pyresample.geometry module

Classes for geometry operations.

class pyresample.geometry.AreaDefinition(area_id, description, proj_id, projection, width, height, area_extent, nprocs=1, lons=None, lats=None, dtype=<class 'numpy.float64'>)

Bases: _ProjectionDefinition

Holds definition of an area.

Parameters:

  • area_id (str) – Identifier for the area

  • description (str) – Human-readable description of the area

  • proj_id (str) – ID of projection

  • projection (dict or str or pyproj.crs.CRS) – Dictionary of PROJ parameters or string of PROJ or WKT parameters. Can also be a pyproj.crs.CRS object.

  • width (int) – x dimension in number of pixels, aka number of grid columns

  • height (int) – y dimension in number of pixels, aka number of grid rows

  • area_extent (list) – Area extent as a list (lower_left_x, lower_left_y, upper_right_x, upper_right_y)

  • nprocs (int, optional) – Number of processor cores to be used for certain calculations

area_id

Identifier for the area

Type:

str

description

Human-readable description of the area

Type:

str

proj_id

ID of projection

Type:

str

projection

Dictionary or string with Proj.4 parameters

Type:

dict or str

width

x dimension in number of pixels, aka number of grid columns

Type:

int

height

y dimension in number of pixels, aka number of grid rows

Type:

int

size

Number of points in grid

Type:

int

area_extent_ll

Area extent in lons lats as a tuple (lower_left_lon, lower_left_lat, upper_right_lon, upper_right_lat)

Type:

tuple

pixel_size_x

Pixel width in projection units

Type:

float

pixel_size_y

Pixel height in projection units

Type:

float

upper_left_extent

Coordinates (x, y) of upper left corner of upper left pixel in projection units

Type:

tuple

pixel_upper_left

Coordinates (x, y) of center of upper left pixel in projection units

Type:

tuple

pixel_offset_x

x offset between projection center and upper left corner of upper left pixel in units of pixels.

Type:

float

pixel_offset_y

y offset between projection center and upper left corner of upper left pixel in units of pixels..

Type:

float

crs

Coordinate reference system object similar to the PROJ parameters in proj_dict and proj_str. This is the preferred attribute to use when working with the pyproj library. Note, however, that this object is not thread-safe and should not be passed between threads.

Type:

pyproj.crs.CRS

crs_wkt

WellKnownText version of the CRS object. This is the preferred way of describing CRS information as a string.

Type:

str

cartesian_coords

Grid cartesian coordinates

Type:

object

Initialize AreaDefinition.

__init__(area_id, description, proj_id, projection, width, height, area_extent, nprocs=1, lons=None, lats=None, dtype=<class 'numpy.float64'>)

Initialize AreaDefinition.

aggregate(**dims)

Return an aggregated version of the area.

property area_extent

Tuple of this area’s extent (xmin, ymin, xmax, ymax).

colrow2lonlat(cols, rows)

Return lons and lats for the given image columns and rows.

Both scalars and arrays are supported. To be used with scarse data points instead of slices (see get_lonlats).

copy(**override_kwargs)

Make a copy of the current area.

This replaces the current values with anything in override_kwargs.

create_areas_def()

Generate YAML formatted representation of this area.

Deprecated. Use dump() instead.

create_areas_def_legacy()

Create area definition in legacy format.

crop_around(other_area)

Crop this area around other_area.

dump(filename=None)

Generate YAML formatted representation of this area.

For the opposite (i.e. to get an AreaDefinition from a YAML-formatted representation), see load_area_from_string().

Parameters:

filename (str or pathlib.Path or file-like object) – Yaml file location to dump the area to.

Returns:

If file is None returns yaml str

classmethod from_area_of_interest(area_id, projection, shape, center, resolution, units=None, **kwargs)

Create an AreaDefinition from center, resolution, and shape.

Parameters:

  • area_id (str) – ID of area

  • projection (dict or str) – Projection parameters as a proj4_dict or proj4_string

  • shape (list) – Number of pixels in the y and x direction (height, width)

  • center (list) – Center of projection (x, y)

  • resolution (list or float) – Size of pixels: (dx, dy). Can be specified with one value if dx == dy

  • units (str, optional) –

    Units that provided arguments should be interpreted as. This can be one of ‘deg’, ‘degrees’, ‘meters’, ‘metres’, and any parameter supported by the cs2cs -lu command. Units are determined in the following priority:

    1. units expressed with each variable through a DataArray’s attrs attribute.

