1 clim_recal.utils.xarray

clim_recal.utils.xarray

1.1 Classes

Name Description
XarrayTimeSeriesCalcManager Manage cacluations over time for .nc files.

1.1.1 XarrayTimeSeriesCalcManager

clim_recal.utils.xarray.XarrayTimeSeriesCalcManager(self, path=climate_data_mount_path() / 'Raw/UKCP2.2/', save_folder=Path('../docs/assets/cpm-raw-medians'), sub_path=Path('latest'), variables=VariableOptions.cpm_values(), runs=RunOptions.preferred_and_first(), method_name='median', time_dim_name='time', regex=CPM_REGEX, source_folders=list())

Manage cacluations over time for .nc files.

1.1.1.1 Attributes

Name Type Description
path PathLike Path to aggreate raw files from, following standard cpm hierarchy.
sub_path PathLike | None A subpath to parse, like ‘latest’ for UKCPM2.2.
save_folder PathLike Where to save resulting summary nc files.
variables Sequence[str | VariableOptions] Which variables to include
runs Sequence[str | RunOptions] Which RunOptions to include.
method_name str Which method to use to aggreate. Must be a standard xarray Dataset method.
time_dim_name str Name of the temporal dim on the joined .nc files.
regex str Check .nc paths to match and then aggregate.
source_folders list List of folders to iterate over, filled via path if None.

1.1.1.2 Examples

>>> tasmax_hads_1980_raw = getfixture('tasmax_hads_1980_raw')
>>> if not tasmax_hads_1980_raw:
...     pytest.skip(mount_or_cache_doctest_skip_message)
>>> tmp_save: Path = getfixture('tmp_path') / 'xarray-time-series-summary-manager'
>>> xr_var_managers = XarrayTimeSeriesCalcManager(save_folder=tmp_save)
>>> save_paths: tuple[Path, ...] = xr_var_managers.save_joined_xr_time_series(stop=2, ts_stop=2)
Aggregating '05' 'tasmax' (1/2)...
Aggregating '06' 'tasmax' (2/2)...
>>> pprint(save_paths)
(...Path('.../median-tasmax-05.nc'),
 ...Path('.../median-tasmax-06.nc'))
>>> pprint(sorted(tmp_save.iterdir()))
[...Path('.../median-tasmax-05.nc'),
 ...Path('.../median-tasmax-06.nc')]

1.2 Functions

Name Description
annual_group_xr_time_series Return and plot a Dataset of time series temporally overlayed.
apply_geo_func Apply a Callable to netcdf_source file and export via to_netcdf.
cftime360_to_date Convert a Datetime360Day into a date.
cftime_range_gen Convert a banded time index a banded standard (Gregorian).
check_xarray_path_and_var_name Check and return a T_Dataset instances and included variable name.
convert_xr_calendar Convert cpm 360 day time series to a standard 365/366 day time series.
cpm_check_converted Check if cpm_xr_time_series is likely already reprojected.
cpm_reproject_with_standard_calendar Convert raw cpm_xr_time_series to an 365/366 days and 27700 coords.
cpm_xarray_to_standard_calendar Convert a CPM nc file of 360 day calendars to standard calendar.
crop_xarray Crop xr_time_series with crop_path shapefile.
ensure_xr_dataset Return xr_time_series as a xarray.Dataset instance.
file_name_to_start_end_dates Return dates of file name with date_format-date_format structure.
gdal_warp_wrapper Execute the gdalwrap function within python.
generate_360_to_standard Return array_to_expand 360 days expanded to 365 or 366 days.
get_cpm_for_coord_alignment Check if cpm_for_coord_alignment is a Dataset, process if a Path.
hads_resample_and_reproject Resample HADs xarray time series to 2.2km.
interpolate_xr_ts_nans Interpolate nan values in a Dataset time series.
join_xr_time_series_var Join a set of xr_time_series files chronologically.
plot_xarray Plot da with **kwargs to path.
region_crop_file_name Generate a file name for a regional crop.
xr_reproject_crs Reproject source_xr to target_xr coordinate structure.

