plots.py

"""
This module contains functions for plotting rainfall rate data using Cartopy and Matplotlib.
It includes utilities for color mapping, coordinate transformations, and plotting.
"""

from pathlib import Path
from typing import Tuple

import cartopy.feature as cfeature
import matplotlib.colors as mcolors
import matplotlib.pyplot as plt
import numpy as np
import xarray as xr
from cartopy.crs import Globe, PlateCarree, Stereographic
from matplotlib.axes import Axes
from pyproj import CRS, Transformer
from scipy.interpolate import griddata
from scipy.spatial import cKDTree

########################################################################################
#                           PROJECTIONS AND COORDINATES                                #
########################################################################################

# Original radar projection
PROJ_WKT = """
PROJCS["unknown",GEOGCS["unknown",DATUM["unknown",SPHEROID["unknown",6378137,298.252840776245]],
PRIMEM["Greenwich",0],UNIT["degree",0.0174532925199433,AUTHORITY["EPSG","9122"]]],
PROJECTION["Polar_Stereographic"],PARAMETER["latitude_of_origin",45],
PARAMETER["central_meridian",0],PARAMETER["false_easting",0],PARAMETER["false_northing",0],
UNIT["metre",1],AXIS["Easting",SOUTH],AXIS["Northing",SOUTH]]
"""
GEOTRANSFORM = (
    -619652.0953618084,
    1000.0,
    0.0,
    -3526818.459196719,
    0.0,
    -999.9999999999997,
)


def project_to_latlon(arr: np.ndarray) -> xr.DataArray:
    """Convert a 2D array from the original projection to lat/lon coordinates."""
    x0, dx, _, y0, _, dy = GEOTRANSFORM
    height, width = arr.shape

    # Create meshgrid of coordinates
    x_coords = x0 + np.arange(width) * dx
    y_coords = y0 + np.arange(height) * dy
    xx, yy = np.meshgrid(x_coords, y_coords)

    # Transform grid coords to lat/lon
    crs_src = CRS.from_wkt(PROJ_WKT)
    crs_dst = CRS.from_epsg(4326)  # WGS84
    to_latlon = Transformer.from_crs(crs_src, crs_dst, always_xy=True)
    lon, lat = to_latlon.transform(xx, yy)

    # Creation of the source DataArray
    da_src = xr.DataArray(arr, dims=("y", "x"), coords={"x": x_coords, "y": y_coords})
    da_src = da_src.assign_coords(lon=(("y", "x"), lon), lat=(("y", "x"), lat))

    # Regular grid in lat/lon
    res_deg = 0.01  # ~1 km
    lat_target = np.arange(lat.min(), lat.max(), res_deg)
    lon_target = np.arange(lon.min(), lon.max(), res_deg)
    lon_grid, lat_grid = np.meshgrid(lon_target, lat_target)

    # Interpolation with griddata
    points = np.column_stack((lon.ravel(), lat.ravel()))
    values = arr.ravel()
    data_interp = griddata(points, values, (lon_grid, lat_grid), method="nearest")

    # The nearest neighbor interpolation can create artefacts on the edges
    # so we mask values using a maximum distance
    tree = cKDTree(points)
    distances, _ = tree.query(
        np.column_stack((lon_grid.ravel(), lat_grid.ravel())), k=1
    )

    # Max radius: diagonal of a target pixel
    max_dist = np.sqrt(2) * res_deg
    mask = distances > max_dist

    # Mask the interpolated data
    data_interp_flat = data_interp.ravel()
    data_interp_flat[mask] = np.nan
    data_interp = data_interp_flat.reshape(lon_grid.shape)

    # Create the final DataArray with the reprojected data
    da_reproj = xr.DataArray(
        data_interp,
        dims=("lat", "lon"),
        coords={"lat": lat_target, "lon": lon_target},
        name="data",
    )
    # Invert latitude axis to match the original orientation
    da_reproj = da_reproj[::-1, :]
    return da_reproj


########################################################################################
#                           COLORS AND COLORMAPS                                       #
########################################################################################


def hex_to_rgb(hex):
    """Converts a hexadecimal color to RGB."""
    return tuple(int(hex[i : i + 2], 16) / 255 for i in (0, 2, 4))


COLORS_RR = [  # 14 colors
    hex_to_rgb("E5E5E5"),
    hex_to_rgb("6600CBFF"),
    hex_to_rgb("0000FFFF"),
    hex_to_rgb("00B2FFFF"),
    hex_to_rgb("00FFFFFF"),
    hex_to_rgb("0EDCD2FF"),
    hex_to_rgb("1CB8A5FF"),
    hex_to_rgb("6BA530FF"),
    hex_to_rgb("FFFF00FF"),
    hex_to_rgb("FFD800FF"),
    hex_to_rgb("FFA500FF"),
    hex_to_rgb("FF0000FF"),
    hex_to_rgb("991407FF"),
    hex_to_rgb("FF00FFFF"),
]
"""list of str: list of colors for the rainfall rate colormap"""

CMAP_RR = mcolors.ListedColormap(COLORS_RR)
"""ListedColormap : rainfall rate colormap"""

BOUNDARIES_RR = [
    0,
    0.1,
    0.4,
    0.6,
    1.2,
    2.1,
    3.6,
    6.5,
    12,
    21,
    36,
    65,
    120,
    205,
    360,
]
"""list of float: boundaries of the rainfall rate colormap"""

NORM_RR = mcolors.BoundaryNorm(BOUNDARIES_RR, CMAP_RR.N, clip=True)
"""BoundaryNorm: norm for the reflectivity colormap"""

########################################################################################
#                           PLOTTING FUNCTIONS                                         #
########################################################################################


def plot_ax_rainfall_rate(
    ax: Axes,
    data: np.ndarray,
    extent: Tuple[float],
    cmap=CMAP_RR,
    norm=NORM_RR,
    title: str = "",
):
    """Plot a rainfall rate image on a given axis."""
    img = ax.imshow(data, extent=extent, cmap=cmap, norm=norm, interpolation="none")
    states_provinces = cfeature.NaturalEarthFeature(
        category="cultural",
        name="admin_1_states_provinces_lines",
        scale="10m",
        facecolor="none",
    )
    ax.add_feature(states_provinces, edgecolor="lightgrey", linewidth=0.5)
    ax.add_feature(cfeature.BORDERS.with_scale("10m"), edgecolor="black", linewidth=1)
    ax.coastlines(resolution="10m", color="black", linewidth=1)
    ax.set_title(title, fontsize=15)
    ax.gridlines(
        crs=PlateCarree(),
        draw_labels=True,
        linewidth=0.4,
        color="lightgrey",
        linestyle=":",
    )
    return img


def plot_map_rain(data: xr.DataArray, title: str, path: Path) -> None:
    """Plot a rainfall rate map."""
    projection = PlateCarree()
    extent = [data.lon.min(), data.lon.max(), data.lat.min(), data.lat.max()]
    fig, ax = plt.subplots(subplot_kw={"projection": projection}, figsize=(10, 7))
    img = plot_ax_rainfall_rate(ax, data.values, title=title, extent=extent)
    cb = fig.colorbar(img, ax=ax, orientation="horizontal", fraction=0.04, pad=0.05)
    cb.set_label(label="Precipitation in mm/h", fontsize=12)
    plt.tight_layout()
    plt.savefig(path)
    plt.close()