    2. units passed to units

    3. units used in projection

    4. meters

  • description (str, optional) – Description/name of area. Defaults to area_id

  • proj_id (str, optional) – ID of projection

  • nprocs (int, optional) – Number of processor cores to be used

  • lons (numpy array, optional) – Grid lons

  • lats (numpy array, optional) – Grid lats

Returns:

AreaDefinition

Return type:

AreaDefinition

classmethod from_cf(cf_file, variable=None, y=None, x=None)

Create an AreaDefinition object from a netCDF/CF file.

Parameters:

  • nc_file (string or object) – path to a netCDF/CF file, or opened xarray.Dataset object

  • variable (string, optional) – name of the variable to load the AreaDefinition from If variable is None the file will be searched for valid CF area definitions

  • y (string, optional) – name of the variable to use as ‘y’ axis of the CF area definition If y is None an appropriate ‘y’ axis will be deduced from the CF file

  • x (string, optional) – name of the variable to use as ‘x’ axis of the CF area definition If x is None an appropriate ‘x’ axis will be deduced from the CF file

Returns:

AreaDefinition

Return type:

AreaDefinition

classmethod from_circle(area_id, projection, center, radius, shape=None, resolution=None, units=None, **kwargs)

Create an AreaDefinition from center, radius, and shape or from center, radius, and resolution.

Parameters:

  • area_id (str) – ID of area

  • projection (dict or str) – Projection parameters as a proj4_dict or proj4_string

  • center (list) – Center of projection (x, y)

  • radius (list or float) – Length from the center to the edges of the projection (dx, dy)

  • shape (list, optional) – Number of pixels in the y and x direction (height, width)

  • resolution (list or float, optional) – Size of pixels: (dx, dy)

  • units (str, optional) –

    Units that provided arguments should be interpreted as. This can be one of ‘deg’, ‘degrees’, ‘meters’, ‘metres’, and any parameter supported by the cs2cs -lu command. Units are determined in the following priority:

    1. units expressed with each variable through a DataArray’s attrs attribute.

    2. units passed to units

    3. units used in projection

    4. meters

  • description (str, optional) – Description/name of area. Defaults to area_id

  • proj_id (str, optional) – ID of projection

  • nprocs (int, optional) – Number of processor cores to be used

  • lons (numpy array, optional) – Grid lons

  • lats (numpy array, optional) – Grid lats

  • optimize_projection – Whether the projection parameters have to be optimized for a DynamicAreaDefinition.

Returns:

AreaDefinition or DynamicAreaDefinition – If shape or resolution are provided, an AreaDefinition object is returned. Else a DynamicAreaDefinition object is returned

Return type:

AreaDefinition or DynamicAreaDefinition

Notes

  • resolution and radius can be specified with one value if dx == dy

classmethod from_epsg(code, resolution)

Create an AreaDefinition object from an epsg code (string or int) and a resolution.

classmethod from_extent(area_id, projection, shape, area_extent, units=None, **kwargs)

Create an AreaDefinition object from area_extent and shape.

Parameters:

  • area_id (str) – ID of area

  • projection (dict or str) – Projection parameters as a proj4_dict or proj4_string

  • shape (list) – Number of pixels in the y and x direction (height, width)

  • area_extent (list) – Area extent as a list (lower_left_x, lower_left_y, upper_right_x, upper_right_y)

  • units (str, optional) –

    Units that provided arguments should be interpreted as. This can be one of ‘deg’, ‘degrees’, ‘meters’, ‘metres’, and any parameter supported by the cs2cs -lu command. Units are determined in the following priority:

    1. units expressed with each variable through a DataArray’s attrs attribute.

    2. units passed to units

    3. units used in projection

    4. meters

  • description (str, optional) – Description/name of area. Defaults to area_id

  • proj_id (str, optional) – ID of projection

  • nprocs (int, optional) – Number of processor cores to be used

  • lons (numpy array, optional) – Grid lons

  • lats (numpy array, optional) – Grid lats

Returns:

AreaDefinition

Return type:

AreaDefinition

classmethod from_ul_corner(area_id, projection, shape, upper_left_extent, resolution, units=None, **kwargs)

Create an AreaDefinition object from upper_left_extent, resolution, and shape.

Parameters:

  • area_id (str) – ID of area

  • projection (dict or str) – Projection parameters as a proj4_dict or proj4_string

  • shape (list) – Number of pixels in the y and x direction (height, width)

  • upper_left_extent (list) – Upper left corner of upper left pixel (x, y)

  • resolution (list or float) – Size of pixels in meters: (dx, dy). Can be specified with one value if dx == dy

  • units (str, optional) –

    Units that provided arguments should be interpreted as. This can be one of ‘deg’, ‘degrees’, ‘meters’, ‘metres’, and any parameter supported by the cs2cs -lu command. Units are determined in the following priority:

    1. units expressed with each variable through a DataArray’s attrs attribute.

