Add additional models to provider JSON

[robbibt]

@tsutterley I've added the following new models to the relevant provider JSON files:

• FES2012
• FES2014_extrapolated
• FES2022_extrapolated
• GOT5.6
• GOT5.6_extrapolated
• TPXO8-atlas-nc

Have tested these locally and they seem to work:

However, I did find that my model definitions that use wildcard matching (e.g. "model_file": "TPXO8_atlas_v1/hf.*_tpxo8_atlas_30c_v1.nc") didn't recognise files in this new implementation, so I had to manually create the model file and constituent lists. I'm not experienced with this stuff so it would be great if you could double-check my changes and make sure I have specified them correctly/matched the right files to the right constituents!

[robbibt]

Is specifying constituents necessary, or will pyTMD auto-guess the constituent names? (previously, I had simply used "model_file": "fes2012/data/*_FES2012_SLEV.nc", but that wasn't being recognised here)

[tsutterley]

Right, the glob string capability isn't being run for the database as I figured the database will be more "maintained".

[robbibt]

The one file I couldn't find a match for was Z0_FES2012_SLEV.nc - have left it out here for now

[robbibt]

Is E2 a match for eps2?

[tsutterley]

I removed the constituents from the database and am running the files through the parser as before. Should make maintenance easier :)

[robbibt]

Is La2 a match for l2? Or lambda2?

[tsutterley]

lambda2

[robbibt]

Had previous used "model_file": "TPXO8_atlas_v1/hf.*_tpxo8_atlas_30c_v1.nc", so have attempted to adapt this to a list of model files

[robbibt]

Code used to test:

from pyTMD.compute import tide_elevations
import pandas as pd
import numpy as np


fes2014_extrap = tide_elevations(
    x=[122.269294],
    y=[-18.057430],
    delta_time=pd.date_range("2020", "2021", periods=50),
    DIRECTORY="/gdata1/data/tide_models_clipped/",
    MODEL="FES2014_extrapolated",
    EPSG=4326,
    TIME="datetime",
    EXTRAPOLATE=True,
    CUTOFF=np.inf,
)

fes2012 = tide_elevations(
    x=[122.269294],
    y=[-18.057430],
    delta_time=pd.date_range("2020", "2021", periods=50),
    DIRECTORY="/gdata1/data/tide_models_clipped/",
    MODEL="FES2012",
    EPSG=4326,
    TIME="datetime",
    EXTRAPOLATE=True,
    CUTOFF=np.inf,
)

fes2022_extrap = tide_elevations(
    x=[122.269294],
    y=[-18.057430],
    delta_time=pd.date_range("2020", "2021", periods=50),
    DIRECTORY="/gdata1/data/tide_models_clipped/",
    MODEL="FES2022_extrapolated",
    EPSG=4326,
    TIME="datetime",
    EXTRAPOLATE=True,
    CUTOFF=np.inf,
)

got56 = tide_elevations(
    x=[122.269294],
    y=[-18.057430],
    delta_time=pd.date_range("2020", "2021", periods=50),
    DIRECTORY="/gdata1/data/tide_models_clipped/",
    MODEL="GOT5.6",
    EPSG=4326,
    TIME="datetime",
    EXTRAPOLATE=True,
    CUTOFF=np.inf,
)

got56_extrap = tide_elevations(
    x=[122.269294],
    y=[-18.057430],
    delta_time=pd.date_range("2020", "2021", periods=50),
    DIRECTORY="/gdata1/data/tide_models_clipped/",
    MODEL="GOT5.6_extrapolated",
    EPSG=4326,
    TIME="datetime",
    EXTRAPOLATE=True,
    CUTOFF=np.inf,
)

tpxo8 = tide_elevations(
    x=[122.269294],
    y=[-18.057430],
    delta_time=pd.date_range("2020", "2021", periods=50),
    DIRECTORY="/gdata1/data/tide_models_clipped/",
    MODEL="TPXO8-atlas-nc",
    EPSG=4326,
    TIME="datetime",
    EXTRAPOLATE=True,
    CUTOFF=np.inf,
)

# Plot
pd.Series(fes2014_extrap, name="FES2014_extrapolated").plot(legend=True, figsize=(10, 5))
pd.Series(fes2012, name="FES2012").plot(legend=True)
pd.Series(fes2022_extrap, name="FES2022_extrapolated").plot(legend=True)
pd.Series(got56, name="GOT5.6").plot(legend=True)
pd.Series(got56_extrap, name="GOT5.6_extrapolated").plot(legend=True)
pd.Series(tpxo8, name="TPXO8-atlas-nc").plot(legend=True)

[tsutterley]

thanks @robbibt! Want to add yourself to the list of contributors?