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

grid (xarray.Dataset or xarray.DataArray) – A 2D grid with one or more data variables.

Returns

table (pandas.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]