    2. units passed to units

    3. units used in projection

    4. meters

  • description (str, optional) – Description/name of area. Defaults to area_id

  • proj_id (str, optional) – ID of projection

  • nprocs (int, optional) – Number of processor cores to be used

  • lons (numpy array, optional) – Grid lons

  • lats (numpy array, optional) – Grid lats

Returns:

AreaDefinition

Return type:

AreaDefinition

geocentric_resolution(ellps='WGS84', radius=None)

Find best estimate for overall geocentric resolution.

This method is extremely important to the results of KDTree-based resamplers like the nearest neighbor resampling. This is used to determine how far the KDTree should be queried for valid pixels before giving up (radius_of_influence). This method attempts to make a best guess at what geocentric resolution (the units used by the KDTree) represents the majority of an area.

To do this this method will:

  1. Create a vertical mid-line and a horizontal mid-line.

  2. Convert these coordinates to geocentric coordinates.

  3. Compute the distance between points along these lines.

  4. Take the histogram of each set of distances and find the bin with the most points.

  5. Take the average of the edges of that bin.

  6. Return the maximum of the vertical and horizontal bin edge averages.

get_area_slices(area_to_cover, shape_divisible_by=None)

Compute the slice to read based on an area_to_cover.

get_array_coordinates_from_lonlat(lon, lat)

Retrieve the array coordinates (float) for a given lon/lat.

If lon,lat is a tuple of sequences of longitudes and latitudes, a tuple of arrays is returned.

Parameters:

  • lon (array_like) – point or sequence of longitudes

  • lat (array_like) – point or sequence of latitudes

Returns:

the array coordinates (cols/rows)

Return type:

floats or arrays of floats

get_array_coordinates_from_projection_coordinates(xm, ym)

Find the floating-point grid cell index for a specified projection coordinate.

If xm, ym is a tuple of sequences of projection coordinates, a tuple of arrays are returned.

Parameters:

  • xm (array_like) – point or sequence of x-coordinates in meters (map projection)

  • ym (array_like) – point or sequence of y-coordinates in meters (map projection)

Returns:

the array coordinates (cols/rows)

Return type:

floats or arrays of floats

get_array_indices_from_lonlat(lon, lat)

Find the closest integer grid cell index for a given lon/lat.

If lon,lat is a point, a ValueError is raised if it is outside the area domain. If lon,lat is a tuple of sequences of longitudes and latitudes, a tuple of masked arrays are returned. The masked values are the actual row and col indexing the grid cell if the area had been big enough, or the numpy default (999999) if invalid.

Parameters:

  • lon (array_like) – point or sequence of longitudes

  • lat (array_like) – point or sequence of latitudes

Returns:

the array indices (cols/rows)

Return type:

ints or masked arrays of ints

Raises:

ValueError – if the return point is outside the area domain

get_array_indices_from_projection_coordinates(xm, ym)

Find the closest integer grid cell index for a specified projection coordinate.

If xm, ym is a point, a ValueError is raised if it is outside the area domain. If xm, ym is a tuple of sequences of projection coordinates, a tuple of masked arrays are returned.

Parameters:

  • xm (array_like) – point or sequence of x-coordinates in meters (map projection)

  • ym (array_like) – point or sequence of y-coordinates in meters (map projection)

Returns:

the array indices (cols/rows)

Return type:

ints or masked arrays of ints

Raises:

ValueError – if the return point is outside the area domain

get_edge_bbox_in_projection_coordinates(vertices_per_side=None, frequency=None)

Return the bounding box in projection coordinates.

get_lonlat(row, col)

Retrieve lon and lat values of single point in area grid.

Parameters:

  • row (int)

  • col (int)

Returns:

(lon, lat)

Return type:

tuple of floats

get_lonlat_from_array_coordinates(cols, rows)

Get the longitude and latitude from (floating) column and row indices.

If cols, rows is a tuple of sequences of array coordinates, a tuple of arrays is returned.

Parameters:

  • cols (array_like) – the column coordinates

  • rows (array_like) – the row coordinates

Returns:

the longitude, latitude in degrees

Return type:

floats or arrays of floats

get_lonlat_from_projection_coordinates(xm, ym)

Get the lonlat from projection coordinates.

If xm, ym is a tuple of sequences of projection coordinates, a tuple of arrays is returned.

