`amescap.Ncdf_wrapper`

Ncdf_wrapper archives data into netCDF format. It serves as a wrapper for creating netCDF files.

Third-party Requirements:

numpy

amescap.FV3_utils

scipy.io

netCDF4

os

datetime

Module Contents

Classes

`Fort`	A class that generates an object from a fort.11 file. The new file
`Ncdf`	netCDF wrapper for archiving data in netCDF format. Alex Kling.

class Fort(filename=None, description_txt='')

Bases: object

A class that generates an object from a fort.11 file. The new file will have netCDF file attributes. Alex Kling.

EX:

f.variables.keys()
f.variables['var'].long_name
f.variables['var'].units
f.variables['var'].dimensions

Create a Fort object using the following:

f=Fort('/Users/akling/test/fort.11/fort.11_0684')

Public methods can be used to generate FV3-like netCDF files:

f.write_to_fixed()
f.write_to_average()
f.write_to_daily()
f.write_to_diurn()

Parameters:: object (_type_) – _description_
Returns:: _description_
Return type:: _type_

class Fort_var(input_vals, name_txt, long_name_txt, units_txt, dimensions_tuple)

Bases: numpy.ndarray

Sub-class that emulates a netCDF-like variable by adding the name, long_name, units, and dimensions attributes to a numpy array. Inner class for fortran_variables (Fort_var) that comprise the Fort file. Alex Kling

A useful resource on subclassing is available at: https://numpy.org/devdocs/reference/arrays.classes.html

Note

Because we use an existing numpy.ndarray to define the object, we do not call __array_finalize__(self, obj)

Parameters:: np.ndarray (_type_) – _description_
Returns:: _description_
Return type:: _type_

__abs__()

__add__(value)

__and__(value)

__array__(dtype=None)

__array_wrap__(obj)

classmethod __class_getitem__(value)

__contains__(key)

__copy__()

__deepcopy__(memo)

__divmod__(value)

__eq__(value): Return self==value.

__float__()

__floordiv__()

__ge__(value): Return self>=value.

__getitem__(key)

__gt__(value): Return self>value.

__iadd__(value)

__iand__(value)

__ifloordiv__(value)

__ilshift__(value)

__imod__(value)

__imul__(value)

__int__()

__invert__()

__ior__(value)

__ipow__(value)

__irshift__(value)

__isub__(value)

__itruediv__(value)

__ixor__(value)

__le__(value): Return self<=value.

__len__()

__lshift__(value)

__lt__(value): Return self<value.

__matmul__(value)

__mod__(value)

__mul__(value)

__ne__(value): Return self!=value.

__neg__()

__or__(value)

__pos__()

__pow__()

__repr__(): Return repr(self).

__rshift__()

__setitem__(key, value)

__str__(): Return str(self).

__sub__(value)

__truediv__(value)

__xor__(value)

all(axis=None, out=None, keepdims=False)

any(axis=None, out=None, keepdims=False)

argmax(axis=None, out=None)

argmin(axis=None, out=None)

argpartition(kth, axis=-1, kind='introselect', order=None)

argsort(axis=-1, kind='quicksort', order=None)

astype(dtype, order='K', casting='unsafe', subok=True, copy=True)

byteswap(inplace=False)

choose(choices, out=None, mode='raise')

clip(min=None, max=None, out=None)

compress(condition, axis=None, out=None)

conj()

conjugate()

copy(order='C')

cumprod(axis=None, dtype=None, out=None)

cumsum(axis=None, dtype=None, out=None)

diagonal(offset=0, axis1=0, axis2=1)

dot(b, out=None)

dump(file)

dumps()

fill(value)

flatten(order='C')

getfield(dtype, offset=0)

item(*args)

itemset(*args)

max(axis=None, out=None)

mean(axis=None, dtype=None, out=None, keepdims=False)

min(axis=None, out=None, keepdims=False)

newbyteorder(new_order='S')

nonzero()

partition(kth, axis=-1, kind='introselect', order=None)

prod(axis=None, dtype=None, out=None, keepdims=False)