1.2.1 annual_group_xr_time_series

clim_recal.utils.xarray.annual_group_xr_time_series(joined_xr_time_series, variable_name, groupby_method='time.dayofyear', method_name='median', time_dim_name='time', regex=CPM_REGEX, start=0, stop=None, step=1, plot_path='annual-aggregated.png', time_stamp=True, **kwargs)

Return and plot a Dataset of time series temporally overlayed.

1.2.1.1 Parameters

Name Type Description Default
joined_xr_time_series T_Dataset | PathLike Provide existing Dataset to aggregate and plot (otherwise check path). required
variable_name str A variable name to specify for data expected. If none that will be extracted and checked from the files directly. required
groupby_method str xarray method to aggregate time of xr_time_series. 'time.dayofyear'
method_name str xarray method to calculate for plot. 'median'
time_dim_name str Name of time dimension in passed files. 'time'
regex str A str to filter files within path CPM_REGEX
start int What point to start indexing path results from. 0
stop int | None What point to stop indexing path restuls from. None
step int How many paths to jump between when iterating between stop and start. 1
plot_path PathLike Path to save plot to. 'annual-aggregated.png'
time_stamp bool Whether to include a time stamp in the plot_path name. True
**kwargs Additional parameters to pass to plot_xarray. {}

1.2.1.2 Examples

>>> tasmax_cpm_1980_raw = getfixture('tasmax_cpm_1980_raw')
>>> if not tasmax_cpm_1980_raw:
...     pytest.skip(mount_or_cache_doctest_skip_message)
>>> tasmax_cpm_1980_raw_path = getfixture('tasmax_cpm_1980_raw_path').parents[1]
>>> results: T_Dataset = annual_group_xr_time_series(
...     tasmax_cpm_1980_raw_path, 'tasmax', stop=3)
>>> results
<xarray.Dataset> ...
Dimensions:    (dayofyear: 360)
Coordinates:
  * dayofyear  (dayofyear) int64 ... 1 2 3 4 5 6 7 ... 355 356 357 358 359 360
Data variables:
    tasmax     (dayofyear) float64 ... 9.2 8.95 8.408 8.747 ... 6.387 8.15 9.132

1.2.2 apply_geo_func

clim_recal.utils.xarray.apply_geo_func(source_path, func, export_folder, new_path_name_func=None, to_netcdf=True, to_raster=False, export_path_as_output_path_kwarg=False, return_results=False, **kwargs)

Apply a Callable to netcdf_source file and export via to_netcdf.

1.2.2.1 Parameters

Name Type Description Default
source_path PathLike netcdf file to apply func to. required
func ReprojectFuncType Callable to modify netcdf. required
export_folder PathLike Where to save results. required
new_path_name_func Callable[[Path], Path] | None Callabe to generate new path to save to. None
to_netcdf bool Whether to call to_netcdf() method on results Dataset. True
to_raster bool Whether to call rio.to_raster() on results Dataset. False
export_path_as_output_path_kwarg bool Whether to add output_path = export_path to kwargs passed to func. Meant for cases calling gdal_warp_wrapper. False
return_results bool Whether to return results, which would be a Dataset or GDALDataset (the latter if gdal_warp_wrapper is used). False
**kwargs Other parameters passed to func call. {}

1.2.2.2 Returns

Type Description
Either a Path to generated file or converted xarray object.

1.2.3 cftime360_to_date

clim_recal.utils.xarray.cftime360_to_date(cf_360)

Convert a Datetime360Day into a date.

1.2.3.1 Examples

>>> cftime360_to_date(Datetime360Day(1980, 1, 1))
datetime.date(1980, 1, 1)

1.2.4 cftime_range_gen

clim_recal.utils.xarray.cftime_range_gen(time_data_array, **kwargs)

Convert a banded time index a banded standard (Gregorian).

