[MITgcm-support] MITgcm Floats Package Depth Interpolation Issue

[John Paul Irving]

Hello All,

I am working on a project utilizing the MITgcm flt package. I am running the model on a curvilinear grid with multiple processors. In my flt input files i prescribe the vertical position of the flts in fractal indices, and the FLT_MAP_K2R function in the flt_mapping.F file converts these fractal indices to a depth in meters. However the floats are not being mapped to their exact positions in the simulation. For particles launched between 50-500 meters, the initial starting positions are less than 1m off the desired depth. But for particles at 1000m launch depth, they are about 20 meters off. I have 45 vertical levels as follows. the dz between layers is 10m in the upper150m, 25m until 600m, 50m until 1000, and then with larger levels to the bottom.

I am having trouble figuring out why it cannot interpolate the exact depth values that i originally used to interpolate to an indice? I also have confusion about why 2 floats launched at the same depth are then read in at slightly different depths?

Was wondering if anyone else has tried to do this and seen this issue, thanks!


John Irving


jpi18 at my.fsu.edu

Florida State University

Department of Earth, Ocean, & Atmospheric Science




John Irving


jpi18 at my.fsu.edu

Florida State University

Department of Earth, Ocean, & Atmospheric Science
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.mitgcm.org/pipermail/mitgcm-support/attachments/20200331/a583cfa6/attachment.html>