verde.grid_to_table

verde.grid_to_table(grid)

Convert a grid to a table with the values and coordinates of each point.

Takes a 2D grid as input, extracts the coordinates and runs them through numpy.meshgrid to create a 2D table. Works for 2D grids and any number of variables. Use cases includes passing gridded data to functions that expect data in XYZ format, such as verde.BlockReduce

Parameters:

gridxarray.Dataset or xarray.DataArray

A 2D grid with one or more data variables.

Returns:

tablepandas.DataFrame

Table with coordinates and variable values for each point in the grid. Column names are taken from the grid. If grid is a xarray.DataArray that doesn’t have a name attribute defined, the column with data values will be called "scalars".

Examples

>>> import xarray as xr
>>> import numpy as np
>>> # Create a sample grid with a single data variable
>>> temperature = xr.DataArray(
...     np.arange(20).reshape((4, 5)),
...     coords=(np.arange(4), np.arange(5, 10)),
...     dims=['northing', 'easting']
... )
>>> print(temperature.values)
[[ 0  1  2  3  4]
 [ 5  6  7  8  9]
 [10 11 12 13 14]
 [15 16 17 18 19]]
>>> # For DataArrays, the data column will be "scalars" by default
>>> table = grid_to_table(temperature)
>>> list(sorted(table.columns))
['easting', 'northing', 'scalars']
>>> print(table.scalars.values)
[ 0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19]
>>> print(table.northing.values)
[0 0 0 0 0 1 1 1 1 1 2 2 2 2 2 3 3 3 3 3]
>>> print(table.easting.values)
[5 6 7 8 9 5 6 7 8 9 5 6 7 8 9 5 6 7 8 9]
>>> # If the DataArray defines a "name", we will use that instead
>>> temperature.name = "temperature_K"
>>> table = grid_to_table(temperature)
>>> list(sorted(table.columns))
['easting', 'northing', 'temperature_K']
>>> # Conversion of Datasets will preserve the data variable names
>>> grid = xr.Dataset({"temperature": temperature})
>>> table  = grid_to_table(grid)
>>> list(sorted(table.columns))
['easting', 'northing', 'temperature']
>>> print(table.temperature.values)
[ 0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19]
>>> print(table.northing.values)
[0 0 0 0 0 1 1 1 1 1 2 2 2 2 2 3 3 3 3 3]
>>> print(table.easting.values)
[5 6 7 8 9 5 6 7 8 9 5 6 7 8 9 5 6 7 8 9]
>>> # Grids with multiple data variables will have more columns.
>>> wind_speed = xr.DataArray(
...     np.arange(20, 40).reshape((4, 5)),
...     coords=(np.arange(4), np.arange(5, 10)),
...     dims=['northing', 'easting']
... )
>>> grid['wind_speed'] = wind_speed
>>> table = grid_to_table(grid)
>>> list(sorted(table.columns))
['easting', 'northing', 'temperature', 'wind_speed']
>>> print(table.northing.values)
[0 0 0 0 0 1 1 1 1 1 2 2 2 2 2 3 3 3 3 3]
>>> print(table.easting.values)
[5 6 7 8 9 5 6 7 8 9 5 6 7 8 9 5 6 7 8 9]
>>> print(table.temperature.values)
[ 0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19]
>>> print(table.wind_speed.values)
[20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39]
>>> # Non-dimensional coordinates are also handled
>>> temperature = xr.DataArray(
...     np.arange(20).reshape((4, 5)),
...     coords=(np.arange(4), np.arange(5, 10)),
...     dims=['northing', 'easting']
... )
>>> temperature = temperature.assign_coords(
...     upward=(
...         ("northing", "easting"),
...         np.arange(20).reshape((4, 5))
...     )
... )
>>> table = grid_to_table(temperature)
>>> list(sorted(table.columns))
['easting', 'northing', 'scalars', 'upward']
>>> print(table.upward.values)
[ 0  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19]