{
  "id": 72663,
  "title": "Hydrogen Spectroscopy as Clue in Supernovae Classification?",
  "url": "/competitions/PLAsTiCC-2018/discussion/72663",
  "author_name": "",
  "post_date": "2018-11-26T00:59:20.852957300Z",
  "votes": 1,
  "comment_count": 6,
  "views": 0,
  "content": "<p>Since joining this competition I have believed that photometry is a more affordable spectroscopy and that clues to composition should be able to be gleaned from the ratios of the passbands and or the deltas in the activities of the passbands.</p>\n\n<p>Looking at <a href=\"https://astrobites.org/2016/12/02/classifying-supernovae/\">this article</a></p>\n\n<p>And <a href=\"https://en.wikipedia.org/wiki/Hydrogen_spectral_series\">this page</a></p>\n\n<p>I am more convinced.  I don't know if I'm going to have time to implement anything using this thought.  Curious if this has already been tried?</p>",
  "messages": [
    {
      "id": "427696",
      "postDate": "11/26/2018 00:59:20",
      "content": "<p>Since joining this competition I have believed that photometry is a more affordable spectroscopy and that clues to composition should be able to be gleaned from the ratios of the passbands and or the deltas in the activities of the passbands.</p>\n\n<p>Looking at <a href=\"https://astrobites.org/2016/12/02/classifying-supernovae/\">this article</a></p>\n\n<p>And <a href=\"https://en.wikipedia.org/wiki/Hydrogen_spectral_series\">this page</a></p>\n\n<p>I am more convinced.  I don't know if I'm going to have time to implement anything using this thought.  Curious if this has already been tried?</p>",
      "rawMarkdown": "Since joining this competition I have believed that photometry is a more affordable spectroscopy and that clues to composition should be able to be gleaned from the ratios of the passbands and or the deltas in the activities of the passbands.\n\nLooking at [this article][1]\n\nAnd [this page][2]\n\nI am more convinced.  I don't know if I'm going to have time to implement anything using this thought.  Curious if this has already been tried?\n\n  [1]: https://astrobites.org/2016/12/02/classifying-supernovae/\n  [2]: https://en.wikipedia.org/wiki/Hydrogen_spectral_series",
      "votes": null
    },
    {
      "id": "427712",
      "postDate": "11/26/2018 02:26:15",
      "content": "<p>This confirms that spectra is key for classifying supernovae.</p>\n\n<p>But we don't have spectra data here, not sure how you can use any of this to help classify what we have.</p>",
      "rawMarkdown": "This confirms that spectra is key for classifying supernovae.\n\nBut we don't have spectra data here, not sure how you can use any of this to help classify what we have.",
      "votes": null
    },
    {
      "id": "427717",
      "postDate": "11/26/2018 02:34:40",
      "content": "<p>Isn't photometry related to spectroscopy?  Spectroscopy is more detailed than photometry but it seems like the wavelengths in the spectrum still correspond to the passbands in a way that would allow you to sense an increased or decreased likelihood of either high / low Hydrogen content.  If there is a burst in a passband that is highly represented in Hydrogen's spectrum and less extreme a burst in a passband that isn't well represented in hydrogens spectrum you might infer that the burst has a a lot of hydrogen involvement.</p>",
      "rawMarkdown": "Isn't photometry related to spectroscopy?  Spectroscopy is more detailed than photometry but it seems like the wavelengths in the spectrum still correspond to the passbands in a way that would allow you to sense an increased or decreased likelihood of either high / low Hydrogen content.  If there is a burst in a passband that is highly represented in Hydrogen's spectrum and less extreme a burst in a passband that isn't well represented in hydrogens spectrum you might infer that the burst has a a lot of hydrogen involvement.",
      "votes": null
    },
    {
      "id": "427728",
      "postDate": "11/26/2018 03:17:46",
      "content": "<blockquote>\n  <p>Isn't photometry related to spectroscopy? </p>\n</blockquote>\n\n<p>Not really, photometry is an extremely crude approximation to spectroscopy.  You cannot identify spectral lines from it.</p>",
      "rawMarkdown": "&gt; Isn't photometry related to spectroscopy? \n\nNot really, photometry is an extremely crude approximation to spectroscopy.  You cannot identify spectral lines from it.",
