FAIR2Adapt Case Study 1 - Case Study Template to be reused for any new FAIR2Adapt Case Study

Context¶

Purpose¶

The goal is to experiment with how to read NorESM data for FAIR2Adapt Case Study 1.

Description¶

In this notebook, we will:

Open online NorESM data
Visualize one field

Contributions¶

Notebook¶

Anne Fouilloux, Simula Research Laboratory (Norway) (reviewer), annefou

Bibliography and other interesting resources¶

The Norwegian Earth System Model (NorESM)

Import Libraries¶

# Install xarray-healpy and dggs libraries for regridding
#!pip install git+https://github.com/IAOCEA/xarray-healpy.git git+https://github.com/xarray-contrib/xdggs.git cmcrameri s3fs shapely

import cartopy.crs as ccrs
import matplotlib as mpl
import matplotlib.pyplot as plt
import numpy as np
import s3fs  # For accessing file storage at Amazon S3
import xarray as xr  # N-dimensional arrays with dimension, coordinate and attribute labels

xr.set_options(display_expand_data=False, display_expand_attrs=False, keep_attrs=True)

<xarray.core.options.set_options at 0x109e75c10>

Open file with Xarray¶

The file is on the cloud (accessible to everyone)
The first time, opening the atmosphere ds typically takes a few minutes
Therefore, we save a local version with selected subset of variables for faster loading the nezxt times

# Define file paths
endpoint_url = "https://server-data.fair2adapt.sigma2.no"
filename_atmosphere = "s3://CS1/data/model/JRAOC20TRNRPv2_hm_sst_2010-01.nc"

# Extract file from S3
client_kwargs = {"endpoint_url": endpoint_url}
s3 = s3fs.S3FileSystem(anon=True, client_kwargs=client_kwargs)

# Opening the atmosphere ds typically takes a few minutes
ds = xr.open_dataset(s3.open(filename_atmosphere))

# Display the subset dataset
ds

Load grid¶

grid_type = "p"
grid_file = "s3://CS1/data/grid/grid.nc"
grid = xr.open_dataset(s3.open(grid_file))
grid

Assign coordinates to Xarray Dataset¶

ds = ds.assign_coords({"lon": grid[grid_type + "lon"]})
ds = ds.assign_coords({"lat": grid[grid_type + "lat"]})
ds

Quick visualization¶

var = "sst"
vmin = -3
vmax = 20
proj = ccrs.NearsidePerspective(
    central_longitude=0.0, central_latitude=80.0, satellite_height=3e6
)
fig, ax = plt.subplots(1, figsize=(8, 4.5), dpi=96, subplot_kw={"projection": proj})

p = (
    ds[var]
    .isel(time=0)
    .plot(
        ax=ax,
        transform=ccrs.PlateCarree(),
        x="lon",
        y="lat",
        add_colorbar=False,
        vmin=vmin,
        vmax=vmax,
        cmap="viridis",
        shading="auto",
        rasterized=True,
    )
)
# Add coastlines and the lat-lon grid
ax.coastlines(resolution="50m", color="black", linewidth=0.5)
ax.stock_img()
gl = ax.gridlines(ylocs=range(15, 76, 15), draw_labels=True)
gl.ylocator = mpl.ticker.FixedLocator([40, 50, 60, 70, 80])

plt.colorbar(p, fraction=0.2, shrink=0.4, label="degC")

ax.set_title("Sea Surface Temperature")

# Define file paths
endpoint_url = "https://server-data.fair2adapt.sigma2.no"
filename_atmosphere = "s3://CS1/data/model/JRAOC20TRNRPv2_hm_U236GF_2010-2018.nc"

# Extract file from S3
client_kwargs = {"endpoint_url": endpoint_url}
s3 = s3fs.S3FileSystem(anon=True, client_kwargs=client_kwargs)

