Fair et al. (2025) Review article: using spaceborne lidar for snow depth retrievals: recent findings and utility for hydrologic applications

Identification

• Journal: ˜The œcryosphere
• Year: 2025
• Date: 2025-11-13
• Authors: Zachary Fair, Carrie Vuyovich, T. Neumann, Justin M. Pflug, David Shean, Ellyn M. Enderlin, Karina Zikan, Hannah Besso, Jessica D. Lundquist, César Deschamps‐Berger, Désirée Treichler
• DOI: 10.5194/tc-19-5671-2025

Research Groups

• Hydrological Sciences Laboratory, NASA Goddard Space Flight Center, Greenbelt, MD, USA
• Earth System Science Interdisciplinary Center, University of Maryland, College Park, MD, USA
• Earth Sciences Division, NASA Goddard Space Flight Center, Greenbelt, MD, USA
• University of Washington, Seattle, WA, USA
• Department of Geosciences, Boise State University, Boise, ID, USA
• Pyrenean Institute of Ecology-CSIC, Zaragoza, Spain
• Department of Geosciences, University of Oslo, Oslo, Norway

Short Summary

This review synthesizes the current status of spaceborne lidar for snow depth retrieval, focusing on the ICESat-2 mission, and evaluates its utility for hydrologic applications. It concludes that while ICESat-2 can achieve centimeter-level accuracy under ideal conditions, challenges persist in complex terrain and with current temporal revisit limitations, necessitating integration with hydrologic models and improved snow-off digital elevation models.

Objective

• To review the current status of research on using spaceborne lidar for snow depth retrievals and evaluate its utility for hydrologic applications.

Study Configuration

• Spatial Scale: Regional to continental, covering key watersheds across the United States (e.g., Tuolumne River Basin, Grand Mesa, Reynolds Creek, Methow Valley), Alaskan tundra, Wolverine Glacier, Hardangervidda (Norway), Northern Xinjiang (China), and the Western United States.
• Temporal Scale: Spans the operational periods of ICESat (2002-2009), GEDI (2018-present), and ICESat-2 (2018-present), focusing on seasonal snow depth measurements and their evolution.

Methodology and Data

• Models used:
  • Differential altimetry (comparing snow-on and snow-free elevation datasets).
  • Backscatter deconvolution (exploiting time delay due to light penetration into the snowpack and ICESat-2 photon counts).
  • Cross-track differencing (using intersecting lidar tracks).
  • Data assimilation frameworks (e.g., ensemble-based data assimilation, Kalman filters/smoothers, kriging, machine learning approaches for combining lidar data with hydrologic models).
• Data sources:
  • Spaceborne Lidar: ICESat (GLAS/GLAH14), GEDI (Level-2A), ICESat-2 (ATL03, ATL06, ATL08, SlideRule).
  • Airborne Lidar: Airborne Snow Observatory (ASO), USGS 3DEP program, University of Alaska Fairbanks (UAF) airborne lidar.
  • Satellite Imagery: Worldview-3 stereo imagery, Pléiades, Copernicus DEM, Landsat, Sentinel-2, MODIS, VIIRS, Sentinel-3, AMSR-2, Sentinel-1A.
  • Reference Digital Elevation Models (DEMs)/Digital Terrain Models (DTMs): SRTM DEM, Kartverket DEM, ArcticDEM, Reference Elevation Model of Antarctica (REMA), 3-D Elevation Program (3DEP).
  • In-situ/Ground-based: Weather stations, snow pit profiles, terrestrial lidar, snow-probe transects.
  • Reanalysis Products: 4 km resolution reanalysis, 24 km resolution reanalysis, ECMWF Reanalysis v5 (ERA5), Modern-Era Retrospective analysis for Research and Applications, v2 (MERRA-2).

Main Results

• Spaceborne lidar, particularly ICESat-2, can provide high-resolution snow depth retrievals with centimeter-level accuracy (e.g., median bias of -4 cm and normalized median absolute deviation (NMAD) of +5.7 cm in Alaskan tundra) under ideal conditions (flat terrain, minimal vegetation).
• Differential altimetry is the most common and consistent method for snow depth retrieval.
• ICESat-2 snow depths generally exhibit a root mean square error (RMSE) up to 33 cm, with improved performance over slopes less than 10° (4–20 cm bias) and with customized processing (SlideRule).
• GEDI shows larger biases (RMSE of 101 cm) compared to ICESat-2, while ICESat reports an RMSE of 47 cm.
• Significant error sources include:
  • Terrain characteristics: Complex topography and steep slopes (e.g., greater than 20°) can lead to errors exceeding 1 meter.
  • DEM accuracy and co-registration: Vertical offsets, especially with coarse DEMs, can result in uncertainties greater than 3 meters. High-quality snow-off DEMs (e.g., from airborne lidar) are critical.
  • Vegetation: Dense canopies weaken lidar returns and introduce biases (e.g., -0.17 to +0.59 meters terrain biases in ATL08 over dense vegetation; increased uncertainties over greater than 60% forest cover). Undergrowth can cause meter-level bias in snow-free DEMs.
  • Lidar penetration in snow: 532 nm lidar (ICESat-2) can experience volumetric scattering, biasing surface elevations by 4–7 cm (photon level) over firn/aged snow, potentially up to 50 cm in modeling studies.
• Current spaceborne lidar missions cannot meet the 1–5 day revisit time requirement for global snow water equivalent (SWE) observations, but infrequent, high-quality observations can still significantly improve modeled SWE estimates, particularly near peak snowpack.
• The small median bias in ICESat-2 snow depth in the Tuolumne Basin suggests that even infrequent ICESat-2 data can be valuable for inferring SWE throughout the snow season.

Contributions

• Provides a comprehensive review of the current status of spaceborne lidar for snow depth retrieval, with a particular focus on the ICESat-2 mission.
• Synthesizes findings from existing literature regarding the accuracy, coverage, and operational potential of ICESat, GEDI, and ICESat-2 for snow depth measurements.
• Identifies and discusses best practices and common error sources (terrain characteristics, DEM accuracy, vegetation, lidar penetration in snow) for spaceborne lidar snow depth retrieval.
• Presents a step-by-step case study for ICESat-2 snow depth retrieval over the Alaskan tundra, demonstrating high accuracy (median bias -4 cm, NMAD +5.7 cm) under favorable conditions.
• Offers recommendations for future studies, including the adoption of consistent accuracy metrics (median bias and NMAD), and outlines requirements for future missions and the integration of lidar data with hydrologic models to improve SWE estimates.
• Highlights the critical need for an accurate, high-resolution global snow-off DEM to fully realize the potential of spaceborne lidar for mid-latitude snow depth observations.

Funding

• Earth Sciences, Division Goddard Institute for Space Studies (grant nos. 80NSSC20K1293 and NPP168273S)
• Co-operative Research grant (no. 012454)

Citation

@article{Fair2025Review,
  author = {Fair, Zachary and Vuyovich, Carrie and Neumann, T. and Pflug, Justin M. and Shean, David and Enderlin, Ellyn M. and Zikan, Karina and Besso, Hannah and Lundquist, Jessica D. and Deschamps‐Berger, César and Treichler, Désirée},
  title = {Review article: using spaceborne lidar for snow depth retrievals: recent findings and utility for hydrologic applications},
  journal = {˜The œcryosphere},
  year = {2025},
  doi = {10.5194/tc-19-5671-2025},
  url = {https://doi.org/10.5194/tc-19-5671-2025}
}