1.2.5 check_xarray_path_and_var_name

clim_recal.utils.xarray.check_xarray_path_and_var_name(xr_time_series, variable_name, ignore_warnings=True)

Check and return a T_Dataset instances and included variable name.

1.2.6 convert_xr_calendar

clim_recal.utils.xarray.convert_xr_calendar(xr_time_series, align_on=DEFAULT_CALENDAR_ALIGN, calendar=CFCalendarSTANDARD, use_cftime=False, missing_value=np.nan, interpolate_na=False, ensure_output_type_is_dataset=False, interpolate_method=DEFAULT_INTERPOLATION_METHOD, keep_crs=True, keep_attrs=True, limit=1, engine=NETCDF4_XARRAY_ENGINE, check_cftime_cols=None, cftime_range_gen_kwargs=None)

Convert cpm 360 day time series to a standard 365/366 day time series.

1.2.6.1 Notes

Short time examples (like 2 skipped out of 8 days) raises: ValueError("date_range_like was unable to generate a range as the source frequency was not inferable.")

1.2.6.2 Parameters

Name Type Description Default
xr_time_series T_DataArray | T_Dataset | PathLike A DataArray or Dataset to convert to calendar time. required
align_on ConvertCalendarAlignOptions Whether and how to align calendar types. DEFAULT_CALENDAR_ALIGN
calendar CFCalendar Type of calendar to convert xr_time_series to. CFCalendarSTANDARD
use_cftime bool Whether to enforce cftime vs datetime64 time format. False
missing_value Any | None Missing value to populate missing date interpolations with. np.nan
keep_crs bool Reapply initial Coordinate Reference System (CRS) after time projection. True
interpolate_na bool Whether to apply temporal interpolation for missing values. False
interpolate_method InterpOptions Which InterpOptions method to apply if interpolate_na is True. DEFAULT_INTERPOLATION_METHOD
keep_attrs bool Whether to keep all attributes on after interpolate_na True
limit int Limit of number of continuous missing day values allowed in interpolate_na. 1
engine XArrayEngineType Which XArrayEngineType to use in parsing files and operations. NETCDF4_XARRAY_ENGINE
extrapolate_fill_value If True, then pass fill_value=extrapolate. See: * https://docs.xarray.dev/en/stable/generated/xarray.Dataset.interpolate_na.html * https://docs.scipy.org/doc/scipy/reference/generated/scipy.interpolate.interp1d.html#scipy.interpolate.interp1d required
check_cftime_cols tuple[str] | None Columns to check cftime format on None
cftime_range_gen_kwargs dict[str, Any] | None Any kwargs to pass to cftime_range_gen None

1.2.6.3 Raises

Type Description
ValueError Likely from xarray calling date_range_like.

1.2.6.4 Returns

Type Description
T_DataArrayOrSet Converted xr_time_series to specified calendar with optional interpolation.

1.2.6.5 Notes

Certain values may fail to interpolate in cases of 360 -> 365/366 (Gregorian) calendar. Examples include projecting CPM data, which is able to fill in measurement values (e.g. tasmax) but the year and month_number variables have nan values

1.2.6.6 Examples

2 Note a new doctest needs to be written to deal

3 with default year vs date parameters

>>> xr_360_to_365_datetime64: T_Dataset = convert_xr_calendar(
...     xarray_spatial_4_years_360_day, align_on="date")
>>> xr_360_to_365_datetime64.sel(
...     time=slice("1981-01-30", "1981-02-01"),
...     space="Glasgow").day_360
<xarray.DataArray 'day_360' (time: 3)>...
Coordinates:
  * time     (time) datetime64[ns] ...1981-01-30 1981-01-31 1981-02-01
    space    <U10 ...'Glasgow'
>>> xr_360_to_365_datetime64_interp: T_Dataset = convert_xr_calendar(
...     xarray_spatial_4_years_360_day, interpolate_na=True)
>>> xr_360_to_365_datetime64_interp.sel(
...     time=slice("1981-01-30", "1981-02-01"),
...     space="Glasgow").day_360
<xarray.DataArray 'day_360' (time: 3)>...
array([0.23789282, 0.5356328 , 0.311945  ])
Coordinates:
  * time     (time) datetime64[ns] ...1981-01-30 1981-01-31 1981-02-01
    space    <U10 ...'Glasgow'
>>> convert_xr_calendar(xarray_spatial_6_days_2_skipped)
Traceback (most recent call last):
   ...
ValueError: `date_range_like` was unable to generate a range as the source frequency was not inferable.

