Different grids issue when running a regression

[Murk89]

I am trying to run a regression with MOD13Q1 NDVI and my WRF climate model outputs. For this I need to assure that the MODIS raster and WRF maps are on the same projection and spatial resolution. So I transformed MODIS'S projection and resolution based on my WRF maps specs using Earth Engine and then converted the file to a netCDF. Using python's xarrays, I can see that the x and y dims of the MOD13Q1 dataset and corresponding WRF's south_north and west_east dims are quite different. Can someone advise on how to align these different datasets/maps in python to be able to run a regression at every grid box? Many thanks.

WRF map: (Time: 1, south_north: 111, west_east: 114)>
()

MOD13Q1: Dimensions:(x: 1378, y: 1203)

[andhuang-CLGX]

If they have both lat / lon coordinates:
xesmf
https://pangeo-xesmf.readthedocs.io/en/latest/notebooks/Rectilinear_grid.html

If not, reproject using rioxarray
https://corteva.github.io/rioxarray/stable/rioxarray.html?highlight=project#rioxarray.raster_array.RasterArray.reproject_match
cartopy.crs.transform_points or pyproj

then
interp_like
https://xarray.pydata.org/en/stable/generated/xarray.DataArray.interp_like.html

[Murk89]

Just to update
After trying to define the crs of the WRF dataset, which is explained here: https://spatialreference.org/ref/sr-org/wrf-lambert-conformal-conic/

The revised code is:
#WRF datset
wrf=xr.open_dataset("wrfout_d02_2007-12-01_00_00_00.nc", decode_coords="all" )
wrf = wrf.rename({'XTIME':'Time'})
wrf = wrf.assign_coords(Time=pd.to_datetime(wrf['Time'].values))

wrf.dims
#returns:
Dimensions: (Time: 1, south_north: 111, west_east: 114)

#renaming the west_east and south_north dims of wrf to y and x. Because otherwise they are not recognized.
wrf = wrf.rename({'west_east':'y'})
wrf = wrf.rename({'south_north':'x'})

#writing the crs, using proj4
crs = CRS.from_string ('+proj=lcc +lat_1=64 +lat_2=15 +lat_0=64 +lon_0=10 +x_0=0 +y_0=0 +ellps=WGS84 +datum=WGS84 +units=m +no_defs')
wrf.T2.rio.write_crs("crs", inplace=True)

#MODIS dataset
ndvi=xr.open_dataset("mean_ndvi.nc", decode_coords="all")
ndvi.dims
#returns:
Frozen({'lat': 5949, 'lon': 12720})

#rename lat lon as y and x
ndvi=ndvi.rename({'lon':'x'})
ndvi=ndvi.rename({'lat':'y'})

#crs of the ndvi dataset
ndvi['Band1'].rio.write_crs("epsg:4326", inplace=True)

trial=ndvi.Band1.rio.reproject_match(wrf.T2)
trial.plot()

the above code produces the following plot:

And I actually need it to look like this to be able to do a regression at each x,y

Can anyone please help in getting this right. I seem to be stuck. Many thanks.

[snowman2]

You need to remove the string quotes around "crs" here: wrf.T2.rio.write_crs("crs", inplace=True)

[Murk89]

Hi,

@snowman2 removing the string quotes generates this plot:

[snowman2]

Mind creating a discussion about this here: https://github.com/corteva/rioxarray/discussions ? Also, providing input datasets are very helpful for debugging.

[Murk89]

Yes why not. I'll create a new post. Hope this can be resolved. I tried uploading the datasets, however .nc files were not supported. Is there a way around it? Thanks.