Easy Xarray Cheat Code

1. Import Xarray

import xarray as xr

2. Open a NetCDF File

ds = xr.open_dataset('file.nc')
print(ds)

3. Create a DataArray

import numpy as np
arr = xr.DataArray(np.random.rand(3, 3), dims=['x', 'y'], coords={'x': [1, 2, 3], 'y': [10, 20, 30]})
print(arr)

4. Create a Dataset

data = xr.Dataset(
{'var1': (('x', 'y'), np.random.rand(3, 3))},
coords={'x': [1, 2, 3], 'y': [10, 20, 30]}
)
print(data)

5. Select Data by Coordinates

# Select single value
value = ds.sel(lat=10, lon=20)

# Select range
subset = ds.sel(lat=slice(5, 15), lon=slice(10, 30))

6. Access Variables

var = ds['variable_name']
print(var)

7. Groupby Operations

# Group by time and calculate the mean
monthly_mean = ds.groupby('time.month').mean()

8. Resample Data

# Resample to monthly means
monthly = ds.resample(time='1M').mean()

9. Interpolate Missing Data

# Linear interpolation
filled = ds.interpolate_na(dim='time')

10. Calculate Statistics

# Calculate mean and standard deviation
mean = ds.mean(dim='time')
std = ds.std(dim='time')

11. Apply Mathematical Operations

# Multiply all values by 2
scaled = ds * 2

12. Plot Data

# Plot a single variable
ds['var1'].plot()

13. Save Dataset to NetCDF

ds.to_netcdf('output.nc')

14. Combine Datasets

# Merge two datasets
combined = xr.merge([ds1, ds2])

15. Convert to Pandas

# Convert to DataFrame
df = ds.to_dataframe()

16. Expand Dimensions

# Add a new dimension
expanded = ds.expand_dims('new_dim')

17. Drop Variables

# Drop a variable
ds = ds.drop_vars('var_to_drop')

18. Mask Data

# Mask values greater than 0.5
masked = ds.where(ds < 0.5)

19. Compute Derived Variables

# Calculate a new variable
ds['new_var'] = ds['var1'] * 2

20. Identify NaNs

# Check for NaN values
nan_mask = ds.isnull()