Dong et al. (2026) Surface Soil Moisture Drydown over the Tibetan Plateau from SMAP: Consistency with In Situ Observations, Spatial Patterns and Controls

Identification

• Journal: Remote Sensing
• Year: 2026
• Date: 2026-03-06
• Authors: Shiyu Dong, Zhongli Zhu, Jinsong Zhang, Ziqi Liu, Qingxia Wu
• DOI: 10.3390/rs18050814

Research Groups

• State Key Laboratory of Earth Surface Processes and Disaster Risk Reduction, Faculty of Geographical Science, Beijing Normal University, Beijing 100875, China
• Key Laboratory of Ecosystem Network Observation and Modeling, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China
• Chinese Academy of Meteorological Sciences, Beijing 100081, China

Short Summary

This study evaluates the consistency of SMAP satellite-derived surface soil moisture drydown timescales (τ) with in situ observations over the Tibetan Plateau, maps its spatial patterns, and identifies dominant environmental controls. It finds that SMAP systematically yields shorter drydown timescales than in situ measurements, primarily due to differences in effective sensing depth and spatial representativeness, with τ exhibiting a clear southeast-to-northwest gradient driven by elevation, soil sand fraction, and vegetation.

Objective

• To assess the consistency in drydown event detection and drydown timescale (τ) estimation between SMAP satellite products and in situ sensor networks over the Tibetan Plateau.
• To map the Tibetan Plateau-wide spatial patterns of τ and identify the dominant controls, including topography, soil texture, and vegetation, in a high-altitude, heterogeneous environment.
• To clarify key sources of uncertainty in satellite-derived drydown dynamics, particularly sensor detection depth and scale mismatch between satellite footprints and point measurements.

Study Configuration

• Spatial Scale: Tibetan Plateau (approximately 2.5 million km²), analyzed at SMAP 36 km EASE-Grid 2.0 pixel resolution. In situ observations from five sensor networks (Tianjun, Maqu, Naqu, Ngari, CTP-SMTMN) across the plateau. Ancillary data (DEM, land cover, aridity index, soil texture, precipitation) were resampled to 36 km.
• Temporal Scale: SMAP data from 2016 to 2025. In situ observations from various periods within 2016–2025, with daily measurements at 6:00 a.m. local time. Drydown events were identified with a maximum length of 40 days.

Methodology and Data

• Models used:
  • Exponential decay model: θ(t) = A * exp(-t/τ) + θw fitted using nonlinear least squares (Levenberg–Marquardt algorithm via lmfit Python package) to identify drydown timescale (τ), amplitude (A), and minimum soil moisture (θw).
  • Partial Least Squares Regression (PLSR): Employed to quantify the sensitivity of τ to environmental factors (mean annual precipitation, potential evapotranspiration, vegetation, soil sand fraction, and elevation).
• Data sources:
  • Satellite: SMAP Level-3 passive soil moisture product (SMAPL3SM_P) at 36 km resolution, descending pass (~06:00 local time).
  • Observation (in situ): Soil moisture measurements at 5 cm depth from five high-density sensor networks on the Tibetan Plateau (Tianjun, Maqu, Naqu, Ngari, and the Central Tibetan Plateau Multiscale Soil Moisture and Temperature Monitoring Network (CTP-SMTMN)). Data obtained from the International Soil Moisture Network (ISMN) and the National Tibetan Plateau Data Center (TPDC).
  • Ancillary:
    • MODIS products: NDVI (MOD13A2) and land cover (MCD12C1, IGBP classification).
    • Climatic Aridity Index (AI) and Potential Evapotranspiration (ET0): 1 km Global-AIPETv3 (1970–2000).
    • Topography: 1 km GTOPO30 Digital Elevation Model (DEM).
    • Soil texture: 250 m Tibetan Plateau digital soil map.
    • Precipitation: Daily 0.1° IMERG Final product of Global Precipitation Measurement.

Main Results

• SMAP vs. In Situ Consistency: SMAP soil moisture showed good overall agreement with in situ observations, with a Pearson correlation coefficient (R) of 0.83 and a Root Mean Square Error (RMSE) of 0.07 cm³ cm⁻³ across all pooled network samples.
• Drydown Timescale Discrepancy: SMAP-derived drydown timescales (τ) were systematically smaller than those from in situ observations. The median τ from in situ observations (4.9–13.34 days) was approximately 1.5 to 3 times larger than that from SMAP (2.67–6.75 days). Overall, the median τ for SMAP was 4.41 days, representing a 56.6% reduction relative to the in situ median of 10.17 days. This faster drying in SMAP was supported by multi-layer in situ data showing increasing τ with depth and attributed to differences in effective sensing depth (0–5 cm for SMAP vs. 5 cm or deeper for in situ) and spatial representativeness.
• Spatial Patterns of τ: The drydown timescale exhibited a distinct southeast-to-northwest decreasing gradient across the Tibetan Plateau. τ values were typically less than 5 days in the arid Qaidam Basin (northern TP) and generally exceeded 10 days in the more humid eastern regions.
• Controls on τ: Partial Least Squares Regression (PLSR) identified elevation, soil sand fraction, and vegetation conditions (NDVI) as primary drivers of spatial τ variability.
  • τ generally increased from hyper-arid to sub-humid conditions.
  • τ decreased with increasing soil sand fraction, indicating faster drying in coarser-textured soils.
  • τ increased with elevation below 4500 m, then stabilized at higher elevations.
  • τ was generally smaller over bare soil, and showed a positive sensitivity to NDVI, suggesting canopy shading reduces soil evaporation for near-surface soil moisture.

Contributions

• Provides the first systematic, event-scale quantification and intercomparison of surface soil moisture drydown timescales (τ) between SMAP satellite products and extensive in situ networks over the complex Tibetan Plateau.
• Diagnoses the combined roles of topography, soil texture, and vegetation in shaping τ in a high-altitude, heterogeneous environment, revealing a pronounced southeast-to-northwest gradient.
• Clarifies key sources of uncertainty in satellite-derived drydown dynamics, specifically highlighting the critical impact of effective sensing depth and spatial representativeness on soil memory characterization.
• Offers observational constraints crucial for understanding land-surface hydrological processes and improving land–atmosphere coupling analyses in alpine regions, with implications for drought diagnostics.

Funding

• National Natural Science Foundation of China, grant number 42271337.

Citation

@article{Dong2026Surface,
  author = {Dong, Shiyu and Zhu, Zhongli and Zhang, Jinsong and Liu, Ziqi and Wu, Qingxia},
  title = {Surface Soil Moisture Drydown over the Tibetan Plateau from SMAP: Consistency with In Situ Observations, Spatial Patterns and Controls},
  journal = {Remote Sensing},
  year = {2026},
  doi = {10.3390/rs18050814},
  url = {https://doi.org/10.3390/rs18050814}
}