3.0.1 cpm_check_converted

clim_recal.utils.xarray.cpm_check_converted(cpm_xr_time_series)

Check if cpm_xr_time_series is likely already reprojected.

3.0.1.1 Parameters

Name Type Description Default
cpm_xr_time_series T_Dataset | PathLike Dataset instance or Path to check. required

3.0.1.2 Returns

Type Description
True if all of the methics are True, else False

3.0.2 cpm_reproject_with_standard_calendar

clim_recal.utils.xarray.cpm_reproject_with_standard_calendar(cpm_xr_time_series, variable_name=None, close_temp_paths=True, force=False)

Convert raw cpm_xr_time_series to an 365/366 days and 27700 coords.

3.0.2.1 Notes

Currently makes UTM coordinate structure

3.0.2.2 Parameters

Name Type Description Default
cpm_xr_time_series T_Dataset | PathLike Dataset (or path to load as Dataset) expected to be in raw UKCPM format, with 360 day years and a rotated coordinate system. required
variable_name str | None Name of variable used, usually a measure of climate change like tasmax and tasmin. None

3.0.2.3 Returns

Type Description
Final xarray Dataset after spatial and temporal changes.

3.0.2.4 Examples

>>> tasmax_cpm_1980_raw = getfixture('tasmax_cpm_1980_raw')
>>> if not tasmax_cpm_1980_raw:
...     pytest.skip(mount_or_cache_doctest_skip_message)
>>> tasmax_cpm_1980_365_day: T_Dataset = cpm_reproject_with_standard_calendar(
...     cpm_xr_time_series=tasmax_cpm_1980_raw,
...     variable_name="tasmax")
Warp:      ...nc ...100%...
Translate: ...tif ...100%...
>>> tasmax_cpm_1980_365_day
<xarray.Dataset>
Dimensions:              (time: 365, x: 493, y: 607)
Coordinates:
  * time                 (time) datetime64[ns]...
  * x                    (x) float64...
  * y                    (y) float64...
    transverse_mercator  |S1...
    spatial_ref          int64...
Data variables:
    tasmax               (time, y, x) float32...
Attributes: (12/18)
    ...
>>> tasmax_cpm_1980_365_day.dims
Frozen...({'time': 365, 'x': 493, 'y': 607})

3.0.3 cpm_xarray_to_standard_calendar

clim_recal.utils.xarray.cpm_xarray_to_standard_calendar(cpm_xr_time_series, include_bnds_index=False)

Convert a CPM nc file of 360 day calendars to standard calendar.

3.0.3.1 Parameters

Name Type Description Default
cpm_xr_time_series T_Dataset | PathLike A raw xarray of the form provided by CPM. required
include_bnds_index bool Whether to fix bnds indexing in returned Dataset. False

3.0.3.2 Returns

Type Description
Dataset calendar converted to standard (Gregorian).

3.0.4 crop_xarray

clim_recal.utils.xarray.crop_xarray(xr_time_series, crop_box, **kwargs)

Crop xr_time_series with crop_path shapefile.

3.0.4.1 Parameters

Name Type Description Default
xr_time_series T_Dataset | PathLike Dataset or path to netcdf file to load and crop. required
crop_box BoundingBoxCoords Instance of BoundingBoxCoords with coords. required
kwargs Any additional parameters to pass to clip {}

3.0.4.2 Returns

Type Description
T_Dataset Spatially cropped xr_time_series Dataset with final_crs spatial coords.

