Source code for improver.precipitation.snow_splitter
# (C) Crown Copyright, Met Office. All rights reserved.
#
# This file is part of 'IMPROVER' and is released under the BSD 3-Clause license.
# See LICENSE in the root of the repository for full licensing details.
"""Module to separate snow and rain contributions from a precipitation diagnostic"""
from typing import Tuple, Union
import numpy as np
from iris import Constraint
from iris.cube import Cube, CubeList
from improver import BasePlugin
from improver.cube_combiner import Combine
from improver.utilities.common_input_handle import as_cubelist
from improver.utilities.cube_checker import assert_spatial_coords_match
[docs]
class SnowSplitter(BasePlugin):
"""A class to separate the contribution of rain or snow from the precipitation
rate/accumulation. This is calculated using the probability of rain and snow
at the surface to determine what fraction of the precipitation rate/accumulation
is rain or snow.
"""
[docs]
def __init__(self, output_is_rain: bool):
"""
Sets up Class
Args:
output_is_rain:
A boolean where True means the plugin will output rain and False means the
output is snow.
"""
self.output_is_rain = output_is_rain
[docs]
def process(self, *cubes: Union[Cube, CubeList]) -> Cube:
"""
Splits the precipitation cube data into a snow or rain contribution.
Whether the output is a rate or accumulation will depend on the precipitation_cube.
output_is_rain will determine whether the outputted cube is a cube of snow or rain.
The probability of rain and snow at the surfaces should only contain 1's where the
phase is present at the surface and 0's where the phase is not present
at the surface. These cubes need to be consistent with each other such that both
rain and snow can't be present at the surface (e.g. at no grid square can both
diagnostics have a probability of 1).
A grid of coefficients is calculated by an arbitrary function that maps
(0,0) -> 0.5, (1,0) -> 1, (0,1) -> 0 where the first coordinate is the probability
for the variable that will be outputted. This grid of coefficients is then multiplied
by the precipitation rate/accumulation to split out the contribution of the desired
variable.
Args:
cubes:
containing:
rain_cube:
Cube of the probability of rain at the surface.
snow_cube:
Cube of the probability of snow at the surface.
precip_cube:
Cube of either precipitation rate or precipitation accumulation.
Returns:
Cube of rain/snow (depending on self.output_is_rain) rate/accumulation (depending on
precipitation cube). The name will be an updated version of precip_cube.name().
"precipitation" will be replaced with "rainfall" or "snowfall" and "lwe\_" will be
removed for rain output.
Raises:
ValueError: If, at some grid square, both snow_cube and rain_cube have a probability of
0
""" # noqa: W605 (flake8 objects to \_ in "lwe\_" that is required for Sphinx)
cubes = as_cubelist(*cubes)
rain_cube, snow_cube, precip_cube = self.separate_input_cubes(cubes)
assert_spatial_coords_match([rain_cube, snow_cube, precip_cube])
if np.any((rain_cube.data + snow_cube.data) > 1):
raise ValueError(
"""There is at least 1 grid square where the probability of snow
at the surface plus the probability of rain at the surface is greater
1."""
)
if self.output_is_rain:
required_cube = rain_cube
other_cube = snow_cube
name = "rainfall"
else:
required_cube = snow_cube
other_cube = rain_cube
name = "snowfall"
# arbitrary function that maps combinations of rain and snow probabilities
# to an appropriate coefficient
coefficient_cube = (required_cube - other_cube + 1) / 2
coefficient_cube.data = coefficient_cube.data.astype(np.float32)
new_name = precip_cube.name().replace("precipitation", name)
if self.output_is_rain:
new_name = new_name.replace("lwe_", "")
output_cube = Combine(operation="*", new_name=new_name)(
[precip_cube, coefficient_cube]
)
return output_cube
