Koralegedara et al. (2025) Springtime soil moisture variability and its changing environmental drivers: a CMIP6 multi-model ensemble analysis for the subtropical East Asian region

Identification

• Journal: Geoscience Letters
• Year: 2025
• Date: 2025-11-07
• Authors: Suranjith Bandara Koralegedara, Wan‐Ru Huang, Tzu‐Yang Chiang, Hai Bui‐Manh
• DOI: 10.1186/s40562-025-00436-z

Research Groups

• Department of Earth Sciences, National Taiwan Normal University, Taipei, Taiwan

Short Summary

This study projects springtime soil moisture changes in the subtropical East Asian region (STEA) using a 14-model CMIP6 ensemble and machine learning, revealing a critical shift from rainfall/runoff dominance to near-surface temperature dominance as the primary environmental driver of soil moisture depletion under future warming. This reorganization leads to progressive drought vulnerability, with surface soil moisture decreasing by 9.1% and total soil moisture by 5.7% by the far-future.

Objective

• To evaluate CMIP6-model performances in simulating springtime surface (0–10 cm) and subsurface (0–100 cm) soil moisture variability in the subtropical East Asian region (STEA).
• To analyze projected springtime soil moisture changes under the high-emission Shared Socioeconomic Pathway 5–8.5 (SSP5-8.5) scenario and identify their key environmental drivers.

Study Configuration

• Spatial Scale: Subtropical East Asian region (STEA) [113°E–122°E, 21°N–26°N], with subdomains Western Subdomain (WSD) [113°E–119°E, 21°N–26°N] and Eastern Subdomain (ESD) [119°E–122°E, 21°N–26°N]. Broader East Asian domain [95°E–145°E, 0°–42°N] for climatological context.
• Temporal Scale: Springtime (February to April) for three periods: historical baseline (1995–2014), mid-future (2041–2060), and far-future (2081–2100).

Methodology and Data

• Models used: 14 CMIP6 models (multi-model ensemble mean, CMIP6-MME), Random Forest (RF) machine learning framework for driver importance ranking.
• Data sources: Monthly mean outputs from 14 CMIP6 models under the SSP5-8.5 scenario; ERA5-Land (ERA5-L) reanalysis data at 0.25° × 0.25° resolution as historical reference. Environmental drivers include rainfall, evapotranspiration, runoff, and 2 m-temperature.

Main Results

• The CMIP6-MME successfully reproduces historical springtime soil moisture patterns over East Asia, achieving Taylor Skill Scores (TSS) of 0.9 for surface soil moisture and 0.8 for total soil moisture over STEA, outperforming individual models. 86.7% (73.3%) of individual models exceeded a TSS threshold of 0.6 for surface (total) soil moisture.
• Springtime soil moisture is projected to progressively decline over STEA under the SSP5-8.5 scenario. Surface soil moisture is projected to decrease by 3.1% (mid-future) and 9.1% (far-future), while total soil moisture is projected to decrease by 2.0% (mid-future) and 5.7% (far-future).
• A fundamental reorganization of environmental drivers for soil moisture variability is identified:
  • Historically (1995–2014), rainfall (Random Forest Importance Score, RFIS: 0.7 for surface, 0.6 for total) and runoff (RFIS: 0.4 for surface, 0.3 for total) were the dominant drivers.
  • By the mid-future (2041–2060), runoff becomes the dominant driver for surface soil moisture (RFIS: 0.5), while rainfall's influence decreases (RFIS: 0.3).
  • By the far-future (2081–2100), near-surface temperature becomes the dominant driver (RFIS: 0.5 for surface, 0.6 for total), surpassing rainfall and runoff.
• This temperature dominance coincides with a projected 2 m-temperature increase of 11.9% in the mid-future and 25.4% in the far-future, while rainfall shows a slight increase of 2.0% in the mid-future followed by a 9.7% decrease in the far-future.

Contributions

• Provides one of the first detailed assessments of multi-layer springtime soil moisture characteristics and their evolving environmental drivers across the subtropical East Asian region (STEA) under climate change.
• Uniquely integrates multi-layer soil moisture analysis with machine learning-based driver attribution to reveal temporal shifts in climate controls.
• Offers the first quantitative assessment of a critical hydrological regime transition in STEA, showing a shift from rainfall/runoff dominance to temperature dominance as the primary driver of soil moisture variability.
• Highlights that traditional rainfall-centric water management strategies may become inadequate, emphasizing the need for adaptive approaches that integrate temperature projections into future water resource management.

Funding

• National Science and Technology Council of Taiwan (NSTC 113-2111-M-003-003, NSTC 113-2625-M-003-003, NSTC 114-2111-M-003-006, NSTC 114-2625-M-003-002)
• NSTC 113-2811-M-003-030 (for S.B. Koralegedara)
• NSTC 113-2811-M-003-016 (for Hai Bui-Manh)
• Open access funding provided by National Taiwan Normal University.

Citation

@article{Koralegedara2025Springtime,
  author = {Koralegedara, Suranjith Bandara and Huang, Wan‐Ru and Chiang, Tzu‐Yang and Bui‐Manh, Hai},
  title = {Springtime soil moisture variability and its changing environmental drivers: a CMIP6 multi-model ensemble analysis for the subtropical East Asian region},
  journal = {Geoscience Letters},
  year = {2025},
  doi = {10.1186/s40562-025-00436-z},
  url = {https://doi.org/10.1186/s40562-025-00436-z}
}

Generated by BiblioAssistant using gemini-2.5-flash (Google API)