Module gdess.operations.geographic

Functions

def closest(data: list, v: dict) -> dict

Find closest point in data to the v point

Parameters

data : list

 

v : dict

 

Returns

dict

Expand source code

def closest(data: list,
            v: dict) -> dict:
    """Find closest point in `data` to the v point

    Parameters
    ----------
    data: `list`
    v: `dict`

    Returns
    -------
    `dict`
    """
    min_entry = min(data, key=lambda p: distance(v['lat'], v['lon'], p['lat'], p['lon']))

    return min_entry

def distance(lat1: float, lon1: float, lat2: float, lon2: float) -> float

Expand source code

def distance(lat1: float,
             lon1: float,
             lat2: float,
             lon2: float) -> float:
    p = 0.017453292519943295
    a = 0.5 - cos((lat2 - lat1) * p) / 2 + cos(lat1 * p) * cos(lat2 * p) * (1 - cos((lon2 - lon1) * p)) / 2
    return 12742 * asin(sqrt(a))

def get_closest_mdl_cell_dict(dataset: xarray.core.dataset.Dataset, lat: float, lon: float, coords_as_dimensions: bool = True) -> dict

Find the point in the model output that is closest to specified lat/lon pair

Examples

To get the data subset at a location: – For CMIP outputs – >>> closest_point = get_closest_mdl_cell_dict(dataset, lat=24.3, lon=137.8, coords_as_dimensions=True) >>> dataset.stack(coord_pair=['lat', 'lon']).isel(coord_pair=closest_point['index']) – For E3SM native grid – >>> closest_mdl_point = get_closest_mdl_cell_dict(dataset, lat=lat, lon=lon, coords_as_dimensions=False) >>> dataset.where(dataset['ncol'] == closest_mdl_point['index'], drop=True)

Parameters

dataset : xarray.Dataset

 

lat : float

 

lon : float

 

coords_as_dimensions

True for dataset variables as independent dimensions, e.g. lat = -90, -89,…,0, 1,… ,90 lon = -180, -179, -178,…,0, 1,… ,180 False for dataset variables where all pairs are enumerated, e.g. lat = -90, -90,… ,-89, -89,… 90, 90 lon = -180, -179,… ,-180, -179,… 179, 180

Returns

dict With lat, lon, and index in Dataset For example, {'lat': 19.5, 'lon': 204.375, 'index': 31555}

Expand source code

def get_closest_mdl_cell_dict(dataset: xr.Dataset,
                              lat: float,
                              lon: float,
                              coords_as_dimensions: bool = True
                              ) -> dict:
    """Find the point in the model output that is closest to specified lat/lon pair

    Examples
    --------
    To get the data subset at a location:
    -- For CMIP outputs --
        >>> closest_point = get_closest_mdl_cell_dict(dataset, lat=24.3, lon=137.8, coords_as_dimensions=True)
        >>> dataset.stack(coord_pair=['lat', 'lon']).isel(coord_pair=closest_point['index'])
    -- For E3SM native grid --
        >>> closest_mdl_point = get_closest_mdl_cell_dict(dataset, lat=lat, lon=lon, coords_as_dimensions=False)
        >>> dataset.where(dataset['ncol'] == closest_mdl_point['index'], drop=True)

    Parameters
    ----------
    dataset : ``xarray.Dataset``
    lat : `float`
    lon : `float`
    coords_as_dimensions
        True for dataset variables as independent dimensions, e.g.
            lat = -90, -89,...,0, 1,... ,90
            lon = -180, -179, -178,...,0, 1,... ,180
        False for dataset variables where all pairs are enumerated, e.g.
            lat = -90,  -90,...  ,-89,  -89,... 90, 90
            lon = -180, -179,... ,-180, -179,... 179, 180

    Returns
    -------
    `dict`
        With lat, lon, and index in Dataset
        For example, {'lat': 19.5, 'lon': 204.375, 'index': 31555}
    """
    obs_station_lat_lon = {'lat': lat, 'lon': lon}

    if coords_as_dimensions:
        coords = dataset.stack(coord_pair=['lat', 'lon']).coord_pair.values
    else:
        coords = zip(dataset['lat'].values, dataset['lon'].values)

    mdl_lat_lon_list = [{'lat': a, 'lon': o, 'index': i}
                        for i, (a, o)
                        in enumerate(coords)]

    # Find it.
    closest_dict = closest(mdl_lat_lon_list, obs_station_lat_lon)

    return closest_dict