<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=us-ascii">
</head>
<body style="word-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;" class="">
Hi Nadav, I just discussed answers to your questions with group in cc.
<div class=""><br class="">
</div>
<div class="">1. We used MITgcm in hydrostatic configuration.<br class="">
<div class=""><br class="">
</div>
<div class="">2. We used KPP because it was the parameterization used in the global model that provided the boundary conditions, so mostly for inertia reasons and due to lack of bandwidth to explore all possible options. But you are right in pointing out that
background diffusivity and viscosity in KPP is meant, in part, to represent internal wave breaking. Note that below the surface boundary layer, KPP includes a Richardson-number based representation of mixing. This is why it makes sense to turn off the background
diffusivity and viscosity terms when you are starting to explicitly resolve internal waves in the simulation.</div>
<div class=""><br class="">
</div>
<div class="">3. Breaking of IWs in our simulations is parameterized by the KPP Richardson-number based term, since our resolution is insufficient to resolve actual breaking. With the exception of the southwest corner of our domain, IWs do not have much opportunity
to shoal in our open-ocean regional simulations. I think this is a big difference with your domain, where much of the vertical mixing physics in your specific case may be driven by interactions with bottom bathymetry.</div>
<div class=""><br class="">
</div>
<div class="">4. For realistic representation of vertical shear, more vertical levels is definitely better. If you can afford to increase vertical levels, you definitely should, and compare with the coarser-vertical-resolution simulations to establish impact
of vertical resolution on the solution.</div>
<div class=""><br class="">
</div>
<div class="">Cheers,</div>
<div class=""><br class="">
<div class="">Dimitris Menemenlis </div>
<div><br class="">
</div>
<div><br class="">
<blockquote type="cite" class="">
<div class="">On Aug 26, 2022, at 5:11 AM, Nadav Mantel <<a href="mailto:nadav.mantel@mail.huji.ac.il" class="">nadav.mantel@mail.huji.ac.il</a>> wrote:</div>
<br class="Apple-interchange-newline">
<div class="">
<div dir="ltr" class="">Hi Dimitris,
<div class=""><br class="">
</div>
<div class="">Thank you for responding to my email.</div>
<div class="">I have a few questions regarding the paper and the study.</div>
<div class="">1. In the study did you run a non-hydrostatic or hydrostatic model?</div>
<div class="">2. Why in the first place use KPP in IW parameterizations? Obviously it is an extremely well known and widely used vertical parameterization package but isn't meant to include IW in the first place. What was the motivation not to research KL10,
or one of the richardson packages (PP81, MY82, GGL90) which were designed for this reason?</div>
<div class="">3. If I understood correctly, in the study you researched IW presence, its shear and other physical aspects of the IW spectrum, yet you didn't look at the shoaling and breaking of said IWs, once again, if I understood correctly because I might
be missing some crucial point. Would the findings of the study (remove KPP background parameters) also hold for this type of parameterization?</div>
<div class="">4. In the study you use high vertical resolutions. I have "inherited" a model which uses 32 layers, in an elongated and rather deep basin. In your opinion, should I increase vertical layers or do the MITgcm packages overcome the reduced resolutions?</div>
<div class=""><br class="">
</div>
<div class="">Thanks so much for your input and for sharing the paper,</div>
<div class="">Nadav</div>
</div>
<br class="">
<div class="gmail_quote">
<div dir="ltr" class="gmail_attr">On Thu, Aug 25, 2022 at 11:15 PM Menemenlis, Dimitris (US 329B) <<a href="mailto:dimitris.menemenlis@jpl.nasa.gov" class="">dimitris.menemenlis@jpl.nasa.gov</a>> wrote:<br class="">
</div>
<blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
<div style="overflow-wrap: break-word;" class="">Hi Navad, here is a recently-published paper that may be relevant to some of the questions that you ask: <a class="">https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2022GL099614</a>
<div class=""><br class="">
</div>
<div class="">But note that the above study pertains to the representation of internal waves in an open ocean region, as opposed to a narrow, elongated gulf.</div>
<div class=""><br class="">
</div>
<div class="">
<div class="">Dimitris Menemenlis </div>
<div class=""><br class="">
<blockquote type="cite" class="">
<div class="">On Aug 25, 2022, at 2:49 AM, Nadav Mantel <<a class="">nadav.mantel@mail.huji.ac.il</a>> wrote:</div>
<br class="">
<div class="">
<div dir="ltr" class="">Hey everyone,
<div class=""><br class="">
</div>
<div class="">This is a topic that comes up every now and then but I would still really appreciate someone's help.</div>
<div class="">I am running a non-hydrostatic regional model for the Gulf of Aqaba\Eilat, an elongated semi-enclosed basin [~200 km long, ~30 km wide ~1000 m deep] with an entry through a shallow sill in the south. The model runs with a horizontal resolution
of 300 m and 32 vertical layers with the differences in depth of 5,10, 12.5, 15, 17.5, ... m and a time step of 10s. I'm forcing tides through the southern boundary using the obcs package. The barotropic flow crossing the Straits of Tiran (our shallow sill)
creates large amplitude (50 m in observations) internal waves along the gulf.</div>
<div class=""><br class="">
</div>
<div class="">We would like to parameterize the shoaling and general breaking of the internal waves using one of the many packages MITgcm has to offer but we have had some troubles.</div>
<div class=""><br class="">
</div>
<div class="">We first intended to use the KL10 package, yet we had uncharacteristic seasonal mixing and when checking the viscosity output we had weird results where in some places we got viscosity coefficients of 3 [m^2/s] whereas the rest of the gulf is
much smaller. It could be possibly due to non-linear EOS, I saw in the mitgcm bug tracker that there are issues with anything but linear EOS use in the package.</div>
<div class=""><br class="">
</div>
<div class="">We moved on to the richardson number packages, starting with pp81 as it is the simplest. Our problem was that we consistently got the maximum viscosity (a default of 1) in almost all of the second layer, creating once more uncharacteristic mixing
with very large diffusivities (60 m^2/s).</div>
<div class=""><br class="">
</div>
<div class="">We thought about using KPP as it was the package used in simulating the gulf up until now, but it wasn't designed for this purpose and we are skeptical it would produce good results. </div>
<div class=""><br class="">
</div>
<div class="">Kind of at a loss at this point, and it would be really amazing to get someone's feedback via zoom call on their experiences with internal wave parameterization using any package for regional models.</div>
<div class=""><br class="">
</div>
<div class="">Thanks,</div>
<div class="">Nadav Mantel</div>
</div>
_______________________________________________<br class="">
MITgcm-support mailing list<br class="">
<a class="">MITgcm-support@mitgcm.org</a><br class="">
<a class="">https://urldefense.us/v3/__http://mailman.mitgcm.org/mailman/listinfo/mitgcm-support__;!!PvBDto6Hs4WbVuu7!Z9iqcR5gWjF-J_Axv-Z5IwjHE3NIjlnukR1MoYYCJsMmMBhhzukH4-GMMAi7fj3nI9TXJBvuY6M$</a>
<br class="">
</div>
</blockquote>
</div>
<br class="">
</div>
</div>
_______________________________________________<br class="">
MITgcm-support mailing list<br class="">
<a class="">MITgcm-support@mitgcm.org</a><br class="">
<a rel="noreferrer" class="">http://mailman.mitgcm.org/mailman/listinfo/mitgcm-support</a><br class="">
</blockquote>
</div>
</div>
</blockquote>
</div>
<br class="">
</div>
</div>
</body>
</html>