Koster et al. (2025) Investigations into the Subseasonal Predictability of Soil Moisture and Streamflow

⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.

Identification

• Journal: Journal of Hydrometeorology
• Year: 2025
• Date: 2025-10-07
• Authors: Randal D. Koster, Yuna Lim, Eunjee Lee, Jana Kolassa
• DOI: 10.1175/jhm-d-25-0010.1

Research Groups

Not explicitly stated in the abstract.

Short Summary

This study investigates the subseasonal predictability of soil moisture and streamflow using an offline land modeling system forced by S2S forecasts. It finds that the offline system's hydrological forecasts are comparable in skill to fully coupled S2S systems, with skill influenced by soil column depth and improved streamflow forecasts achievable through climatological precipitation forcing.

Objective

• To examine the predictability of hydrological variables (soil moisture and streamflow) at subseasonal time scales using an offline (land-only) modeling system forced by state-of-the-art subseasonal to seasonal (S2S) forecast products.

Study Configuration

• Spatial Scale: Not explicitly defined, but focuses on land processes.
• Temporal Scale: Subseasonal to seasonal (S2S) time scales, with examination of skill at longer leads.

Methodology and Data

• Models used: An unspecified offline (land-only) modeling system. Comparisons are made against fully coupled S2S systems.
• Data sources:
  • Meteorological forcing: State-of-the-art subseasonal to seasonal (S2S) forecast products.
  • Validation: In situ measurements of soil moisture and streamflow.
  • Additional forcing: Climatological precipitation.

Main Results

• The offline land modeling system's hydrological forecasts (soil moisture and streamflow) demonstrate skill comparable to or exceeding those generated by fully coupled S2S systems.
• Subseasonal streamflow forecasts exhibit lower skill than soil moisture forecasts, primarily attributed to the poor quality of subseasonal precipitation predictions.
• Streamflow forecast skill can be significantly enhanced by incorporating climatological precipitation forcing during part or all of the forecast period, thereby increasing the influence of soil moisture initialization.
• Forecast skill for both soil moisture and streamflow is sensitive to the modeled soil column depth; deeper soil columns, associated with longer soil moisture memory, provide greater advantages at longer lead times.

Contributions

• Illustrates the capability of an offline land modeling system to extract additional hydrological prediction skill from existing S2S forecasts.
• Advances understanding of hydrological predictability and the physical factors that govern it, particularly at subseasonal scales.
• Identifies the sensitivity of hydrological forecast skill to modeled soil column depth and soil moisture memory.
• Proposes a practical method to improve subseasonal streamflow forecast skill by leveraging climatological precipitation forcing to compensate for poor precipitation predictions.

Funding

Not explicitly stated in the abstract.

Citation

@article{Koster2025Investigations,
  author = {Koster, Randal D. and Lim, Yuna and Lee, Eunjee and Kolassa, Jana},
  title = {Investigations into the Subseasonal Predictability of Soil Moisture and Streamflow},
  journal = {Journal of Hydrometeorology},
  year = {2025},
  doi = {10.1175/jhm-d-25-0010.1},
  url = {https://doi.org/10.1175/jhm-d-25-0010.1}
}