      "votes": null
    },
    {
      "id": "427746",
      "postDate": "11/26/2018 04:12:15",
      "content": "<p>Supernova types are defined by the features in their spectra. Unfortunately, with photometry, you simply don't have enough signal-to-noise or resolution to be able to identify those spectral features. Spectroscopy takes a lot of telescope time, and for LSST we won't be able to get spectra for the vast majority of the objects that are discovered. The whole point of this challenge is to figure out how to identify objects without having their spectra.</p>",
      "rawMarkdown": "Supernova types are defined by the features in their spectra. Unfortunately, with photometry, you simply don't have enough signal-to-noise or resolution to be able to identify those spectral features. Spectroscopy takes a lot of telescope time, and for LSST we won't be able to get spectra for the vast majority of the objects that are discovered. The whole point of this challenge is to figure out how to identify objects without having their spectra.",
      "votes": null
    },
    {
      "id": "427811",
      "postDate": "11/26/2018 07:42:59",
      "content": "<p>I wasn't expecting to be able to identify spectral lines.  I'm thinking more along the lines of the UV / IR ratio may have a pattern for some objects and not for others.  It sounds like people are skeptical there will be enough resolution to draw any conclusions from this sort of analysis.  I did see some interesting things when I was tracking UV / IR ratio early on.  I put that aside because it was taking too long to extract the information.</p>",
      "rawMarkdown": "I wasn't expecting to be able to identify spectral lines.  I'm thinking more along the lines of the UV / IR ratio may have a pattern for some objects and not for others.  It sounds like people are skeptical there will be enough resolution to draw any conclusions from this sort of analysis.  I did see some interesting things when I was tracking UV / IR ratio early on.  I put that aside because it was taking too long to extract the information.",
      "votes": null
    },
    {
      "id": "427852",
      "postDate": "11/26/2018 09:18:26",
      "content": "<blockquote>\n  <p>UV / IR ratio  ...  It sounds like people are skeptical there will be enough resolution to draw any conclusions from this sort of analysis. </p>\n</blockquote>\n\n<p>This is not what was discussed in the article and in the page you pointed to.  </p>",
      "rawMarkdown": "&gt; UV / IR ratio  ...  It sounds like people are skeptical there will be enough resolution to draw any conclusions from this sort of analysis. \n\nThis is not what was discussed in the article and in the page you pointed to.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 427712,
      "author_name": "cpmpml",
      "author_url": "",
      "post_date": "11/26/2018 02:26:15",
      "content": "<p>This confirms that spectra is key for classifying supernovae.</p>\n\n<p>But we don't have spectra data here, not sure how you can use any of this to help classify what we have.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 427717,
      "author_name": "jimpsull",
      "author_url": "",
      "post_date": "11/26/2018 02:34:40",
      "content": "<p>Isn't photometry related to spectroscopy?  Spectroscopy is more detailed than photometry but it seems like the wavelengths in the spectrum still correspond to the passbands in a way that would allow you to sense an increased or decreased likelihood of either high / low Hydrogen content.  If there is a burst in a passband that is highly represented in Hydrogen's spectrum and less extreme a burst in a passband that isn't well represented in hydrogens spectrum you might infer that the burst has a a lot of hydrogen involvement.</p>",
      "votes": null,
      "replies": [
        {
          "id": 427728,
          "author_name": "cpmpml",
          "author_url": "",
          "post_date": "11/26/2018 03:17:46",
          "content": "<blockquote>\n  <p>Isn't photometry related to spectroscopy? </p>\n</blockquote>\n\n<p>Not really, photometry is an extremely crude approximation to spectroscopy.  You cannot identify spectral lines from it.</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 427746,
      "author_name": "kyleboone",
      "author_url": "",
