Unexpected Keyword Argument in Xarray

[SHsabbir25]

!pip install earthengine-api -q
!pip install xee -q
!pip install geemap -q
import ee
import geemap
import xarray as xr
import xee

!pip install -U xee xarray earthengine-api geemap

import ee
import geemap
import xarray as xr
ee.Authenticate()
ee.Initialize(project='ee-sabbirhossainshah')

import geemap
map = geemap.Map(basemap = 'SATELLITE')
map

roi = map.draw_last_feature.geometry()
roi

lst = ee.ImageCollection("MODIS/061/MOD11A2").filterDate('2020','2021').select('LST_Day_1km').mean().multiply(0.02).subtract(273.15)
ndvi = ee.ImageCollection("MODIS/061/MOD13Q1").filterDate('2020','2021').select('NDVI').mean().multiply(0.0001)
stack = ee.Image.cat([lst, ndvi]).select([0,1],['lst','ndvi'])
stack

ds1km = xr.open_dataset(stack, engine = 'ee', crs = 'EPSG:4326', scale = 0.01, geometry = roi)
ds250m = xr.open_dataset(stack, engine ='ee', crs = 'EPSG:4326', scale = 0.0025, geometry = roi)

ds1km

While running the code in google collab why shows error

TypeError Traceback (most recent call last)
/tmp/ipykernel_2657/2152062775.py in <cell line: 0>()
----> 1 ds1km = xr.open_dataset(stack, engine = 'ee', crs = 'EPSG:4326', scale = 0.01, geometry = roi)
2 ds250m = xr.open_dataset(stack, engine ='ee', crs = 'EPSG:4326', scale = 0.0025, geometry = roi)
3
4 ds1km

/usr/local/lib/python3.12/dist-packages/xarray/backends/api.py in open_dataset(filename_or_obj, engine, chunks, cache, decode_cf, mask_and_scale, decode_times, decode_timedelta, use_cftime, concat_characters, decode_coords, drop_variables, create_default_indexes, inline_array, chunked_array_type, from_array_kwargs, backend_kwargs, **kwargs)
604
605 overwrite_encoded_chunks = kwargs.pop("overwrite_encoded_chunks", None)
--> 606 backend_ds = backend.open_dataset(
607 filename_or_obj,
608 drop_variables=drop_variables,

TypeError: EarthEngineBackendEntrypoint.open_dataset() got an unexpected keyword argument 'scale' or geometry

please help me as soon as possible
Thanks in advance

[jdbcode]

Hi! It looks like you have encountered the changes introduced in Xee v0.1.0.

In this version, the library transitioned to a new grid parameter system, which is why your code is failing with an "unexpected keyword argument" error for scale and geometry. The API no longer accepts those parameters directly in xr.open_dataset.

What Changed?

• Grid Parameters: Instead of passing scale and geometry directly, you must now pass crs, crs_transform, and shape_2d.
• Helper Functions: The xee.helpers module is now the standard way to generate these parameters, ensuring your data is correctly aligned.
• Dimension Ordering: Xee now uses (time, y, x) order to match CF conventions. You likely won't need to use .transpose() anymore.

For more details, please see the Xee v0.1.0 Migration Guide.

Updated Code Snippet

Here is how you can update your workflow using the new helpers.fit_geometry function:

!pip install -q xee

import ee
import shapely
import xarray as xr
from xee import helpers  # Import the xee helpers module
import matplotlib.pyplot as plt

# 1. Authentication and Initialization
ee.Authenticate()
project_id = 'PROJECT_ID' # Replace with your project ID
ee.Initialize(project=project_id)

# 2. Define your AOI and helper
# The 'fit_geometry' helper replaces the old direct arguments
def get_grid_params(roi, grid_crs, grid_scale):
    # This generates the required 'crs', 'crs_transform', and 'shape_2d' 
    # arguments automatically for xr.open_dataset
    grid_params = helpers.fit_geometry(
        geometry=roi,
        geometry_crs='EPSG:4326',
        grid_crs=grid_crs,
        grid_scale=grid_scale # Note: Y-scale should be negative for north-up
    )
    return grid_params

# A rectangle in Washington State for testing
roi = shapely.geometry.box(-122.5, 46.5, -121.5, 47.5)

# 3. Prepare your data
lst = ee.ImageCollection("MODIS/061/MOD11A2").filterDate('2020','2021').select('LST_Day_1km').mean().multiply(0.02).subtract(273.15)
ndvi = ee.ImageCollection("MODIS/061/MOD13Q1").filterDate('2020','2021').select('NDVI').mean().multiply(0.0001)
stack = ee.Image.cat([lst, ndvi]).select([0,1],['lst','ndvi'])

# 4. Generate grid parameters using the helper
# grid_scale is (x_scale, y_scale)
grid_1km = get_grid_params(roi, 'EPSG:4326', (0.01, -0.01))
grid_250m = get_grid_params(roi, 'EPSG:4326', (0.0025, -0.0025))

# 5. Open dataset using the new engine call
# We unpack the dictionary returned by the helper
ds1km = xr.open_dataset(stack, engine='ee', **grid_1km)
ds250m = xr.open_dataset(stack, engine='ee', **grid_250m)

print(ds1km)

# 6. Plotting
# Note: No .transpose() is needed as v0.1.0 uses (time, y, x) order by default
fig, axes = plt.subplots(1, 2, figsize=(15, 6))

ds1km.lst.isel(time=0).plot(ax=axes[0])
axes[0].set_title('MODIS LST (1km Scale)')

ds250m.lst.isel(time=0).plot(ax=axes[1])
axes[1].set_title('MODIS LST (250m Scale)')

plt.tight_layout()
plt.show()