Parameters:

  • xm (array_like) – the x projection coordinates in meters

  • ym (array_like) – the y projection coordinates in meters

Returns:

the longitude, latitude in degrees

Return type:

floats or arrays of floats

get_lonlats(nprocs=None, data_slice=None, cache=False, dtype=None, chunks=None)

Return lon and lat arrays of area.

Note that this historically this method always returns longitude/latitudes on the geodetic (unprojected) model of the Earth used by the Coordinate Reference System (CRS) for this area. However, this is not true for shifted datums. For example, a projection including a PROJ.4 parameter like +pm=180 to shift longitudes/latitudes 180 degrees, will return degrees on the +pm=0 equivalent of the geodetic CRS.

Parameters:

  • nprocs (int, optional) – Number of processor cores to be used. Defaults to the nprocs set when instantiating object

  • data_slice (slice object, optional) – Calculate only coordinates for specified slice

  • cache (bool, optional) – Store the result internally for later reuse. Requires data_slice to be None.

  • dtype (numpy.dtype, optional) – Data type of the returned arrays

  • chunks (int or tuple, optional) – Create dask arrays and use this chunk size

Returns:

(lons, lats) – Grids of area lons and and lats

Return type:

tuple of numpy arrays

get_lonlats_dask(chunks=None, dtype=None)

Get longitudes and latitudes.

get_proj_coords(data_slice=None, dtype=None, chunks=None)

Get projection coordinates of grid.

Parameters:

  • data_slice (slice object, optional) – Calculate only coordinates for specified slice

  • dtype (numpy.dtype, optional) – Data type of the returned arrays

  • chunks (int or tuple, optional) – Create dask arrays and use this chunk size

Returns:

  • (target_x, target_y) (tuple of numpy arrays) – Grids of area x- and y-coordinates in projection units

  • .. versionchanged:: 1.11.0 – Removed ‘cache’ keyword argument and add ‘chunks’ for creating dask arrays.

get_proj_coords_dask(chunks=None, dtype=None)

Get projection coordinates.

get_proj_vectors(dtype=None, chunks=None)

Calculate 1D projection coordinates for the X and Y dimension.

Parameters:

  • dtype (numpy.dtype) – Numpy data type for the returned arrays

  • chunks (int or tuple) – Return dask arrays with the chunk size specified. If this is a tuple then the first element is the Y array’s chunk size and the second is the X array’s chunk size.

Returns:

  • tuple ((X, Y) where X and Y are 1-dimensional numpy arrays)

  • The data type of the returned arrays can be controlled with the

  • dtype keyword argument. If chunks is provided then dask arrays

  • are returned instead.

get_proj_vectors_dask(chunks=None, dtype=None)

Get projection vectors.

get_projection_coordinates_from_array_coordinates(cols, rows)

Get the projection coordinate from the array coordinates.

If cols, rows is a tuple of sequences of array coordinates, a tuple of arrays is returned.

Parameters:

  • cols (array_like) – the column coordinates

  • rows (array_like) – the row coordinates

Returns:

the projection coordinates x, y in meters

Return type:

floats or arrays of floats

get_projection_coordinates_from_lonlat(lon, lat)

Get the projection coordinate from longitudes and latitudes.

If lon,lat is a tuple of sequences of longitudes and latitudes, a tuple of arrays is returned.

Parameters:

  • lon (array_like) – point or sequence of longitudes

  • lat (array_like) – point or sequence of latitudes

Returns:

the projection coordinates x, y in meters

Return type:

floats or arrays of floats

get_xy_from_lonlat(lon, lat)

Retrieve closest x and y coordinates.

Retrieve the closest x and y coordinates (column, row indices) for the specified geolocation (lon,lat) if inside area. If lon,lat is a point a ValueError is raised if the return point is outside the area domain. If lon,lat is a tuple of sequences of longitudes and latitudes, a tuple of masked arrays are returned.

Parameters:

  • lon – point or sequence (list or array) of longitudes

  • lat – point or sequence (list or array) of latitudes

Returns:

tuple of points/arrays

Return type:

(x, y)

get_xy_from_proj_coords(xm, ym)

Find closest grid cell index for a specified projection coordinate.

If xm, ym is a tuple of sequences of projection coordinates, a tuple of masked arrays are returned.

Parameters:

  • xm (list or array) – point or sequence of x-coordinates in meters (map projection)

  • ym (list or array) – point or sequence of y-coordinates in meters (map projection)

Returns:

column and row grid cell indexes as 2 scalars or arrays

Return type:

x, y

Raises:

ValueError – if the return point is outside the area domain

property is_geostationary

Whether this area is in a geostationary satellite projection or not.

lonlat2colrow(lons, lats)

Return image columns and rows for the given lons and lats.

