• chiliedogg@lemmy.world
    link
    fedilink
    English
    arrow-up
    8
    ·
    edit-2
    11 months ago

    Why follow a line when I could make a DEM from some LIDAR data, then run Aspect and Accumulation functions and dileneate watersheds?

    • KrokanteBamischijf@feddit.nl
      link
      fedilink
      English
      arrow-up
      13
      ·
      11 months ago

      You’re now playing GIS DnD:

      The LiDAR dataset you’re using was scanned in a forested area and doesn’t include any secondary return data. As a result, your watersheds are occluded and the data doesn’t provide the greatest cartography.

      What do you do?

        • KrokanteBamischijf@feddit.nl
          link
          fedilink
          English
          arrow-up
          1
          ·
          11 months ago

          Success! You manage to build a somewhat useful TIN from the data. Upon further inspection, the contours of the watersheds you were looking for are vaguely visible. Occlusion from the surrounding trees has had an impact on the dataset though, and it seems the noise wasn’t filtered out of the dataset properly, leaving you with the occasional ridiculously stretched triangle to work with. Generating nice vector data from this will prove challenging.

          How do you proceed?

          • chiliedogg@lemmy.world
            link
            fedilink
            English
            arrow-up
            1
            ·
            11 months ago

            Knowing this isn’t old growth and that the trees have been harvested for lumber multiple times, I summon my unpaid intern and have her search for maps containing contours in a year closely following a tree harvest.

            I cast Quadrangle with a conformity bonus from my UTM state plane.

            • KrokanteBamischijf@feddit.nl
              link
              fedilink
              English
              arrow-up
              1
              ·
              11 months ago

              [consults with other DMs]: “Wait…they can’t do that, can they?”… “Really? Well, damn”.

              Your intern somehow manages to convince the local forestry corporation to share an old field map of the region. Problem is, the file is not georeferenced and she is having trouble doing so. Not being familiar with affine transformations, she picks some horrible references and her attempt has an error of a couple meters.

              This is your chance to teach your intern a valuable skill and you’re very close to finalizing the map of the region. What is your next move?