Glaciers — GLIMS outlines & WGMS mass balance (live)¶

Beyond the RGI snapshot, the backend reaches two more open sources:

• GLIMS — time-series glacier outlines, queried by bounding box through the GLIMS WFS (a glacier can have several outlines from different dates).
• WGMS — the Fluctuations of Glaciers database: mass balance, front variation (length change), and glacier state, as tidy tables.

This notebook draws GLIMS outlines for an Alpine box and plots a glacier's measured mass balance over time.

Setup¶

In [1]:

import tempfile
from pathlib import Path

import matplotlib.pyplot as plt

from earthlens import EarthLens

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

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

GLIMS — time-series outlines by bounding box¶

Use data_source='glims' with a bounding box. max_features= caps the number of outlines returned by the WFS. The result is a FeatureCollection in EPSG:4326.

In [2]:

glims = EarthLens(
    data_source='glims',
    variables=['glims:outlines'],
    lat_lim=[45.9, 46.05],
    lon_lim=[7.0, 7.2],
    max_features=300,
    path=workdir,
).download()

len(glims), str(glims.crs)

glims = EarthLens( data_source='glims', variables=['glims:outlines'], lat_lim=[45.9, 46.05], lon_lim=[7.0, 7.2], max_features=300, path=workdir, ).download() len(glims), str(glims.crs)

2026-06-28 17:55:13 | INFO | pyramids.base.config | Logging is configured.

2026-06-28 17:55:17.290 | INFO     | earthlens.glaciers.backend:_log_citation:442 - Glaciers source citation: GLIMS and NSIDC (2005, updated current year), Global Land Ice Measurements from Space glacier database, NSIDC-0272 (CC-BY-4.0); cite the contributing analysts.

2026-06-28 17:55:17.291 | INFO     | earthlens.glaciers.backend:download:392 - Glaciers glims:outlines: returned a FeatureCollection (289 outline(s)).

Out[2]:

(289, 'EPSG:4326')

GLIMS is multi-temporal: the src_date column shows that the same glacier can appear with outlines from different acquisition dates.

In [3]:

glims[['glac_id', 'glac_name', 'src_date']].head()

glims[['glac_id', 'glac_name', 'src_date']].head()

Out[3]:

	glac_id	glac_name	src_date
11	G006984E45950N	du Tour Noir	2022-08-12 00:00:00+00:00
12	G006984E45950N	du Tour Noir	2022-08-12 00:00:00+00:00
13	G006984E45950N	du Tour Noir	2022-08-12 00:00:00+00:00
14	G006984E45950N	du Tour Noir	2022-08-12 00:00:00+00:00
15	G006984E45950N	du Tour Noir	2022-08-12 00:00:00+00:00

In [4]:

ax = glims.plot(figsize=(8, 6), facecolor='none', edgecolor='steelblue')
ax.set_title('GLIMS glacier outlines (time series) — Alpine box')
ax.set_xlabel('Longitude')
ax.set_ylabel('Latitude')
plt.tight_layout()
plt.show()

ax = glims.plot(figsize=(8, 6), facecolor='none', edgecolor='steelblue') ax.set_title('GLIMS glacier outlines (time series) — Alpine box') ax.set_xlabel('Longitude') ax.set_ylabel('Latitude') plt.tight_layout() plt.show()

WGMS — a glacier's mass-balance record¶

Now the tabular side. data_source='wgms' with variables=['wgms:mass_balance'] returns a DataFrame. The first call downloads and caches the Fluctuations of Glaciers archive (~37 MB) once. We filter to the famous Aletsch glacier by name.

In [5]:

df = EarthLens(
    data_source='wgms',
    variables=['wgms:mass_balance'],
    glacier_name='aletsch',
    path=workdir,
).download()

df[['glacier_name', 'year', 'annual_balance']].dropna().head()

df = EarthLens( data_source='wgms', variables=['wgms:mass_balance'], glacier_name='aletsch', path=workdir, ).download() df[['glacier_name', 'year', 'annual_balance']].dropna().head()

2026-06-28 17:55:22.007 | INFO     | earthlens.glaciers.backend:_log_citation:442 - Glaciers source citation: WGMS (2026), Fluctuations of Glaciers Database, DOI 10.5904/wgms-fog-2026-02-10.

2026-06-28 17:55:22.012 | INFO     | earthlens.glaciers.backend:download:400 - Glaciers wgms:mass_balance: 66 row(s) written to C:\Users\main\AppData\Local\Temp\glaciers-0ki13_ri\wgms_mass_balance.csv.

Out[5]:

	glacier_name	year	annual_balance
0	GROSSER ALETSCH	1940	0.897
1	GROSSER ALETSCH	1941	-0.919
2	GROSSER ALETSCH	1942	-1.323
3	GROSSER ALETSCH	1943	-1.203
4	GROSSER ALETSCH	1944	-1.316

The annual_balance column is the net mass balance in millimetres water equivalent — negative years mean the glacier lost mass. Let's plot the record over time for the longest-running series in the selection.

In [6]:

series = df.dropna(subset=['annual_balance']).copy()
top = series['glacier_name'].value_counts().index[0]
one = series[series['glacier_name'] == top].sort_values('year')
top, len(one)

series = df.dropna(subset=['annual_balance']).copy() top = series['glacier_name'].value_counts().index[0] one = series[series['glacier_name'] == top].sort_values('year') top, len(one)

Out[6]:

('GROSSER ALETSCH', 66)

In [7]:

fig, ax = plt.subplots(figsize=(9, 4))
colors = ['tab:red' if v < 0 else 'tab:blue' for v in one['annual_balance']]
ax.bar(one['year'], one['annual_balance'], color=colors)
ax.axhline(0, color='0.3', linewidth=0.8)
ax.set_title(f'WGMS annual mass balance — {top}')
ax.set_xlabel('Year')
ax.set_ylabel('Annual balance (mm w.e.)')
plt.tight_layout()
plt.show()

fig, ax = plt.subplots(figsize=(9, 4)) colors = ['tab:red' if v < 0 else 'tab:blue' for v in one['annual_balance']] ax.bar(one['year'], one['annual_balance'], color=colors) ax.axhline(0, color='0.3', linewidth=0.8) ax.set_title(f'WGMS annual mass balance — {top}') ax.set_xlabel('Year') ax.set_ylabel('Annual balance (mm w.e.)') plt.tight_layout() plt.show()

The predominance of red (negative) bars in recent decades is the measured signature of sustained glacier mass loss.

Takeaway¶

GLIMS gives you dated outlines for change studies; WGMS gives you the in-situ measurement record. Both come through the same EarthLens(...).download() call — the output shape follows the dataset. See the usage guide for the full list of selectors.