3.0.4.3 Examples

>>> from clim_recal.utils.data import GlasgowCoordsEPSG27700
>>> from numpy.testing import assert_allclose
>>> tasmax_cpm_1980_raw = getfixture('tasmax_cpm_1980_raw')
>>> if not tasmax_cpm_1980_raw:
...     pytest.skip(mount_or_cache_doctest_skip_message)
>>> tasmax_cpm_1980_365_day: T_Dataset = cpm_reproject_with_standard_calendar(
...     cpm_xr_time_series=tasmax_cpm_1980_raw,
...     variable_name="tasmax")
>>> cropped = crop_xarray(
...     tasmax_cpm_1980_365_day,
...     crop_box=GlasgowCoordsEPSG27700)
>>> assert_allclose(cropped.rio.bounds(),
...                 GlasgowCoordsEPSG27700.as_rioxarray_tuple(),
...                 rtol=.01)
>>> tasmax_cpm_1980_365_day.sizes
Frozen({'x': 529, 'y': 653, 'time': 365})
>>> cropped.sizes
Frozen({'x': 10, 'y': 8, 'time': 365})

3.0.5 ensure_xr_dataset

clim_recal.utils.xarray.ensure_xr_dataset(xr_time_series, default_name='to_convert')

Return xr_time_series as a xarray.Dataset instance.

3.0.5.1 Parameters

Name Type Description Default
xr_time_series T_DataArrayOrSet Instance to check and if necessary to convert to Dataset. required
default_name Name to give returned Dataset if xr_time_series.name is empty. 'to_convert'

3.0.5.2 Returns

Type Description
T_Dataset Converted (or original) Dataset.

3.0.5.3 Examples

>>> ensure_xr_dataset(xarray_spatial_4_days)
<xarray.Dataset>...
Dimensions:      (time: 5, space: 3)
Coordinates:
  * time         (time) datetime64[ns] ...1980-11-30 1980-12-01 ... 1980-12-04
  * space        (space) <U10 ...'Glasgow' 'Manchester' 'London'
Data variables:
    xa_template  (time, space) float64 ...0.5488 0.7152 ... 0.9256 0.07104

3.0.6 file_name_to_start_end_dates

clim_recal.utils.xarray.file_name_to_start_end_dates(path, date_format=CLI_DATE_FORMAT_STR)

Return dates of file name with date_format-date_format structure.

3.0.6.1 Parameters

Name Type Description Default
path PathLike Path to file required
date_format str Format of date for strptime CLI_DATE_FORMAT_STR

3.0.6.2 Examples

The examples below are meant to demonstrate usage, and the significance of when the last date is included or not by default.

>>> from .core import date_range_generator
>>> tif_365_path: Path = (Path('some') /
...     'folder' /
...     'pr_rcp85_land-cpm_uk_2.2km_06_day_20761201-20771130.tif')
>>> start_date, end_date = file_name_to_start_end_dates(tif_365_path)
>>> start_date
datetime.datetime(2076, 12, 1, 0, 0)
>>> end_date
datetime.datetime(2077, 11, 30, 0, 0)
>>> dates: tuple[date, ...] =  tuple(
...     date_range_generator(start_date=start_date,
...                          end_date=end_date,
...                          inclusive=True))
>>> dates[:3]
(datetime.datetime(2076, 12, 1, 0, 0),
 datetime.datetime(2076, 12, 2, 0, 0),
 datetime.datetime(2076, 12, 3, 0, 0))
>>> len(dates)
365
>>> tif_366_path: Path = (Path('some') /
...     'folder' /
...     'pr_rcp85_land-cpm_uk_2.2km_06_day_20791201-20801130.tif')
>>> from pandas import date_range
>>> dates = date_range(*file_name_to_start_end_dates(tif_366_path))
>>> len(dates)
366

