He et al. (2025) Benchmarking and evaluating the NASA Land Information System (version 7.5.2) coupled with the refactored Noah-MP land surface model (version 5.0)

Identification

• Journal: Geoscientific model development
• Year: 2025
• Date: 2025-11-12
• Authors: Tzu‐Shun Lin, David M. Mocko, Ronnie Abolafia‐Rosenzweig, Jerry Wegiel, Sujay V. Kumar
• DOI: 10.5194/gmd-18-8439-2025

Research Groups

• NSF National Center for Atmospheric Research (NCAR), Boulder, Colorado, USA
• NASA Goddard Space Flight Center, Greenbelt, Maryland, USA
• Science Applications International Corporation, Greenbelt, Maryland, USA

Short Summary

This study integrates the refactored Noah-MP version 5.0 land surface model with the NASA Land Information System (LIS) version 7.5.2 to enhance interoperability and evaluates its global and regional performance against the previous LIS/Noah-MPv4.0.1 for key land surface variables. The results show that LIS/Noah-MPv5.0 generally performs similarly or better than its predecessor, with slight degradations in simulated surface soil moisture and snow water equivalent.

Objective

• To integrate the refactored community Noah-MP version 5.0 model with the NASA Land Information System (LIS) version 7.5.2 to streamline synchronization, development, and maintenance of Noah-MP within LIS, thereby enhancing their interoperability and applicability.
• To evaluate and compare 5-year (2018–2022) global and regional benchmark simulations of LIS/Noah-MPv5.0 and LIS/Noah-MPv4.0.1 for a set of key land surface variables.

Study Configuration

• Spatial Scale: Global (approximately 10 km resolution) and regional over the contiguous U.S. (CONUS) (0.125° resolution).
• Temporal Scale: 5-year evaluation period (2018–2022) following a 5-year spin-up (regional simulations started 2013, global simulations started 2018). Hourly atmospheric forcing data.

Methodology and Data

• Models used:
  • NASA Land Information System (LIS) version 7.5.2
  • Noah-MP version 5.0 (refactored community version)
  • Noah-MP version 4.0.1 (previous LIS version)
• Data sources:
  • Atmospheric Forcing:
    • Global: U.S. Air Force (USAF) atmospheric forcing reanalysis (hourly, approximately 10 km).
    • CONUS: North American Land Data Assimilation System (NLDAS-2) atmospheric forcing data (hourly, 0.125°).
  • Static Land Parameters:
    • Land type map: Moderate Resolution Imaging Spectroradiometer (MODIS) satellite data.
    • Leaf Area Index (LAI) and Stem Area Index (SAI): MODIS monthly climatological (2000–2008).
    • Soil type map: State Soil Geographic (STATSGO)/Food and Agriculture Organization (FAO) soil database.
  • Reference Data for Evaluation:
    • Surface soil moisture: Soil Moisture Active Passive (SMAP) version 8 Level 3 satellite data (daily, 36 km); International Soil Moisture Network (ISMN) ground station hourly measurements.
    • Latent heat flux (LH): Global Land Evaporation Amsterdam Model (GLEAMv3.8a) reanalysis data (daily, 0.25°); FLUXCOM-X-BASE observation-based data (hourly, 0.05°).
    • Snow Water Equivalent (SWE) and snow depth: NOAA National Weather Service’s National Operational Hydrologic Remote Sensing Center (NOHRSC) Snow Data Assimilation System (SNODAS) data (daily, 1 km); ERA5-Land reanalysis data (daily, 0.1°).
    • Snow cover fraction: MODIS Terra Snow Cover version 6 data (daily, 500 m).
    • Surface albedo: MODIS Terra/Aqua merged data (daily, 0.05°).

Main Results

• LIS/Noah-MPv5.0 generally outperforms or is similar to LIS/Noah-MPv4.0.1 in simulating key land surface variables, with slight degradation in simulated surface soil moisture and snow water equivalent (SWE).
• Both LIS/Noah-MPv4.0.1 and LIS/Noah-MPv5.0 simulations effectively capture the spatial and seasonal distributions of observed soil moisture, latent heat (LH), SWE, snow depth, snow cover, and surface albedo, exhibiting similar bias patterns.
• Soil Moisture: Both models tend to underestimate soil moisture over wet soil regimes and overestimate over dry soil regimes. LIS/Noah-MPv5.0 shows slightly higher soil moisture (global mean bias of 0.008 m³ m⁻³ vs. 0.003 m³ m⁻³ for v4.0.1 compared to SMAP) across most regions.
• Latent Heat Flux: Model bias patterns generally follow those of soil moisture. LIS/Noah-MPv5.0 exhibits lower LH across many non-polar regions, which reduces the global mean LH bias from 0.99 W m⁻² (LIS/Noah-MPv4.0.1) to -0.39 W m⁻² (LIS/Noah-MPv5.0).
• Snow Water Equivalent (SWE): Model SWE bias patterns are primarily influenced by precipitation and temperature forcing uncertainties. LIS/Noah-MPv5.0 shows slightly lower SWE values (global mean bias of -13.2 mm) compared to LIS/Noah-MPv4.0.1 (global mean bias of -10.1 mm).
• Snow Depth: Model bias patterns for snow depth generally align with those of SWE, with a global annual mean bias of approximately 0.06 m for both simulations.
• Snow Cover Fraction: LIS/Noah-MPv4.0.1 consistently overestimates snow cover globally (mean bias of 0.11). LIS/Noah-MPv5.0, with updated snow cover parameters, effectively reduces these overestimates across global snowpacks, lowering the mean bias to 0.07.
• Surface Albedo: Both models show widespread overestimates of surface albedo over mid-latitude and high-latitude regions, but significant underestimates in the Sahara Desert and Antarctica. LIS/Noah-MPv5.0 demonstrates an overall reduction in surface albedo across mid- and high-latitudes due to lower snow cover, degrading the global mean bias from -0.018 to -0.033.

Contributions

• Successfully integrated the refactored community Noah-MPv5.0 with the NASA LISv7.5.2 using a GitHub submodule mechanism, significantly streamlining code synchronization, development, and maintenance, and enhancing interoperability.
• Provided a comprehensive, systematic benchmarking and evaluation of the new LIS/Noah-MPv5.0 against the previous LIS/Noah-MPv4.0.1 across global and regional scales for a suite of key land surface variables.
• Quantified the performance differences between the two model versions and identified specific model deficiencies (e.g., in plant hydraulics, snow physics, and background albedo), motivating future targeted improvements in coupled canopy-snowpack-soil processes and input data.
• Contributed to establishing a "scorecard" type of practice for Land Surface Models (LSMs), offering a valuable reference for model development and assessment.

Funding

• NASA grant no. 80NSSC24K0121
• National Science Foundation (NSF) (for supporting the NSF National Center for Atmospheric Research (NCAR))

Citation

@article{He2025Benchmarking,
  author = {He, Cenlin and Lin, Tzu‐Shun and Mocko, David M. and Abolafia‐Rosenzweig, Ronnie and Wegiel, Jerry and Kumar, Sujay V.},
  title = {Benchmarking and evaluating the NASA Land Information System (version 7.5.2) coupled with the refactored Noah-MP land surface model (version 5.0)},
  journal = {Geoscientific model development},
  year = {2025},
  doi = {10.5194/gmd-18-8439-2025},
  url = {https://doi.org/10.5194/gmd-18-8439-2025}
}