<html style="direction: ltr;">
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
    <style type="text/css">body p { margin-bottom: 0cm; margin-top: 0pt; } </style>
  </head>
  <body bidimailui-charset-is-forced="true" style="direction: ltr;"
    text="#000000" bgcolor="#FFFFFF">
    <br>
    <br>
    <div class="moz-cite-prefix">On 08/31/2017 11:03 PM, Ken Mankoff
      wrote:<br>
    </div>
    <blockquote type="cite"
      cite="mid:B3BE7D14-2DDD-496E-8787-1314692940A7@gmail.com">
      <meta http-equiv="content-type" content="text/html; charset=utf-8">
      <div>Hi Micha,</div>
      <div><br>
      </div>
      <div>I understand what you wrote (I think). I get that the basin
        product from r.watershed does not change with SFD or MFD. I
        think this is because the flow direction raster from r.watershed
        only provides the primary flow direction.</div>
    </blockquote>
    Yes, each pixel gets a single flow direction, one of 8 directions.<br>
    <br>
    <blockquote type="cite"
      cite="mid:B3BE7D14-2DDD-496E-8787-1314692940A7@gmail.com">
      <div><br>
      </div>
      <div>But the accumulation map doesn't know about boundaries or
        basins, does it? At a divide, can water can flow equally in all
        8 directions? If so, I think that at the boundary of the basin
        delineated by </div>
    </blockquote>
    That can happen only for a single pixel at the very peak of a
    mountain. <br>
    <blockquote type="cite"
      cite="mid:B3BE7D14-2DDD-496E-8787-1314692940A7@gmail.com">
      <div>r.water.outlet there may be a cell that contributed 49%. The
        flow direction would show this cell flows away from the basin
        boundary because 51% of it does do that, so it is not in the
        basin. If I use this basin as a mask, I'm losing 49% of that
        cell, and the many upstream cells that contribute to it. <br>
      </div>
    </blockquote>
    You won't have many upstream cells for those cells along the basin
    boundary, only the few that drain exactly along the watershed
    divide. And these cells will eventually become part of one basin or
    the other. So I guess theoretically you could "loose" some flow
    accumulation: 49% of the drainage from those few cells right along
    the watershed divide. But this would be an accurate representation
    of reality, since that 49% is actually not part of the basin you're
    investigating.<br>
    <br>
    <blockquote type="cite"
      cite="mid:B3BE7D14-2DDD-496E-8787-1314692940A7@gmail.com">
      <div><br>
      </div>
      <div>The inefficient method, running r.watershed 14,000 times,
        never considers basins and is therefore not impacted by this
        issue. <br>
      </div>
    </blockquote>
    Not sure I understand that.<br>
    The only way that r.watershed can return different results is if you
    input a different elevation grid. The r.watershed module is strictly
    a geo-morphological analysis - nothing to do with real runoff, only
    the terrain, slopes, flow direction, etc. It creates a theoretical
    model of where runoff would go. The only situation that I can
    envision where you would rerun r.watershed is when massive earthwork
    was done, and you have a new/revised elevation dataset.<br>
    <br>
    <blockquote type="cite"
      cite="mid:B3BE7D14-2DDD-496E-8787-1314692940A7@gmail.com">
      <div><br>
          -k. 