3.0.7 gdal_warp_wrapper

clim_recal.utils.xarray.gdal_warp_wrapper(input_path, output_path, output_crs=BRITISH_NATIONAL_GRID_EPSG, output_x_resolution=None, output_y_resolution=None, copy_metadata=True, return_path=True, format=GDALNetCDFFormatStr, multithread=True, warp_dict_options=DEFAULT_WARP_DICT_OPTIONS, use_tqdm_progress_bar=True, tqdm_file_name_chars=TQDM_FILE_NAME_PRINT_CHARS_INDEX, resampling_method=None, supress_warnings=True, **kwargs)

Execute the gdalwrap function within python.

This is following a script in the bash/ folder that uses this programme:

f=$1 # The first argument is the file to reproject
fn=${f/Raw/Reprojected_infill} # Replace Raw with Reprojected_infill in the filename
folder=`dirname $fn` # Get the folder name
mkdir -p $folder # Create the folder if it doesn't exist
gdalwarp -t_srs 'EPSG:27700' -tr 2200 2200 -r near -overwrite $f "${fn%.nc}.tif" # Reproject the file

3.0.7.1 Parameters

Name Type Description Default
input_path PathLike Path with CPRUK files to resample. srcDSOrSrcDSTab in Warp. required
output_path PathLike Path to save resampled input_path file(s) to destNameOrDestDS in Warp. required
output_crs str Coordinate system to convert input_path file(s) to. dstSRS in WarpOptions. BRITISH_NATIONAL_GRID_EPSG
format GDALFormatsType | str | None Format to convert input_path to in output_path. GDALNetCDFFormatStr
output_x_resolution int | None Resolution of x cordinates to convert input_path file(s) to. xRes in WarpOptions. None
output_y_resolution int | None Resolution of y cordinates to convert input_path file(s) to. yRes in WarpOptions. None
copy_metadata bool Whether to copy metadata when possible. True
return_path bool Return the resulting path if True, else the new GDALDataset. True
format GDALFormatsType | str | None Format to write new file to. GDALNetCDFFormatStr
multithread bool Whether to use multithread to speed up calculations. True
kwargs Any additional parameters to pass to WarpOption. {}

3.0.8 generate_360_to_standard

clim_recal.utils.xarray.generate_360_to_standard(array_to_expand)

Return array_to_expand 360 days expanded to 365 or 366 days.

This may be dropped if cpm_reproject_with_standard_calendar is successful.

3.0.9 get_cpm_for_coord_alignment

clim_recal.utils.xarray.get_cpm_for_coord_alignment(cpm_for_coord_alignment, skip_reproject=False, cpm_regex=CPM_REGEX)

Check if cpm_for_coord_alignment is a Dataset, process if a Path.

3.0.9.1 Parameters

Name Type Description Default
cpm_for_coord_alignment PathLike | T_Dataset | None Either a Path or a file or folder with a cpm file to align to or a xarray.Dataset. If a folder, the first file matching cpm_regex will be used. It will then be processed via cpm_reproject_with_standard_calendar for comparability and use alongside cpm files. required
skip_reproject bool Whether to skip calling cpm_reproject_with_standard_calendar. False
cpm_regex str A regular expression to filter suitable files if cpm_for_coord_alignment is a folder Path. CPM_REGEX

3.0.9.2 Returns

Type Description
An xarray.Dataset coordinate structure to align HADs coordinates.

3.0.10 hads_resample_and_reproject

clim_recal.utils.xarray.hads_resample_and_reproject(hads_xr_time_series, variable_name, cpm_to_match, cpm_to_match_func=cpm_reproject_with_standard_calendar, x_dim_name=HADS_RAW_X_COLUMN_NAME, y_dim_name=HADS_RAW_Y_COLUMN_NAME)

Resample HADs xarray time series to 2.2km.