      "post_date": "11/26/2018 04:12:15",
      "content": "<p>Supernova types are defined by the features in their spectra. Unfortunately, with photometry, you simply don't have enough signal-to-noise or resolution to be able to identify those spectral features. Spectroscopy takes a lot of telescope time, and for LSST we won't be able to get spectra for the vast majority of the objects that are discovered. The whole point of this challenge is to figure out how to identify objects without having their spectra.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 427811,
      "author_name": "jimpsull",
      "author_url": "",
      "post_date": "11/26/2018 07:42:59",
      "content": "<p>I wasn't expecting to be able to identify spectral lines.  I'm thinking more along the lines of the UV / IR ratio may have a pattern for some objects and not for others.  It sounds like people are skeptical there will be enough resolution to draw any conclusions from this sort of analysis.  I did see some interesting things when I was tracking UV / IR ratio early on.  I put that aside because it was taking too long to extract the information.</p>",
      "votes": null,
      "replies": [
        {
          "id": 427852,
          "author_name": "cpmpml",
          "author_url": "",
          "post_date": "11/26/2018 09:18:26",
          "content": "<blockquote>\n  <p>UV / IR ratio  ...  It sounds like people are skeptical there will be enough resolution to draw any conclusions from this sort of analysis. </p>\n</blockquote>\n\n<p>This is not what was discussed in the article and in the page you pointed to.  </p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "427696": "Since joining this competition I have believed that photometry is a more affordable spectroscopy and that clues to composition should be able to be gleaned from the ratios of the passbands and or the deltas in the activities of the passbands.\n\nLooking at [this article][1]\n\nAnd [this page][2]\n\nI am more convinced.  I don't know if I'm going to have time to implement anything using this thought.  Curious if this has already been tried?\n\n  [1]: https://astrobites.org/2016/12/02/classifying-supernovae/\n  [2]: https://en.wikipedia.org/wiki/Hydrogen_spectral_series",
    "427712": "This confirms that spectra is key for classifying supernovae.\n\nBut we don't have spectra data here, not sure how you can use any of this to help classify what we have.",
    "427717": "Isn't photometry related to spectroscopy?  Spectroscopy is more detailed than photometry but it seems like the wavelengths in the spectrum still correspond to the passbands in a way that would allow you to sense an increased or decreased likelihood of either high / low Hydrogen content.  If there is a burst in a passband that is highly represented in Hydrogen's spectrum and less extreme a burst in a passband that isn't well represented in hydrogens spectrum you might infer that the burst has a a lot of hydrogen involvement.",
    "427728": "&gt; Isn't photometry related to spectroscopy? \n\nNot really, photometry is an extremely crude approximation to spectroscopy.  You cannot identify spectral lines from it.",
    "427746": "Supernova types are defined by the features in their spectra. Unfortunately, with photometry, you simply don't have enough signal-to-noise or resolution to be able to identify those spectral features. Spectroscopy takes a lot of telescope time, and for LSST we won't be able to get spectra for the vast majority of the objects that are discovered. The whole point of this challenge is to figure out how to identify objects without having their spectra.",
    "427811": "I wasn't expecting to be able to identify spectral lines.  I'm thinking more along the lines of the UV / IR ratio may have a pattern for some objects and not for others.  It sounds like people are skeptical there will be enough resolution to draw any conclusions from this sort of analysis.  I did see some interesting things when I was tracking UV / IR ratio early on.  I put that aside because it was taking too long to extract the information.",
    "427852": "&gt; UV / IR ratio  ...  It sounds like people are skeptical there will be enough resolution to draw any conclusions from this sort of analysis. \n\nThis is not what was discussed in the article and in the page you pointed to."
  },
  "source": "meta"
}