Interpolation

class kadlu.geospatial.interpolation.BaseGeospatialInterpolator(value, name='GeospatialInterpolator', origin=None, **coors)[source]

Bases: object

Parent class for all interpolators.

Child classes must implement the _eval method.

Interpolators accept the following coordinates,

lat: latitude in degrees

lon: longitudes in degrees

depth: depth in meters below the sea surface

epoch: time in hours since 2000-01-01 00:00:00

Args:

value: array-like: Data values
lat: array-like: Latitudes in degrees.
lon: array-like: Longitudes in degrees.
depth: array-like: Depths in meters below the sea surface.
epoch: array-like: Time in hours since 2000-01-01 00:00:00.
name: str: Name used to identify the interpolator.
origin: tuple(float,float): Reference location used as the origo of XY coordinate system. If not specified, the center point of the lat,lon coordinates will be used.

Attrs:

value: numpy array: Data values
coordinates: dict: Coordinates arrays

class kadlu.geospatial.interpolation.IrregularGridGeospatialInterpolator(value, name='IrregularGridGeospatialInterpolator', origin=None, method='cubic', **coors)[source]

Bases: BaseGeospatialInterpolator

Attrs:

dims: list(str): Names of the dimensions that are being interpolated

property method

class kadlu.geospatial.interpolation.RegularGridGeospatialInterpolator(value, name='RegularGridGeospatialInterpolator', origin=None, method=None, **coors)[source]

Bases: BaseGeospatialInterpolator

Args:

method: str: For interpolation of data on a 2d lat-lon grid, it is recommended to leave method unspecified (i.e. method=None)

property method

kadlu.geospatial.interpolation.get_interpolator(value, **kwargs)[source]

Convenience function for initialising a interpolator for a given set of data.

Args:

value: array-like: Data values

Keyword args:

lat: array-like: Latitudes in degrees.
lon: array-like: Longitudes in degrees.
depth: array-like: Depths in meters below the sea surface.
epoch: array-like: Time in hours since 2000-01-01 00:00:00.
origin: tuple(float,float): Reference location used as the origo of XY coordinate system. If not specified, the center point of the lat,lon coordinates will be used.
method: str: Preferred interpolation method. Allowed values are: nearest, linear, slinear, cubic. For high-dimensional data with large numbers of nodes (> 1E6), it is recommended to use one of the two simpler interpolation methods (nearest or linear) to reduce memory usage. For data that require interpolation in latitude and longitude only (e.g. bathymetry) the default interpolation method is a bivariate spline method adapted specifically to spherical coordinates with proper handling of the poles and anti-meridian discontinuity.
name: str: Name used to identify the interpolator. Optional.
max_size: int: Maximum size (no. bins) along any of the axes of the regular interpolation grid. Optional. Only relevant if input data is on an irregular grid.
grid_shape: dict(): Shape of the regular interpolation grid. Optional. Only relevant if input data is on an irregular grid.
bin_size: dict(): Bin sizes of the regular interpolation grid. Optional. Overwrites grid_shape. Only relevant if input data is on an irregular grid.
irreg_method: str: Interpolation method used for mapping data from an irregular to a regular grid. Options are nearest and linear. Default is nearest.

Returns:

interp: RegularGridGeospatialInterpolator: Interpolator