Glaciers — quickstart (live)¶

The glaciers backend fetches glacier outlines and fluctuations from three open sources and returns them in the natural shape for each: a pyramids FeatureCollection for outlines (RGI 7.0, GLIMS) and a pandas.DataFrame for fluctuations (WGMS).

This quickstart pulls RGI 7.0 glacier outlines for a small bounding box over the French Alps and draws them on a map. By the end you will know how to request outlines for an area of interest and what the result looks like.

All three sources are open — no account, no API key.

Setup¶

We need the EarthLens facade and matplotlib. The backend writes its output under path=, which we point at a throwaway temp directory. Its downloads are cached separately, under glaciers/ in the shared earthlens cache directory, so a re-run reuses them; set EARTHLENS_CACHE to relocate that.

In [ ]:

import tempfile
from pathlib import Path

import matplotlib.pyplot as plt

from earthlens.core import EarthLens

workdir = Path(tempfile.mkdtemp(prefix='glaciers-'))
workdir

import tempfile from pathlib import Path import matplotlib.pyplot as plt from earthlens.core import EarthLens workdir = Path(tempfile.mkdtemp(prefix='glaciers-')) workdir

Request RGI outlines for a bounding box¶

Pick the dataset with variables=['rgi:outlines'] and pass a bounding box as lat_lim / lon_lim. Behind the scenes the backend maps the bbox to the overlapping GTN-G region (here region 11, Central Europe), downloads that region's shapefile once, reads it with pyramids, and clips it to your box. The result is a FeatureCollection (a GeoDataFrame) in EPSG:4326.

In [ ]:

fc = EarthLens(
    data_source='glaciers',
    variables=['rgi:outlines'],
    lat_lim=[45.8, 46.05],
    lon_lim=[6.8, 7.2],  # a small box over the Mont Blanc area
    path=workdir,
).download()

type(fc).__name__, len(fc), str(fc.crs)

fc = EarthLens( data_source='glaciers', variables=['rgi:outlines'], lat_lim=[45.8, 46.05], lon_lim=[6.8, 7.2], # a small box over the Mont Blanc area path=workdir, ).download() type(fc).__name__, len(fc), str(fc.crs)

Each row is one glacier. The attribute table carries the RGI id, the glacier name, its area, and elevation statistics — let's peek at a few columns.

In [ ]:

fc[['rgi_id', 'glac_name', 'area_km2', 'zmin_m', 'zmax_m']].head()

fc[['rgi_id', 'glac_name', 'area_km2', 'zmin_m', 'zmax_m']].head()

Visualise the outlines¶

Because a FeatureCollection is a GeoDataFrame, we can draw it directly. We shade each glacier by its area so the larger ice bodies stand out.

In [ ]:

ax = fc.plot(
    column='area_km2',
    cmap='Blues',
    edgecolor='0.4',
    legend=True,
    figsize=(8, 6),
)
ax.set_title('RGI 7.0 glacier outlines — Mont Blanc area')
ax.set_xlabel('Longitude')
ax.set_ylabel('Latitude')
plt.tight_layout()
plt.show()

ax = fc.plot( column='area_km2', cmap='Blues', edgecolor='0.4', legend=True, figsize=(8, 6), ) ax.set_title('RGI 7.0 glacier outlines — Mont Blanc area') ax.set_xlabel('Longitude') ax.set_ylabel('Latitude') plt.tight_layout() plt.show()

The polygons are the real RGI 7.0 glacier outlines clipped to our box — the large blue body near the centre is the Mer de Glace / Mont Blanc glacier complex. The total glacierised area in view:

In [ ]:

round(float(fc['area_km2'].sum()), 1)  # km^2 of ice in the bbox

round(float(fc['area_km2'].sum()), 1) # km^2 of ice in the bbox

Takeaway¶

One EarthLens(...).download() call returns ready-to-map glacier outlines clipped to your area of interest. Next:

• Catalog & behaviour — the datasets, the GTN-G regions, the per-instance output shape, and why aggregate= is rejected (no network).
• GLIMS & WGMS — time-series outlines and the mass-balance record.