Both scalars and arrays are supported. Same as get_xy_from_lonlat, renamed for convenience.

property name

Return area name.

property outer_boundary_corners

Return the lon,lat of the outer edges of the corner points.

property proj4_string

Return projection definition as Proj.4 string.

property proj_str

Return PROJ projection string.

This is no longer the preferred way of describing CRS information. Switch to the crs or crs_wkt properties for the most flexibility.

property projection_x_coords

Return projection X coordinates.

property projection_y_coords

Return projection Y coordinates.

property resolution

Return area resolution in X and Y direction.

to_cartopy_crs()

Convert projection to cartopy CRS object.

to_odc_geobox()

Convert AreaDefinition to ODC GeoBox.

See: https://odc-geo.readthedocs.io/en/latest/

update_hash(existing_hash=None)

Update a hash, or return a new one if needed.

class pyresample.geometry.BaseDefinition(lons=None, lats=None, nprocs=1)

Bases: object

Base class for geometry definitions.

Changed in version 1.8.0: BaseDefinition no longer checks the validity of the provided longitude and latitude coordinates to improve performance. Longitude arrays are expected to be between -180 and 180 degrees, latitude -90 to 90 degrees. Use check_and_wrap() to preprocess your arrays.

Initialize BaseDefinition.

__init__(lons=None, lats=None, nprocs=1)

Initialize BaseDefinition.

boundary(vertices_per_side=None, force_clockwise=False, frequency=None)

Retrieve the AreaBoundary object.

Parameters:

  • vertices_per_side – (formerly frequency) The number of points to provide for each side. By default (None) the full width and height will be provided.

  • force_clockwise – Perform minimal checks and reordering of coordinates to ensure that the returned coordinates follow a clockwise direction. This is important for compatibility with pyresample.spherical.SphPolygon where operations depend on knowing the inside versus the outside of a polygon. These operations assume that coordinates are clockwise. Default is False.

property corners

Return the corners centroids of the current area.

get_area()

Get the area of the convex area defined by the corners of the curren area.

get_area_extent_for_subset(row_LR, col_LR, row_UL, col_UL)

Calculate extent for a subdomain of this area.

Rows are counted from upper left to lower left and columns are counted from upper left to upper right.

Parameters:

  • row_LR (int) – row of the lower right pixel

  • col_LR (int) – col of the lower right pixel

  • row_UL (int) – row of the upper left pixel

  • col_UL (int) – col of the upper left pixel

Returns:

Area extent (LL_x, LL_y, UR_x, UR_y) of the subset

Return type:

area_extent (tuple)

Author:

Ulrich Hamann

get_area_slices(area_to_cover)

Compute the slice to read based on an area_to_cover.

get_bbox_lonlats(vertices_per_side=None, force_clockwise=True, frequency=None)

Return the bounding box lons and lats sides.

Parameters:

  • vertices_per_side (int | None) – The number of points to provide for each side. By default (None) the full width and height will be provided.

  • frequency (int | None) – Deprecated, use vertices_per_side

  • force_clockwise (bool) – Perform minimal checks and reordering of coordinates to ensure that the returned coordinates follow a clockwise direction. This is important for compatibility with pyresample.spherical.SphPolygon where operations depend on knowing the inside versus the outside of a polygon. These operations assume that coordinates are clockwise. Default is True.

Return type:

tuple

Returns:

Two lists of four elements each. The first list is longitude coordinates, the second latitude. Each element is a numpy array representing a specific side of the geometry. The order of the arrays is first row (index 0), last column, last row, and first column. The arrays are sliced (ordered) in a way to ensure that the coordinates follow a clockwise path. In the usual case this results in the coordinates starting in the north-west corner. In the case where the data is oriented with the first pixel (row 0, column 0) in the south-east corner, the coordinates will start in that corner. Other orientations that are detected to follow a counter-clockwise path will be reordered to provide a clockwise path in order to be compatible with other parts of pyresample (ex. pyresample.spherical.SphPolygon).

get_boundary_lonlats()

Return Boundary objects.

get_cartesian_coords(nprocs=None, data_slice=None, cache=False)

Retrieve cartesian coordinates of geometry definition.

Parameters:

  • nprocs (int, optional) – Number of processor cores to be used. Defaults to the nprocs set when instantiating object

  • data_slice (slice object, optional) – Calculate only cartesian coordnates for the defined slice

  • cache (bool, optional) – Store result the result. Requires data_slice to be None

Returns:

cartesian_coords

Return type:

numpy array

get_edge_lonlats(vertices_per_side=None, frequency=None)

Get the concatenated boundary of the current swath.

get_lonlat(row, col)

Retrieve lon and lat of single pixel.

