Working with xarray

For gridded analysis and plotting, load layers directly into an xarray.Dataset with load_layer. This requires the optional xarray extra:

pip install "pyo-oracle[xarray]"

Load a subset as xarray

import pyo_oracle as pyo

ds_id = "thetao_baseline_2000_2019_depthsurf"
constraints = pyo.build_constraints(
    ds_id,
    latitude=(-60, 60),
    longitude=(-180, 180),
    latitude_step=4,
    longitude_step=4,
)

ds = pyo.load_layer(ds_id, constraints=constraints, variables=["thetao_mean"], fmt="xarray")
ds

ds is a standard xarray.Dataset with time, latitude and longitude coordinates, so the whole xarray ecosystem is available.

Quick map

import matplotlib.pyplot as plt

ds["thetao_mean"].isel(time=0).plot(figsize=(10, 5), cmap="thermal")
plt.title("Mean sea surface temperature (baseline)")
plt.show()

For publication-quality maps with coastlines and projections, combine with cartopy:

import cartopy.crs as ccrs

ax = plt.axes(projection=ccrs.Robinson())
ds["thetao_mean"].isel(time=0).plot(
    ax=ax, transform=ccrs.PlateCarree(), cmap="thermal"
)
ax.coastlines()
plt.show()

pandas instead

If you prefer tabular data (e.g. to join with occurrence records for species distribution modelling), use the default fmt="pandas":

df = pyo.load_layer(ds_id, constraints=constraints, variables=["thetao_mean"])
df.head()