        <div><br>
        </div>
        <div>Please excuse brevity. Sent from pocket computer with tiny
          non-haptic feedback keyboard. </div>
      </div>
      <div><br>
        On 31 Aug 2017, at 21:18, Micha Silver <<a
          href="mailto:tsvibar@gmail.com" moz-do-not-send="true">tsvibar@gmail.com</a>>
        wrote:<br>
        <br>
      </div>
      <blockquote type="cite">
        <div>
          <meta http-equiv="Content-Type" content="text/html;
            charset=utf-8">
          The r.water.outlet module takes as input a flow direction
          raster that needs to be created first by r.watershed. So the
          SFD/MFD question is irrelevant at this stage. When you first
          ran r.watershed you chose which method to use for determining
          flow direction for each pixel. Further, SFD/MFD influences
          only the stream routing, not the total number of cells in the
          basin. I'm pretty sure that if you run r.watershed once with
          MFD and again with SFD, you'll get the same basin, only with
          slightly different stream networks. <br>
          AFAIK there should never be a situation where water is
          directed out of the basin. So all cells that flow into cell C
          are then directed downstream to your final drainage point.<br>
          <br>
          <div class="moz-cite-prefix">On 08/31/2017 10:04 PM, Ken
            Mankoff wrote:<br>
          </div>
          <blockquote type="cite"
            cite="mid:81312416-5630-44ED-9B5B-EBC2CCF0FE42@gmail.com">
            <meta http-equiv="content-type" content="text/html;
              charset=utf-8">
            <div>Yes. This! What you wrote. </div>
            <div><br>
            </div>
            <div>But the issue is that r.water.outlet make basins based
              on SFD, right? What if there are 10,000 cells that feed
              into cell C at x,y, and then cell C feeds 49% (based on
              MFD) into the basin. These 10,000 cells are not included
              in the r.water.outlet basin, are they?<br>
              <br>
                -k. 
              <div><br>
              </div>
              <div>Please excuse brevity. Sent from pocket computer with
                tiny non-haptic feedback keyboard. </div>
            </div>
            <div><br>
              On 31 Aug 2017, at 20:51, Micha Silver <<a
                href="mailto:tsvibar@gmail.com" moz-do-not-send="true">tsvibar@gmail.com</a>>
              wrote:<br>
              <br>
            </div>
            <blockquote type="cite">
              <div>
                <meta http-equiv="Content-Type" content="text/html;
                  charset=utf-8">
                I'm also not clear what you are asking. But risking a
                guess:<br>
                You could run r.water.outlet *1 time* to get the basin.
                Then use that raster as a MASK, so that the next process
                will address only the pixels within the basin. Now do a
                loop with r.univar on all 14,000 flow rasters, and
                you'll get 14,000 results with total, min, max, mean,
                etc of the basin pixels for each of the flow rasters.<br>
                <br>
                --<br>
                Micha<br>
                <br>
                <div class="moz-cite-prefix">On 08/31/2017 09:30 PM,
                  Thomas Adams wrote:<br>
                </div>
                <blockquote type="cite"
cite="mid:CAGxgkWhJ8PWAh+Av1u5KUcLHvqa5NZZvqx+SQzJ0eVz2h3e-0A@mail.gmail.com">
                  <div dir="ltr">
                    <div>
                      <div>Ken,<br>
                        <br>
                      </div>
                      You "want 14,000 values" of what?? Your original
                      email stated you were "trying to determine flow
                      past a drainage basin outlet" -- r.watershed does
                      NOT do this, if indeed this is what you want. And
                      you say you have "14,000 flow rasters to be used
                      as input" -- what exactly are these 'flow
                      rasters'; what is your goal? I may not
                      understand...<br>
                      <br>
                    </div>
                    Tom<br>
                    <div class="gmail_extra"><br>
                      <div class="gmail_quote">On Thu, Aug 31, 2017 at
                        2:24 PM, Ken Mankoff <span dir="ltr"><<a
                            href="mailto:mankoff@gmail.com"
                            target="_blank" moz-do-not-send="true">mankoff@gmail.com</a>></span>
                        wrote:<br>
                        <blockquote class="gmail_quote"
                          style="margin:0px 0px 0px
                          0.8ex;border-left:1px solid
                          rgb(204,204,204);padding-left:1ex">
                          <div dir="auto">
                            <div>Hi Tom,</div>
                            <div><br>
                            </div>
                            <div>I have 1 DEM and 14,000 flow rasters to
                              be used as input. I want 14,000 values,
                              one at a specific coordinate from each acc
                              output. </div>
                            <div><br>
                            </div>
                            <div>I can do this by running r.watershed
                              14,000 times. That is slow, unless I'm
                              missing something (e.g. It works with
                              I.group variables or Time Series data more
                              efficiently). </div>
                            <div><br>
                            </div>
                            <div>An alternative approach is possible if
                              I knew the complete drainage basin *and*
                              the fractional value of each cell that
                              contributed to the basin. In this case I
                              don't need to route. But basins from
                              r.watershed or r.water.outlet, I think,
                              use SFD not MFD (no cell is ever in 2
                              basins, are they?), and I don't know how
                              to get the fractional contribution from
                              each cell. </div>
                            <div><br>
                                -k. 