Parameters:

  • row (int)

  • col (int)

Returns:

(lon, lat)

Return type:

tuple of floats

get_lonlats(data_slice=None, chunks=None, **kwargs)

Get longitude and latitude arrays representing this geometry.

Returns:

(lon, lat) – If chunks is provided then the arrays will be dask arrays with the provided chunk size. If chunks is not provided then the returned arrays are the same as the internal data types of this geometry object (numpy or dask).

Return type:

tuple of numpy arrays

get_lonlats_dask(chunks=None)

Get the lon lats as a single dask array.

intersection(other)

Return the corners of the intersection polygon of the current area with other.

Parameters:

other (object) – Instance of subclass of BaseDefinition

Returns:

(corner1, corner2, corner3, corner4)

Return type:

tuple of points

property is_geostationary

Whether this geometry is in a geostationary satellite projection or not.

overlap_rate(other)

Get how much the current area overlaps an other area.

Parameters:

other (object) – Instance of subclass of BaseDefinition

Returns:

overlap_rate

Return type:

float

overlaps(other)

Test if the current area overlaps the other area.

This is based solely on the corners of areas, assuming the boundaries to be great circles.

Parameters:

other (object) – Instance of subclass of BaseDefinition

Returns:

overlaps

Return type:

bool

update_hash(existing_hash=None)

Update the hash.

class pyresample.geometry.CoordinateDefinition(lons, lats, nprocs=1)

Bases: BaseDefinition

Base class for geometry definitions defined by lons and lats only.

Initialize CoordinateDefinition.

__init__(lons, lats, nprocs=1)

Initialize CoordinateDefinition.

append(other)

Append another coordinate definition to existing one.

concatenate(other)

Concatenate coordinate definitions.

geocentric_resolution(ellps='WGS84', radius=None, nadir_factor=2)

Calculate maximum geocentric pixel resolution.

If lons is a xarray.DataArray object with a resolution attribute, this will be used instead of loading the longitude and latitude data. In this case the resolution attribute is assumed to mean the nadir resolution of a swath and will be multiplied by the nadir_factor to adjust for increases in the spatial resolution towards the limb of the swath.

Parameters:

  • ellps (str) – PROJ Ellipsoid for the Cartographic projection used as the target geocentric coordinate reference system. Default: ‘WGS84’. Ignored if radius is provided.

  • radius (float) – Spherical radius of the Earth to use instead of the definitions in ellps.

  • nadir_factor (int) – Number to multiply the nadir resolution attribute by to reflect pixel size on the limb of the swath.

Returns: Estimated maximum pixel size in meters on a geocentric

coordinate system (X, Y, Z) representing the Earth.

Raises: RuntimeError if a simple search for valid longitude/latitude

data points found no valid data points.

exception pyresample.geometry.DimensionError

Bases: ValueError

Wrap ValueError.

class pyresample.geometry.DynamicAreaDefinition(area_id=None, description=None, projection=None, width=None, height=None, area_extent=None, resolution=None, optimize_projection=False)

Bases: object

An AreaDefintion containing just a subset of the needed parameters.

The purpose of this class is to be able to adapt the area extent and shape of the area to a given set of longitudes and latitudes, such that e.g. polar satellite granules can be resampled optimally to a given projection.

Note that if the provided projection is geographic (lon/lat degrees) and the provided longitude and latitude data crosses the anti-meridian (-180/180), the resulting area will be the smallest possible in order to contain that data and avoid a large area spanning from -180 to 180 longitude. This means the resulting AreaDefinition will have a right-most X extent greater than 180 degrees. This does not apply to data crossing the north or south pole as there is no “smallest” area in this case.

area_id

The name of the area.

description

The description of the area.

projection

The dictionary or string or CRS object of projection parameters. Doesn’t have to be complete. If not complete, proj_info must be provided to freeze to “fill in” any missing parameters.

width

x dimension in number of pixels, aka number of grid columns

height

y dimension in number of pixels, aka number of grid rows

shape

Corresponding array shape as (height, width)

area_extent

The area extent of the area.

resolution

Resolution of the resulting area as (pixel_size_x, pixel_size_y) or a scalar if pixel_size_x == pixel_size_y.

optimize_projection

Whether the projection parameters have to be optimized.

Initialize the DynamicAreaDefinition.

__init__(area_id=None, description=None, projection=None, width=None, height=None, area_extent=None, resolution=None, optimize_projection=False)

Initialize the DynamicAreaDefinition.

compute_domain(corners, resolution=None, shape=None, projection=None)

Compute shape and area_extent from corners and [shape or resolution] info.

