ICESat-2 (Ice, Cloud, and land Elevation Satellite 2)

NASA’s photon-counting lidar altimeter — 10 kHz, 6 simultaneous beams, ~3 trillion measurements collected to date. Primary mission was ice-sheet topography; the data turned out to be transformative for inland water altimetry, sea ice freeboard, forest canopy, and even ocean-floor bathymetry in clear shallow water.

What it sees

• 532 nm green laser photon-counting (not waveform) — single-photon detection enables measurements through thin clouds and over dark surfaces
• 6 beams arranged in 3 strong/weak pairs, ~3 km cross-track spacing → multiple parallel tracks per overpass
• 17 m along-track sampling per beam at the surface (effective resolution depends on aggregation)
• 91-day exact repeat orbit — same ground track returned to every 91 days; sub-cycle of 31 days for less-strict revisit

Why it matters

• Ice-sheet elevation change at unprecedented precision (cm/yr) — the canonical Antarctica + Greenland mass-balance signal that drives sea-level rise projections.
• Inland water altimetry (ATL13) — rivers, lakes, reservoirs sampled at the same precision as ice. Used to fill gaps SWOT can’t reach.
• Land + canopy (ATL08) — vegetation height + ground topography globally.
• Sea ice freeboard (ATL10) — Arctic + Antarctic sea ice thickness when paired with snow depth from passive microwave.

Where to get the data

• earthaccess Python: short_name="ATL03", ATL06, ATL08, etc.
• icepyx: the canonical Python toolkit — pip install icepyx — wraps NSIDC’s search + download + read
• NSIDC DAAC: nsidc.org/data/icesat-2/data-sets
• OpenAltimetry: web-app visualization at openaltimetry.org
• Cloud-optimized HDF5 since 2024-25 — significantly faster than legacy

What it enables

• Greenland + Antarctic ice-sheet mass balance (paired with GRACE-FO for cross-validation)
• Sea-level rise attribution (ice contribution component)
• Sea ice thickness time series (Arctic, Southern Ocean)
• River-stage measurements in remote basins (Amazon tributaries, Siberian rivers)
• Lake-level + reservoir-storage anomaly tracking
• Forest canopy height (paired with GEDI for biomass)
• Shallow bathymetry in clear water (lakes, coastal reefs)

Gotchas

• ATL03 is huge. Photon-level products are billions of returns; downloads can hang (harmony-py #122). Filter aggressively by bounding box + time + photon-quality flag before downloading.
• 6 beams are NOT equivalent. Strong vs weak beam pairs. Use the right beam for the science (strong beams for ice, all 6 for canopy density).
• 91-day repeat is exact, not arbitrary. You can wait 30-60 days for the same track again if your AOI doesn’t land on a current orbit pass. Plan timing.
• Photon counting != waveform. Geometry of returns is different from full-waveform GEDI. Don’t apply GEDI canopy-metric algorithms to ICESat-2 photons.
• icepyx is the right entry point, not raw HDF5. Saves weeks of learning curve.

Related missions

• GRACE-FO (s12 →): gravity-derived mass change — complementary cross-validation for ice-sheet mass balance.
• SWOT (s11 →): wide-swath altimetry — complements ICESat-2’s narrow-track-but-precise approach.
• GEDI (s09 →): canopy lidar (separate technology), complementary footprint.
• CryoSat-2 (ESA): radar altimeter for sea ice — complementary to ICESat-2’s lidar.

Related datasets