<html><head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
  </head>
  <body>
    <font size="+1"><font face="monospace">Andy -<br>
        <br>
        Perhaps I am still missing something about the overall setup,
        but if the domain is periodic in the lateral (vertical)
        direction and the initial data is axially symmetric, why does
        the noise does not reappear at the bottom (left side in the
        movie)? It seems to grow stronger away from the equator.<br>
      </font></font><br>
    <font size="+1"><font face="monospace"><font size="+1"><font face="monospace">So it seems like the outer boundary
            conditions, perhaps? I imagine perturbations are created
            from the very beginning. Plotfiles have limited accuracy and
            using checkpoint files is preferred for the debugging
            purposes. Also, one can always produce specific diagnostic
            directly from the code, perhaps something simply such as
            output a certain data at certain mesh locations.<br>
            <br>
            The question then is what is the origin of perturbations and
            why they grow.<br>
            <br>
            As for the source, early versions of the FLASH USM solver
            were known to produce spurious, small scale magnetic fields,
            which was due to numerical issues involved in spectral
            decomposition. I believe that problem was successfully
            resolved, at least judging from our simulation results. I
            would not completely exclude this possibility, and perhaps a
            simplified version of your problem could offer some insight
            here.<br>
            <br>
          </font></font>There are usually two reasons for the observed
        zone-to-zone oscillations to appear. Either there is a diffusion
        process that is not resolved correctly in time and the solution
        blows up, or else small perturbations in one solution component
        are not correctly coupled to other components. The former
        problem is due to violation of the CFL condition, and the latter
        is known as the odd-even decoupling and typically caused by
        approximation used in solving the Riemann problem (flux
        calculation).<br>
        <br>
        I do not think you are using any diffusion in the current setup
        (I do not know what long wavelength perturbations are and how
        you suppress them, though). But you can experiment with various
        approximate Riemann solvers provided with the code.<br>
        <br>
        There was some discussion of modeling strongly stratified
        systems with equilibria in the context of solar atmosphere. I
        remember there were some stability issues reported.<br>
        <br>
        Tomek<br>
        --</font></font><br>
    <div class="moz-cite-prefix">On 4/9/21 11:21 AM, Andy Sha Liao
      wrote:<br>
    </div>
    <blockquote type="cite" cite="mid:CAKfEzWB0DwNavLbad=PPxpZkRE+=JvD1fNAA+e5J+zBTQDvY1Q@mail.gmail.com">
      
      <div dir="ltr">
        <div dir="ltr">
          <div dir="ltr">Tomek,
            <div><br>
            </div>
            <div><br>
            </div>
            <div>I suppress the long wave oscillations and plot the
              lineout for radial velocity over alfvén velocity along z
              (lateral direction) at r={.0200,.0500,.1000} cm in
              {cyan,magenta,green}: <a href="https://urldefense.com/v3/__https://drive.google.com/file/d/14T52ybx9ukQDee3zzaHdd-8VRs2Tp-C0/view?usp=sharing__;!!PhOWcWs!mVh2uuX9UIKjIAFE20HKpSEMByw5oYYQd5CayQbNnKO6h7Bs6OnpI7mbzIFWQQ$" target="_blank" moz-do-not-send="true">https://drive.google.com/file/d/14T52ybx9ukQDee3zzaHdd-8VRs2Tp-C0/view?usp=sharing</a></div>
            <div><br>
            </div>
            <div>Zooming in on the green lineout: <a href="https://urldefense.com/v3/__https://drive.google.com/file/d/1wqay_ZmPtpvwuej8C0A9ZhqCdJszy1-l/view?usp=sharing__;!!PhOWcWs!mVh2uuX9UIKjIAFE20HKpSEMByw5oYYQd5CayQbNnKO6h7Bs6OnpI7mG4QBS8w$" target="_blank" moz-do-not-send="true">https://drive.google.com/file/d/1wqay_ZmPtpvwuej8C0A9ZhqCdJszy1-l/view?usp=sharing</a></div>
            <div><br>
            </div>
            <div>The zoomed movie starts at frame 33, but the teeth are
              observable much earlier, when their amplitudes are as low
              ~1 ppm of their centroid value.    </div>
            <div><br>
            </div>
            <div>The curious thing is that the teeth grow independently
              of the lateral mean value of the quantity, and before they
              become large enough to become disruptive, their pattern
              stays coherent. </div>
            <div><br>
            </div>
            <div><br>
            </div>
            <div>Andy</div>
          </div>
        </div>
      </div>
      <br>
      <div class="gmail_quote">
        <div dir="ltr" class="gmail_attr">On Thu, Apr 8, 2021 at 4:20 PM
          Tomasz Plewa <<a href="mailto:tplewa@fsu.edu" target="_blank" moz-do-not-send="true">tplewa@fsu.edu</a>>
          wrote:<br>
        </div>
        <blockquote class="gmail_quote" style="margin:0px 0px 0px
          0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
          <div> <font size="+1"><font face="monospace">Andy -<br>
                <br>
                I would think that with that many cells per scale height
                the mesh resolution is adequate.<br>
              </font></font><br>
            <font size="+1"><font face="monospace"><font size="+1"><font face="monospace">I have a sense something happens
                    near the lower-left boundary (perhaps across the
                    equatorial plane), and a perturbation from that
                    region sweeps through the rest of the domain.<br>
                    <br>
                  </font></font>Whatever happens breaks the assumed
                radial symmetry and lateral uniformity. I would be more
                interested in lateral (momentum, pressure gradients,
                field gradients) rather than radial components as the
                former should nominally remain zero at all times. You
                could try to suppress lateral fluxes and see whether any
                particular part of the domain eventually becomes a
                source of perturbations. In either case, the simulation
                is in trouble as soon as lateral perturbations develop
                and their source is not controlled.<br>
                <br>
                I assume the mesh is uniform, correct?<br>
                <br>
                In the plots, is the symmetry axis at the left and
                equatorial plane at the bottom?<br>
                <br>
                Not sure if assuming reflecting boundary at Rmax is
                justified? Depending how it is implemented in the code,
                there might be a jump in some magnetic field components.
