1.4. Readers for input data

Depending on the input data you are using, you might have to add a reader to read them. Several types of data are already managed, but you can any other input data by adding it to the readers in the Readers directory.

cd Readers

Three sets of readers exist:

topo_reader.py	Reader for topographic/bathymetric data (etopo5, etopo2, etopo1, srtm30, homonim, gebco…)
ibc_reader.py	Reader for initial and boundary conditions datasets (mercator, eccov4, soda…)
tides_reader.py	Reader for tides datasets (tpxo,fes2014…)

They basically contain a routine, lookvar, to associate the input variable names to commonly used variable names in the croco_pytools with a dictionnary.

In the following we describe the three type of reader:

Note

To access variable name in a netcdf, ncdump is your friend

1.4.1. Topography

To build bathymetry from a dataset, the dictionnary needs at least four keys:

lon	Longitude
lat	Latitude
topo	Name of the 2-D bathymetric field
zaxis	Direction of z axis (up or down)

A supplementary key srtm needs to be added and set to True if the input dataset is STRM30plus

1.4.2. Initial and boundary conditions

Reader for initial and boundary conditions (ibc) handles 2D and 3D fields and their evolution in time. It requires their coordinates:

depth	Z-coordinates variables
lonr, lonu, lonv	Longitude name for Rho,U,V grid. If only one grid put the same name everywhere
latr, latu, latv	Latitude name for Rho,U,V grid. If only one grid put the same name everywhere
time, time_dim	Time variable and the associated time dimension

And the following fields:

ssh	Sea Surface Elevation variable name
u	Eastward velocity (m/s) variable name
v	Northward velocity (m/s) variable name
temp	temperature (C) variable name
salt	salinity (PSU) variable name

You can also add other tracers if needed following the same pattern:

'tracer':'tracer_name_in_dataset'

1.4.3. Tides

As for ibc_reader, you need to specify the keys for geographical coordinates

lonr, lonu, lonv	Longitude name for Rho, U, V grid. If only one grid put the same name everywhere
latr, latu, latv	Latitude name for Rho, U, V grid. If only one grid put the same name everywhere

Tide data consist in three fields (elevation, easwtward velocity, northward velocity), and each field has two components. Depending on the input dataset, those two components can be either Amplitude/Phase or Real/Imaginary part.

If your dataset follows an Amplitude/Phase, keys to declare are:

H_a	Elevation amplitude
H_p	Elevation phase
U_a	Eastward velocity amplitude
U_p	Eastward velocity phase
V_a	Northward velocity amplitude
V_p	Northward velocity phase

Or if it follows a Real/Imaginary format, keys to declare are:

H_r	Elevation real part
H_i	Elevation imaginary part
U_r	Eastward velocity real part
U_i	Eastward velocity imaginary part
V_r	Northward velocity real part
V_i	Northward velocity imaginary part