ptp(axis=None, out=None)

put(indices, values, mode='raise')

ravel(order='C')

repeat(repeats, axis=None)

reshape(shape, order='C')

resize(new_shape, refcheck=True)

round(decimals=0, out=None)

searchsorted(v, side='left', sorter=None)

setfield(val, dtype, offset=0)

setflags(write=None, align=None, uic=None)

sort(axis=-1, kind='quicksort', order=None)

squeeze(axis=None)

std(axis=None, dtype=None, out=None, ddof=0, keepdims=False)

sum(axis=None, dtype=None, out=None, keepdims=False)

swapaxes(axis1, axis2)

take(indices, axis=None, out=None, mode='raise')

tobytes(order='C')

tofile(fid, sep='', format='%s')

tolist()

tostring(order='C')

trace(offset=0, axis1=0, axis2=1, dtype=None, out=None)

transpose(*axes)

var(axis=None, dtype=None, out=None, ddof=0, keepdims=False)

view(dtype=None, type=None)

close()

write_to_average(day_average=5): Create average file (e.g., N-day averages [N=5 usually])

write_to_daily(): Create daily file (continuous time series)

write_to_diurn(day_average=5): Create diurn file (variables organized by time of day & binned (typically 5-day bins)

write_to_fixed(): Create fixed file (all static variables)

class Ncdf(filename=None, description_txt='', action='w', ncformat='NETCDF4_CLASSIC')

Bases: object

netCDF wrapper for archiving data in netCDF format. Alex Kling.

Usage:

from netcdf_wrapper import Ncdf

Fgeo = 0.03 # W/m2, a constant
sfcT = np.ones((24,8)) # surface temperature

# Create file
filename = "/path/to/myfile.nc"
description = "results from new simulation, Alex 01-01-19"
Log = Ncdf(filename, description)

# Save the constant (``Fgeo``) to the file
Log.add_constant('Fgeo', Fgeo, "geothermal flux", "W/m2")

# Save the sfcT array to the file
Log.add_dimension('Nx', 8)
Log.add_dimension('time', 24)

Log.log_variable('sfcT', sfcT, ('time', 'Nx'),
                 'soil temperature', 'K')

Log.close()

Parameters:: object (_type_) – _description_
Returns:: netCDF file

add_constant(variable_name, value, longname_txt='', units_txt='')

add_dim_with_content(dimension_name, DATAin, longname_txt='', units_txt='', cart_txt='')

Function to define a dimension and add a variable at the same time. Equivalent to add_dimension() followed by log_axis1D():

lon_array = np.linspace(0, 360)

EX:

Log.add_dim_with_content("lon", lon_array, "longitudes",
                         "degree", "X")

add_dimension(dimension_name, length)

close()

copy_Ncaxis_with_content(Ncdim_var): Copy a netCDF DIMENSION variable (e.g., Ncdim = f.variables["lon"]). If the dimension does not exist yet, it will be created

copy_Ncvar(Ncvar, swap_array=None): Copy a netCDF variable from another file (e.g., Ncvar = f.variables["ucomp"]). All dimensions must already exist. If swap_array is provided, the original values are swapped with this array.

copy_all_dims_from_Ncfile(Ncfile_in, exclude_dim=[], time_unlimited=True): Copy all variables, dimensions, and attributes from another netCDF file

copy_all_vars_from_Ncfile(Ncfile_in, exclude_var=[])

log_axis1D(variable_name, DATAin, dim_name, longname_txt='', units_txt='', cart_txt='')

EX:

Log.log_axis1D("areo", areo, "time", "degree", "T")

log_variable(variable_name, DATAin, dim_array, longname_txt='', units_txt='', datatype='f4')

EX:

Log.log_variable("sfcT", sfcT, ("time", "Nx"),
                 "soil temperature", "K")

merge_files_from_list(Ncfilename_list, exclude_var=[])

print_dimensions()

print_variables()

amescap.Ncdf_wrapper

Module Contents

Classes

`amescap.Ncdf_wrapper`