                I would think that zero gradient/outflow conditions
                might be safer as any perturbations will likely feed
                back into the domain, and one would want to avoid that.
                (It seems Tummel et al. actually stress that particular
                point.)<br>
                <br>
                Have you tried a 1D version of this setup?<br>
                <br>
                Which version of FLASH are you using?<br>
                <br>
                Tomek<br>
                --</font></font><br>
            <div>On 4/8/21 11:12 AM, Andy Sha Liao wrote:<br>
            </div>
            <blockquote type="cite">
              <div dir="ltr">
                <div dir="ltr">
                  <div dir="ltr">
                    <div dir="ltr">Tomasz, 
                      <div><br>
                      </div>
                      <div><br>
                      </div>
                      <div>Thanks for taking interest in our problem.
                        Let me send you a few movies from my GDrive so
                        you have a better picture.</div>
                      <div><br>
                      </div>
                      <div>First, the movie from which the stills were
                        sent to you previously:</div>
                      <div><br>
                      </div>
                      <div><a href="https://urldefense.com/v3/__https://drive.google.com/file/d/1pKKQzVHn8Qm9RVMNpX0CINJlqULoIfzw/view?usp=sharing__;!!PhOWcWs!nWLN84YtVshbYV71fKfEPMDMQ9Jrl6lAUw98Vz8sWBW6dtvRjvzPqMCWBYUCXQ$" target="_blank" moz-do-not-send="true">https://drive.google.com/file/d/1pKKQzVHn8Qm9RVMNpX0CINJlqULoIfzw/view?usp=sharing</a><br>
                      </div>
                      <div><br>
                      </div>
                      <div>The movie shows density, radial momentum, and
                        azimuthal magnetic field.</div>
                      <div><br>
                      </div>
                      <div>Next, a movie of the radial lineout near the
                        equatorial plane of the domain, of radial
                        velocity as a fraction of the characteristic
                        alfven speed ~240 km/s. I show results from two
                        different interpolation orders, olive is 3rd
                        order, magenta is 2nd order:</div>
                      <div><br>
                      </div>
                      <div><a href="https://urldefense.com/v3/__https://drive.google.com/file/d/1pCh_Lb4P6TbBWxNOItLotb7-SoWLqvmq/view?usp=sharing__;!!PhOWcWs!nWLN84YtVshbYV71fKfEPMDMQ9Jrl6lAUw98Vz8sWBW6dtvRjvzPqMDM5Tj5lA$" target="_blank" moz-do-not-send="true">https://drive.google.com/file/d/1pCh_Lb4P6TbBWxNOItLotb7-SoWLqvmq/view?usp=sharing</a><br>
                      </div>
                      <div><br>
                      </div>
                      <div>The problem is not the long waves, but the
                        sawteeth oscillations. In other simulations, I
                        suppressed the long waves, but the sawteeth
                        still came in on schedule.  </div>
                      <div><br>
                      </div>
                      <div>To answer your questions, the scale height,
                        or characteristic length of the pinch is 0.0910
                        cm, as found in the reference in the previous
                        message, and the resolution is 32 cells per
                        characteristic length. We also ran up to 256
                        cells per characteristic length, but the problem
                        doesn't go away.</div>
                      <div><br>
                      </div>
                      <div><br>
                      </div>
                      <div>Andy</div>
                    </div>
                  </div>
                </div>
              </div>
            </blockquote>
            <br>
          </div>
        </blockquote>
      </div>
    </blockquote>
    <br>
  </body>
</html>