                              <div><br>
                              </div>
                              <div>Please excuse brevity. Sent from
                                pocket computer with tiny non-haptic
                                feedback keyboard. </div>
                            </div>
                            <div>
                              <div class="gmail-h5">
                                <div><br>
                                  On 31 Aug 2017, at 19:59, Thomas Adams
                                  <<a href="mailto:tea3rd@gmail.com"
                                    target="_blank"
                                    moz-do-not-send="true">tea3rd@gmail.com</a>>
                                  wrote:<br>
                                  <br>
                                </div>
                                <blockquote type="cite">
                                  <div>
                                    <div dir="ltr">
                                      <div>
                                        <div>
                                          <div>Ken,<br>
                                            <br>
                                          </div>
                                          I'm confused about what you
                                          are trying to do with
                                          r.watershed, because the
                                          output from the module is:<br>
                                          <br>
                                          accumulation=name <br>
                                          Name for output accumulation
                                          raster map <br>
                                          Number of cells that drain
                                          through each cell <br>
                                          tci=name <br>
                                          Name for output topographic
                                          index ln(a / tan(b)) map <br>
                                          spi=name <br>
                                          Stream power index a * tan(b)
                                          <br>
                                          Name for output raster map <br>
                                          drainage=name <br>
                                          Name for output drainage
                                          direction raster map <br>
                                          basin=name <br>
                                          Name for output basins raster
                                          map <br>
                                          stream=name <br>
                                          Name for output stream
                                          segments raster map <br>
                                          half_basin=name <br>
                                          Name for output half basins
                                          raster map <br>
                                          Each half-basin is given a
                                          unique value <br>
                                          length_slope=name <br>
                                          Name for output slope length
                                          raster map <br>
                                          Slope length and steepness
                                          (LS) factor for USLE <br>
                                          slope_steepness=name <br>
                                          Name for output slope
                                          steepness raster map <br>
                                          Slope steepness (S) factor for
                                          USLE <br>
                                          <br>
                                        </div>
                                        I think you want a hydrologic
                                        model, and r.watershed is NOT
                                        that. What are you trying to
                                        obtain?<br>
                                        <br>
                                      </div>
                                      Tom<br>
                                      <div>
                                        <div><br>
                                          <br>
                                          <br>
                                        </div>
                                      </div>
                                      <div class="gmail_extra"><br>
                                        <div class="gmail_quote">On Thu,
                                          Aug 31, 2017 at 1:47 PM, Ken
                                          Mankoff <span dir="ltr"><<a
href="mailto:mankoff@gmail.com" target="_blank" moz-do-not-send="true">mankoff@gmail.com</a>></span>
                                          wrote:<br>
                                          <blockquote
                                            class="gmail_quote"
                                            style="margin:0px 0px 0px
                                            0.8ex;border-left:1px solid
rgb(204,204,204);padding-left:1ex">
                                            <div dir="ltr">Hi List,
                                              <div><br>
                                              </div>
                                              <div>I'm trying to
                                                determine flow past a
                                                drainage basin outlet.
                                                The complicating factor
                                                is that I need to do
                                                this each day for 40
                                                years. If I do
                                                "r.watershed" ~14,000
                                                times I'll get the
                                                results, but it will
                                                take 3 days. It seems
                                                that r.watershed is
                                                likely calculating many
                                                things each time through
                                                the loop. Is there a
                                                more efficient way to
                                                this? A flag to
                                                r.watershed that isn't
                                                documented? Something
                                                with time-series?</div>
                                              <div><br>
                                              </div>
                                              <div>Alternatively,
                                                because I only need the
                                                flow at the outlet, I
                                                could calculate the
                                                basin, not route the
                                                flow, and instead sum
                                                the values in the basin.
