10-100 arcminutes, or even many degrees in
                  > length with a complex structure.
                  >
                  > This doesn't mean we can't have a standard approach. For instance here's a
                  > suggestion for observers trying to reduce filtered images of bright comets
                  > such as PanSTARRS: When you integrate the counts in a circular aperture of
                  > increasing size, don't place this aperture on the 'photocentre' (= centre of
                  > gravity of the signal) like one does for a variable star. Instead, move the
                  > aperture around the image until you find the maximum integrated signal. As
                  > you go to larger apertures, this position gradually moves away from the
                  > position of the nucleus in the direction of the tail. This is in effect what
                  > visual observers do when estimating brightness. By this approach, the larger
                  > measuring aperture sizes correspond to integrated magnitudes for what are in
                  > effect different tail lengths. So when you compare CCD and visual
                  > magnitudes, you should read off the CCD magnitude from the 'growth curve' as
                  > shown in the above link at the same tail length reported by the visual
                  > observer.
                  >
                  > Also, to be able to accurately link CCD and visual observers, CCD
                  > photometricists need to use either a Johnson V filter or the Green FITS
                  > image from an RGB colour image taken with a DSLR camera or the like. Watch
                  > out with the latter - here you will be best to pick comparison stars having
                  > a similar B-V colour magnitude as the target if you don't want to do the
                  > full business of transforming the stellar magnitude by correcting for colour
                  > differences. Airmass corrections need to be applied in a rigorous way of
                  > course but this is not much of an issue at altitudes above say 5 degrees.
                  >
                  > Now who's going to volunteer to write that all-singing all-dancing bright
                  > comet photometry package? Richard Berry - Are you out there somewhere?
                  >
                  > Richard Miles
                  > BAA
                  >
                Your message has been successfully submitted and would be delivered to recipients shortly.