3.0.11 interpolate_xr_ts_nans

clim_recal.utils.xarray.interpolate_xr_ts_nans(xr_ts, original_xr_ts=None, check_cftime_cols=None, interpolate_method=DEFAULT_INTERPOLATION_METHOD, keep_crs=True, keep_attrs=True, limit=1, cftime_range_gen_kwargs=None, **kwargs)

Interpolate nan values in a Dataset time series.

3.0.11.1 Notes

For details and details of keep_attrs, limit and **kwargs parameters see: https://docs.xarray.dev/en/stable/generated/xarray.DataArray.interpolate_na.html

3.0.11.2 Parameters

Name Type Description Default
xr_ts T_Dataset Dataset to interpolate via interpolate_na. Requires a time coordinate. required
original_xr_ts T_Dataset | None A Dataset to compare the conversion process with. If not provided, set to the original xr_ts as a reference. None
check_cftime_cols tuple[str] | None tuple of column names in a cftime format to check. None
interpolate_method InterpOptions Which of the xarray interpolation methods to use. DEFAULT_INTERPOLATION_METHOD
keep_crs bool Whether to ensure the original crs is kept via rio.write_crs. True
keep_attrs bool Passed to keep_attrs in interpolate_na. See Notes. True
limit int How many nan are allowed either side of data point to interpolate. See Notes. 1
cftime_range_gen_kwargs dict[str, Any] | None Any cftime_range_gen arguments to use with check_cftime_cols calls. None

3.0.11.3 Returns

Type Description
Dataset where xr_ts nan values are iterpolated with respect to the time coordinate.

3.0.12 join_xr_time_series_var

clim_recal.utils.xarray.join_xr_time_series_var(path, variable_name=None, method_name='median', time_dim_name='time', regex=CPM_REGEX, start=0, stop=None, step=1)

Join a set of xr_time_series files chronologically.

3.0.12.1 Parameters

Name Type Description Default
path PathLike Path to collect files to process from, filtered via regex. required
variable_name str | None A variable name to specify for data expected. If none that will be extracted and checked from the files directly. None
method_name str What method to use to summarise each time point results. 'median'
time_dim_name str Name of time dimension in passed files. 'time'
regex str A str to filter files within path CPM_REGEX
start int What point to start indexing path results from. 0
stop int | None What point to stop indexing path restuls from. None
step int How many paths to jump between when iterating between stop and start. 1

3.0.12.2 Examples

>>> tasmax_cpm_1980_raw = getfixture('tasmax_cpm_1980_raw')
>>> if not tasmax_cpm_1980_raw:
...     pytest.skip(mount_or_cache_doctest_skip_message)
>>> tasmax_cpm_1980_raw_path = getfixture('tasmax_cpm_1980_raw_path').parents[1]
>>> results: T_Dataset = join_xr_time_series_var(
...     tasmax_cpm_1980_raw_path,
...     'tasmax', stop=3)
>>> results
<xarray.Dataset> ...
Dimensions:  (time: 1080)
Coordinates:
  * time     (time) object ... 1980-12-01 12:00:00 ... 1983-11-30 12:00:00
Data variables:
    tasmax   (time) float64 ... 8.221 6.716 6.499 7.194 ... 8.456 8.153 5.501

3.0.13 plot_xarray

clim_recal.utils.xarray.plot_xarray(da, path=None, time_stamp=False, return_path=True, **kwargs)

Plot da with **kwargs to path.

3.0.13.1 Parameters

Name Type Description Default
da T_DataArrayOrSet xarray objec to plot. required
path PathLike | None File to write plot to. None
time_stamp bool Whather to add a datetime str of time of writing in file name. False
kwargs Additional parameters to pass to plot. {}

3.0.13.2 Examples

>>> example_path: Path = (
...     getfixture('tmp_path') / 'test-path/example.png')
>>> image_path: Path = plot_xarray(
...     xarray_spatial_4_days, example_path)
>>> example_path == image_path
True
>>> example_time_stamped: Path = (
...      example_path.parent / 'example-stamped.png')
>>> timed_image_path: Path = plot_xarray(
...     xarray_spatial_4_days, example_time_stamped,
...     time_stamp=True)
>>> example_time_stamped != timed_image_path
True
>>> print(timed_image_path)
/.../test-path/example-stamped_...-...-..._...png

3.0.14 region_crop_file_name

clim_recal.utils.xarray.region_crop_file_name(crop_region, file_name)

Generate a file name for a regional crop.

