There are time delays and slightly different range vectors fro each transducer. The idea about the current zarr structure is that we keep the data in a xarray structure. We should keep these data without regridding/averaging, and we need a separate range vector per frequency. Hiowever, we should align the range bins by shifting up down to maximize the correspondence without doing andy interpolation/regridding.
@iambaim check with @ingeke and provide a suggestion.
There are time delays and slightly different range vectors fro each transducer. The idea about the current zarr structure is that we keep the data in a xarray structure. We should keep these data without regridding/averaging, and we need a separate range vector per frequency. Hiowever, we should align the range bins by shifting up down to maximize the correspondence without doing andy interpolation/regridding.
@iambaim check with @ingeke and provide a suggestion.