# Opening the atmosphere ds typically takes a few minutes
ds = xr.open_dataset(s3.open(filename_atmosphere))

# Display the subset dataset
ds

ds.U236GF

def create_plot(ds, var, vmin, vmax, time=np.nan, sigma=True):
    proj = ccrs.NearsidePerspective(
        central_longitude=0.0, central_latitude=80.0, satellite_height=3e6
    )
    fig, ax = plt.subplots(1, figsize=(8, 4.5), dpi=96, subplot_kw={"projection": proj})
    if time is not np.nan:
        p = (
            ds[var]
            .isel(time=0)
            .plot(
                ax=ax,
                transform=ccrs.PlateCarree(),
                x="lon",
                y="lat",
                add_colorbar=False,
                vmin=vmin,
                vmax=vmax,
                cmap="viridis",
                shading="auto",
                rasterized=True,
            )
        )
    elif sigma:
        p = ds[var].plot(ax=ax, add_colorbar=False, vmin=vmin, vmax=vmax)
    else:
        p = ds[var].plot(
            ax=ax,
            transform=ccrs.PlateCarree(),
            x="lon",
            y="lat",
            add_colorbar=False,
            vmin=vmin,
            vmax=vmax,
            cmap="viridis",
            shading="auto",
            rasterized=True,
        )
    # Add coastlines and the lat-lon grid
    ax.coastlines(resolution="50m", color="black", linewidth=0.5)
    ax.stock_img()
    gl = ax.gridlines(ylocs=range(15, 76, 15), draw_labels=True)
    gl.ylocator = mpl.ticker.FixedLocator([40, 50, 60, 70, 80])
    plt.colorbar(p, fraction=0.2, shrink=0.4, label="degC")
    ax.set_title(ds[var].attrs["long_name"])

var = "U236GF"
vmin = ds[var].min().values
vmax = ds[var].max().values
print(vmin, vmax)
# create_plot(ds, var, vmin, vmax, sigma=True)

# Define file paths
endpoint_url = "https://server-data.fair2adapt.sigma2.no"
filename_atmosphere = "s3://CS1/data/model/JRAOC20TRNRPv2_hm_U236LH_2010-2018.nc"

# Extract file from S3
client_kwargs = {"endpoint_url": endpoint_url}
s3 = s3fs.S3FileSystem(anon=True, client_kwargs=client_kwargs)

# Opening the atmosphere ds typically takes a few minutes
ds = xr.open_dataset(s3.open(filename_atmosphere))

# Display the subset dataset
ds

ds.U236LH

# Define file paths
endpoint_url = "https://server-data.fair2adapt.sigma2.no"
filename_atmosphere = "s3://CS1/data/model/JRAOC20TRNRPv2_hm_U236PPG_2010-2018.nc"

# Extract file from S3
client_kwargs = {"endpoint_url": endpoint_url}
s3 = s3fs.S3FileSystem(anon=True, client_kwargs=client_kwargs)

# Opening the atmosphere ds typically takes a few minutes
ds = xr.open_dataset(s3.open(filename_atmosphere))

# Display the subset dataset
ds

ds.U236PPG

# Define file paths
endpoint_url = "https://server-data.fair2adapt.sigma2.no"
filename_atmosphere = "s3://CS1/data/model/JRAOC20TRNRPv2_hm_U236SF_2010-2018.nc"

# Extract file from S3
client_kwargs = {"endpoint_url": endpoint_url}
s3 = s3fs.S3FileSystem(anon=True, client_kwargs=client_kwargs)

# Opening the atmosphere ds typically takes a few minutes
ds = xr.open_dataset(s3.open(filename_atmosphere))

# Display the subset dataset
ds

ds.U236SF

# Define file paths
endpoint_url = "https://server-data.fair2adapt.sigma2.no"
filename_atmosphere = (
    "s3://CS1/data/model/NSSP245frc2NRP205p2v2_hm_I129se40_205001-205912.nc"
)

# Extract file from S3
client_kwargs = {"endpoint_url": endpoint_url}
s3 = s3fs.S3FileSystem(anon=True, client_kwargs=client_kwargs)

# Opening the atmosphere ds typically takes a few minutes
ds = xr.open_dataset(s3.open(filename_atmosphere))

# Display the subset dataset
ds

ds.I129se40

# Define file paths
endpoint_url = "https://server-data.fair2adapt.sigma2.no"
filename_atmosphere = (
    "s3://CS1/data/model/NSSP585frc2NRP205p2v2_hm_I129se40_205001-205912.nc"
)

# Extract file from S3
client_kwargs = {"endpoint_url": endpoint_url}
s3 = s3fs.S3FileSystem(anon=True, client_kwargs=client_kwargs)

# Opening the atmosphere ds typically takes a few minutes
ds = xr.open_dataset(s3.open(filename_atmosphere))

# Display the subset dataset
ds

ds.I129se40

Case Study Template