Parameters:

  • corners (Sequence) – 4-element sequence representing the outer corners of the region. Note that corners represents the center of pixels, while area_extent represents the edge of pixels. The four values are (xmin_corner, ymin_corner, xmax_corner, ymax_corner). If the x corners are None then the full extent (area of use) of the projection will be used. When needed, area of use is taken from the PROJ library or in the case of a geographic lon/lat projection -180/180 is used. A RuntimeError is raised if the area of use is needed (when x corners are None) and area of use can’t be determined.

  • resolution (float | tuple[float, float] | None) – Spatial resolution in projection units (typically meters or degrees). If not specified then shape must be provided. If a scalar then it is treated as the x and y resolution. If a tuple then x resolution is the first element, y is the second.

  • shape (tuple[int, int] | None) – Number of pixels in the area as a 2-element tuple. The first is number of rows, the second number of columns.

  • projection (CRS | dict | str | int | None) – PROJ.4 definition string, dictionary, integer EPSG code, or pyproj CRS object.

Note that shape is (rows, columns) and resolution is (x_size, y_size); the dimensions are flipped.

freeze(lonslats=None, resolution=None, shape=None, proj_info=None, antimeridian_mode=None)

Create an AreaDefinition from this area with help of some extra info.

Parameters:

  • lonlats (SwathDefinition or tuple) – The geographical coordinates to contain in the resulting area. A tuple should be (lons, lats). If a SwathDefinition is provided, and it has a “bounding_box” attribute, it will be used instead of the full longitude and latitude to avoid potentially slow computations.

  • resolution – the resolution of the resulting area.

  • shape – the shape of the resulting area.

  • proj_info – complementing parameters to the projection info.

  • antimeridian_mode –

    How to handle lon/lat data crossing the anti-meridian of the projection. This currently only affects lon/lat geographic projections and data cases not covering the north or south pole. The possible options are:

    • ”modify_extents”: Set the X bounds to the edges of the data, but

      add 360 to the right-most bound. This has the effect of making the area coordinates continuous from the left side to the right side. However, this means that some coordinates will be outside the coordinate space of the projection. Although most PROJ and pyresample functionality can handle this there may be some edge cases.

    • ”modify_crs”: Change the prime meridian of the projection

      from 0 degrees longitude to 180 degrees longitude. This has the effect of putting the data on a continuous coordinate system. However, this means that comparing data resampled to this resulting area and an area not over the anti-meridian would be more difficult.

    • ”global_extents”: Ignore the bounds of the data and use -180/180

      degrees as the west and east bounds of the data. This will generate a large output area, but with the benefit of keeping the data on the original projection. Note that some resampling methods may produce artifacts when resampling on the edge of the area (the anti-meridian).

  • created (Shape parameters are ignored if the instance is)

  • True. (with the optimize_projection flag set to)

property pixel_size_x

Pixel width in projection units.

property pixel_size_y

Pixel height in projection units.

class pyresample.geometry.GridDefinition(lons, lats, nprocs=1)

Bases: CoordinateDefinition

Grid defined by lons and lats.

Parameters:

  • lons (numpy array)

  • lats (numpy array)

  • nprocs (int, optional) – Number of processor cores to be used for calculations.

shape

Grid shape as (rows, cols)

Type:

tuple

size

Number of elements in grid

Type:

int

lons

Grid lons

Type:

object

lats

Grid lats

Type:

object

cartesian_coords

Grid cartesian coordinates

Type:

object

Initialize GridDefinition.

__init__(lons, lats, nprocs=1)

Initialize GridDefinition.

exception pyresample.geometry.IncompatibleAreas

Bases: ValueError

Error when the areas to combine are not compatible.

exception pyresample.geometry.InvalidArea

Bases: ValueError

Error to be raised when an area is invalid for a given purpose.

class pyresample.geometry.StackedAreaDefinition(*definitions, **kwargs)

Bases: _ProjectionDefinition

Definition based on muliple vertically stacked AreaDefinitions.

Initialize StackedAreaDefinition based on definitions.

kwargs used here are nprocs and dtype (see AreaDefinition).

__init__(*definitions, **kwargs)

Initialize StackedAreaDefinition based on definitions.

kwargs used here are nprocs and dtype (see AreaDefinition).

append(definition)

Append another definition to the area.

get_lonlats(nprocs=None, data_slice=None, cache=False, dtype=None, chunks=None)

Return lon and lat arrays of the area.

get_lonlats_dask(chunks=None, dtype=None)

Return lon and lat dask arrays of the area.

property height

Return height of the area definition.

property proj4_string

Return projection definition as Proj.4 string.

property proj_str

Return projection definition as Proj.4 string.

squeeze()

Generate a single AreaDefinition if possible.

update_hash(the_hash=None)

Update the hash.

property width

Return width of the area definition.

class pyresample.geometry.SwathDefinition(lons, lats, nprocs=1, crs=None)

Bases: CoordinateDefinition

Swath defined by lons and lats.