3.0.14.1 Parameters

Name Type Description Default
crop_region str | RegionOptions | None Region name to include in cropped file name. required
file_name PathLike File name to add crop_region name to. required

3.0.14.2 Examples

>>> region_crop_file_name(
...    'Glasgow',
...    'tasmax.nc')
'crop_Glasgow_tasmax.nc'
>>> region_crop_file_name(
...    'Glasgow',
...    'tasmax_hadukgrid_uk_2_2km_day_19800601-19800630.nc')
'crop_Glasgow_tasmax_hads_19800601-19800630.nc'
>>> region_crop_file_name(
...     'Glasgow',
...     'tasmax_rcp85_land-cpm_uk_2.2km_05_day_std_year_19861201-19871130.nc')
'crop_Glasgow_tasmax_cpm_05_19861201-19871130.nc'

3.0.15 xr_reproject_crs

clim_recal.utils.xarray.xr_reproject_crs(xr_time_series, x_dim_name=CPM_RAW_X_COLUMN_NAME, y_dim_name=CPM_RAW_Y_COLUMN_NAME, time_dim_name=TIME_COLUMN_NAME, variable_name=None, final_crs=BRITISH_NATIONAL_GRID_EPSG, match_xr_time_series=None, match_xr_time_series_load_func=None, match_xr_time_series_load_kwargs=None, resampling_method=DEFAULT_RESAMPLING_METHOD, nodata=np.nan, **kwargs)

Reproject source_xr to target_xr coordinate structure.

3.0.15.1 Parameters

Name Type Description Default
xr_time_series T_Dataset | PathLike Dataset or PathLike to load and reproject. required
x_dim_name str str name of x spatial dimension in xr_time_series. Default matches CPM UK projections. CPM_RAW_X_COLUMN_NAME
y_dim_name str str name of y spatial dimension in xr_time_series. Default matches CPM UK projections. CPM_RAW_Y_COLUMN_NAME
time_dim_name str str name of time dimension in xr_time_series. TIME_COLUMN_NAME
variable_name str | None Name of datset to apply projection to within xr_time_series. Inferred if None assuming only one data_var attribute. None
final_crs str Coordinate system str to project xr_time_series to. BRITISH_NATIONAL_GRID_EPSG
resampling_method Resampling rasterio resampling method to apply. DEFAULT_RESAMPLING_METHOD

3.0.15.2 Examples

>>> tasmax_hads_1980_raw = getfixture('tasmax_hads_1980_raw')
>>> if not tasmax_hads_1980_raw:
...     pytest.skip(mount_or_cache_doctest_skip_message)
>>> tasmax_hads_1980_raw.dims
FrozenMappingWarningOnValuesAccess({'time': 31,
                                    'projection_y_coordinate': 1450,
                                    'projection_x_coordinate': 900,
                                    'bnds': 2})
>>> tasmax_hads_2_2km: T_Dataset = xr_reproject_crs(
...     tasmax_hads_1980_raw,
...     variable_name="tasmax",
...     x_dim_name=HADS_RAW_X_COLUMN_NAME,
...     y_dim_name=HADS_RAW_Y_COLUMN_NAME,
...     resolution=(CPM_RESOLUTION_METERS,
...                 CPM_RESOLUTION_METERS),)
>>> tasmax_hads_2_2km.dims
FrozenMappingWarningOnValuesAccess({'x': 410,
                                    'y': 660,
                                    'time': 31})
Back to top