Base Classes

This is a reference API class listing.

class rompy.core.BaseModel(run_id='run_0001', model=None, template=None, checkout=None, settings=None, output_dir=None)

class rompy.core.BaseGrid

An object which provides an abstract representation of a grid in some geographic space

This is the base class for all Grid objects. The minimum representation of a grid are two NumPy array’s representing the vertices or nodes of some structured or unstructured grid, its bounding box and a boundary polygon. No knowledge of the grid connectivity is expected.

bbox(buffer=0.0)

Returns a bounding box for the spatial grid

This function returns a list [ll_x, ll_y, ur_x, ur_y] where ll_x, ll_y (ur_x, ur_y) are the lower left (upper right) x and y coordinates bounding box of the model domain

boundary(tolerance=0.2)

Returns a boundary polygon as a Shapely Polygon object. Sub-classes can override the private method Grid._get_boundary() but must return a Shapely polygon object.

tolerancefloat

See https://shapely.readthedocs.io/en/stable/manual.html#object.simplify

polygon : shapely.Polygon see https://shapely.readthedocs.io/en/stable/manual.html#Polygon

boundary_points(tolerance=0.2)

Convenience function to convert boundary Shapely Polygon to arrays of coordinates

tolerancefloat

Passed to Grid.boundary See https://shapely.readthedocs.io/en/stable/manual.html#object.simplify

class rompy.swan.SwanModel(run_id='run_0001', template=None, checkout=None, settings=None, output_dir=None)

class rompy.swan.SwanGrid(gridtype=None, x0=None, y0=None, rot=None, dx=None, dy=None, nx=None, ny=None, exc=None, gridfile=None)

nearby_spectra(ds_spec, dist_thres=0.05, plot=True)

Find points nearby and project to the boundary

ds_spec: xarray.Dataset

an XArray dataset of wave spectra at a number of points. Dataset variable names standardised using wavespectra.read_* functions.

See https://wavespectra.readthedocs.io/en/latest/api.html#input-functions

dist_thres: float, optional [Default: 0.05]

Maximum distance to translate the input spectra to the grid boundary

plot: boolean, optional [Default: True]

Generate a plot that shows the applied projections

xarray.Dataset

A subset of ds_spec with lat and lon coordinates projected to the boundary

class rompy.intake.NetCDFFCStackSource(*args, **kwargs)

An extension of intake-xarray in an opinionated fashion to open a stack of forecast model results stored as netcdf using xarray.

Expands urlpath with the product of values passed in parameter fmt_fields to establish a finite set of urls to stack. Uses fsspec so that the file system can be on a local or http. Useful for subsetting catalogs of netcdf files. Defers to XArray for file opening and backend driver selection.

urlpathstr

Templated path to scan for source file(s). May include format pattern strings. Some examples:

• {{ CATALOG_DIR }}/data/{year}/air_{month}.nc

• https://some.thredds.server/thredds/catalog/model/{year}/{month}/catalog.html

fn_fmtstr, optional

Templated file name to check for existence in the broadcast urlpath Example:

• air_{month}.nc

fmt_fields: dict

Dictionary of keys corresponding to urlpath format entries. The product of all values is broadcast across urlpath to establish a finite set of locations to look for source files.

url_replace:

Dictionary of string substitutions to be made on found urls. Example:

• {‘catalog’:’dodsC’}

ds_filters:

Dictionary of common dataset manipulations that are applied in order during dataset preprocessing. Configurable from catalog entry yaml. Filters currently available are [‘sort’,’subset’,’crop’,’timenorm’,’rename’] Examples:

• {‘rename’:{‘dir’:’mean_dir’},

  ‘sort’:[‘direction’], ‘timenorm’:’hour’, ‘crop’:{‘lon’:slice(-32.2,-33.2),

  ‘lat’:slice(114.,115.)}

  }

Additional documentation to follow - see source code for xarray implementation details.

startdt, enddt: datetime, optional

Start and end dates to crop final stacked dataset to.

hindcast: bool, optional

Rather than return a stack of forecasts, return a dataset with a unique time dimension selecting the minimum lead time for each time point. Useful for establishing a pseudo-reanalysis dataset from a stack of forecasts.

chunksint or dict, optional

Chunks is used to load the new dataset into dask arrays. chunks={} loads the dataset with dask using a single chunk for all arrays.

xarray_kwargs: dict

Additional xarray kwargs for xr.open_dataset().

storage_options: dict

If using a remote fs (whether caching locally or not), these are the kwargs to pass to that FS.