Parameters:

  • lons (numpy array)

  • lats (numpy array)

  • nprocs (int, optional) – Number of processor cores to be used for calculations.

  • crs (pyproj.CRS,) – The CRS to use. longlat on WGS84 by default.

shape

Swath shape

Type:

tuple

size

Number of elements in swath

Type:

int

ndims

Swath dimensions

Type:

int

lons

Swath lons

Type:

object

lats

Swath lats

Type:

object

cartesian_coords

Swath cartesian coordinates

Type:

object

Initialize SwathDefinition.

__init__(lons, lats, nprocs=1, crs=None)

Initialize SwathDefinition.

aggregate(**dims)

Aggregate the current swath definition by averaging.

For example, averaging over 2x2 windows: sd.aggregate(x=2, y=2)

compute_bb_proj_params(proj_dict)

Compute BB projection parameters.

compute_optimal_bb_area(proj_dict=None, resolution=None)

Compute the “best” bounding box area for this swath with proj_dict.

By default, the projection is Oblique Mercator (omerc in proj.4), in which case the right projection angle alpha is computed from the swath centerline. For other projections, only the appropriate center of projection and area extents are computed.

The height and width are computed so that the resolution is approximately the same across dimensions.

copy()

Copy the current swath.

pyresample.geometry.combine_area_extents_vertical(area1, area2)

Combine the area extents of areas 1 and 2.

pyresample.geometry.concatenate_area_defs(area1, area2, axis=0)

Append area2 to area1 and return the results.

pyresample.geometry.daskify_2in_2out(func)

Daskify the coordinate conversion functions.

pyresample.geometry.enclose_areas(*areas, area_id='joint-area')

Return the smallest areadefinition enclosing one or more others.

From one or more AreaDefinition objects (most usefully at least two), which shall differ only in extent, calculate the smallest AreaDefinition that encloses all. Touches only the area_extent; projection and units must be identical in all input areas and will be unchanged in the resulting area. When the input areas \(i=1..n\) have extent \((a_i, b_i, c_i, d_i)\), the resulting area will have extent \((\\min_i{a_i}, \\min_i{b_i}, \\max_i{c_i}, \\max_i{d_i})\).

Parameters:

  • *areas (AreaDefinition) – AreaDefinition objects to enclose.

  • area_id (Optional[str]) – Name of joint area, defaults to “joint-area”.

pyresample.geometry.get_array_hashable(arr)

Compute a hashable form of the array arr.

Works with numpy arrays, dask.array.Array, and xarray.DataArray.

pyresample.geometry.get_full_geostationary_bounding_box_in_proj_coords(geos_area, nb_points=50)

Get the valid boundary geos projection coordinates of the full disk.

Parameters:

  • geos_area – Geostationary area definition to get the bounding box for.

  • nb_points – Number of points on the polygon

pyresample.geometry.get_geostationary_angle_extent(geos_area)

Get the max earth (vs space) viewing angles in x and y.

pyresample.geometry.get_geostationary_bounding_box(geos_area, nb_points=50)

Get the bbox in lon/lats of the valid pixels inside geos_area.

Parameters:

  • geos_area – Geostationary area definition to get the bounding box for.

  • nb_points – Number of points on the polygon

pyresample.geometry.get_geostationary_bounding_box_in_lonlats(geos_area, nb_points=50)

Get the bbox in lon/lats of the valid pixels inside geos_area.

Parameters:

  • geos_area – Geostationary area definition to get the bounding box for.

  • nb_points – Number of points on the polygon

pyresample.geometry.get_geostationary_bounding_box_in_proj_coords(geos_area, nb_points=50)

Get the bbox in geos projection coordinates of the valid pixels inside geos_area.

Parameters:

  • geos_area – Geostationary area definition to get the bounding box for.

  • nb_points – Number of points on the polygon.

pyresample.geometry.masked_ints(func)

Return masked integer arrays when returning array indices.

pyresample.geometry.ordered_dump(data, stream=None, Dumper=<class 'yaml.dumper.Dumper'>, **kwds)

Dump the data to YAML in ordered fashion.

pyresample.geometry.preserve_scalars(func)

Preserve scalars through the coordinate conversion functions.