                                                I assume this would take
                                                seconds or minutes
                                                rather than days. In
                                                this case I'm not sure
                                                of the best way to
                                                define the basin. I
                                                tried doing
                                                r.water.outlet upstream
                                                from the outlet, but I
                                                think this uses SFD,
                                                which means the basin
                                                may be significantly
                                                underestimated.</div>
                                              <div><br>
                                              </div>
                                              <div>I also tried
                                                inverting/flipping the
                                                DEM and then running
                                                r.watershed with
                                                convergence=1, and a
                                                flow equal to 0
                                                everywhere except 1000
                                                at the outlet (now the
                                                source due to the
                                                inversion) to see where
                                                it flooded upstream (now
                                                downstream due to the
                                                inversion). This didn't
                                                seem to work... because
                                                basins are filled and
                                                flow routes to the edge
                                                of the DEM, I could not
                                                pick out the </div>
                                              <div><br>
                                              </div>
                                              <div>Any advice how to
                                                either a) efficiently
                                                route 14,000 FLOW
                                                rasters over 1 DEM or b)
                                                determine the full basin
                                                will be much
                                                appreciated.</div>
                                              <div><br>
                                              </div>
                                              <div>Thanks,</div>
                                              <div><br>
                                              </div>
                                              <div>    -k.</div>
                                            </div>
                                            <br>
______________________________<wbr>_________________<br>
                                            grass-user mailing list<br>
                                            <a
                                              href="mailto:grass-user@lists.osgeo.org"
                                              target="_blank"
                                              moz-do-not-send="true">grass-user@lists.osgeo.org</a><br>
                                            <a
                                              href="https://lists.osgeo.org/mailman/listinfo/grass-user"
                                              rel="noreferrer"
                                              target="_blank"
                                              moz-do-not-send="true">https://lists.osgeo.org/mailma<wbr>n/listinfo/grass-user</a><br>
                                          </blockquote>
                                        </div>
                                        <br>
                                        <br>
                                        <div dir="ltr">
                                          <div>
                                            <div dir="ltr">
                                              <div dir="ltr">
                                                <div><br>
                                                </div>
                                              </div>
                                            </div>
                                          </div>
                                        </div>
                                      </div>
                                    </div>
                                  </div>
                                </blockquote>
                              </div>
                            </div>
                          </div>
                        </blockquote>
                      </div>
                      <br>
                      <div dir="ltr">
                        <div>
                          <div dir="ltr">
                            <div dir="ltr">
                              <div><br>
                              </div>
                            </div>
                          </div>
                        </div>
                      </div>
                    </div>
                  </div>
                  <br>
                  <fieldset class="mimeAttachmentHeader"></fieldset>
                  <br>
                  <pre wrap="">_______________________________________________
grass-user mailing list
<a class="moz-txt-link-abbreviated" href="mailto:grass-user@lists.osgeo.org" moz-do-not-send="true">grass-user@lists.osgeo.org</a>
<a class="moz-txt-link-freetext" href="https://lists.osgeo.org/mailman/listinfo/grass-user" moz-do-not-send="true">https://lists.osgeo.org/mailman/listinfo/grass-user</a></pre>
                </blockquote>
                <br>
                <pre class="moz-signature" cols="72">-- 
Micha Silver
Ben Gurion Univ.
Sde Boker, Remote Sensing Lab
cell: +972-523-665918</pre>
              </div>
            </blockquote>
          </blockquote>
          <br>
          <pre class="moz-signature" cols="72">-- 
Micha Silver
Ben Gurion Univ.
Sde Boker, Remote Sensing Lab
cell: +972-523-665918</pre>
        </div>
      </blockquote>
    </blockquote>
    <br>
    <pre class="moz-signature" cols="72">-- 
Micha Silver
Ben Gurion Univ.
Sde Boker, Remote Sensing Lab
cell: +972-523-665918</pre>
  </body>
</html>