GHSL — built-up surface¶

Download GHS-BUILT-S 2020 at 100 m over the Lisbon AOI: the built-up surface (m² per cell).

Setup¶

The imports for the whole notebook. pyramids provides Dataset (reading the GeoTIFF); earthlens provides the unified EarthLens entry point. tempfile gives us a throwaway output directory for the download.

In [ ]:

import tempfile

import matplotlib.pyplot as plt
import numpy as np
from pyramids.dataset import Dataset

from earthlens import EarthLens

import tempfile import matplotlib.pyplot as plt import numpy as np from pyramids.dataset import Dataset from earthlens import EarthLens

Download GHS-BUILT-S¶

We build the request first — source ghsl, the GHS_BUILT_S variable, the 2020 epoch, and a bounding box over Lisbon. path points at a temporary directory so the downloaded GeoTIFF lands somewhere disposable.

In [ ]:

out = tempfile.mkdtemp()
builtup = EarthLens(
    data_source='ghsl',
    variables=['GHS_BUILT_S'],
    start='2020-01-01',
    end='2020-12-31',
    aoi=[-9.20, 38.68, -9.10, 38.78],
    path=out,
)

out = tempfile.mkdtemp() builtup = EarthLens( data_source='ghsl', variables=['GHS_BUILT_S'], start='2020-01-01', end='2020-12-31', aoi=[-9.20, 38.68, -9.10, 38.78], path=out, )

GHSL is a file-writing backend, so download() returns the list of written output paths. The download step is kept separate from construction so each is easy to read and re-run.

In [ ]:

paths = builtup.download(progress_bar=False)
paths

paths = builtup.download(progress_bar=False) paths

Read and clean the raster¶

Read the first (only) written file into a Dataset, pull out its array, and mask the GHSL nodata fill (negative values) to NaN so it plots cleanly.

In [ ]:

arr = np.asarray(Dataset.read_file(str(paths[0])).read_array(), dtype='float32')
arr = arr[0] if arr.ndim == 3 else arr
arr = np.where(arr < 0, np.nan, arr)

arr = np.asarray(Dataset.read_file(str(paths[0])).read_array(), dtype='float32') arr = arr[0] if arr.ndim == 3 else arr arr = np.where(arr < 0, np.nan, arr)

Plot the built-up surface¶

Render the masked array: each cell holds the built-up surface in m², so denser urban cores show up brightest on the cividis colour map.

In [ ]:

plt.figure(figsize=(6, 5))
plt.imshow(arr, cmap='cividis')
plt.colorbar(label='built m2 / cell')
plt.title('GHS-BUILT-S 2020 (100 m) — Lisbon')
plt.axis('off')
plt.show()

plt.figure(figsize=(6, 5)) plt.imshow(arr, cmap='cividis') plt.colorbar(label='built m2 / cell') plt.title('GHS-BUILT-S 2020 (100 m) — Lisbon') plt.axis('off') plt.show()