Lumban-Gaol et al. (2025) Peat subsidence and dynamics in Midden-Delfland, the Netherlands, from time series InSAR analysis and the SPAMS model

Identification

• Journal: Geoderma
• Year: 2025
• Date: 2025-10-21
• Authors: Yustisi Lumban-Gaol, Philip Conroy, Simon A. N. van Diepen, Freek van Leijen, Ramon F. Hanssen
• DOI: 10.1016/j.geoderma.2025.117551

Research Groups

• Department of Geoscience and Remote Sensing, Delft University of Technology, The Netherlands
• National Research and Innovation Agency, Indonesia

Short Summary

This study estimates and analyzes peat subsidence in the Midden-Delfland region, The Netherlands, using Sentinel-1 InSAR data and the SPAMS model. It reveals an average subsidence rate of −5.4 ± 0.7 mm/year, with irreversible subsidence strongly correlated with dry climatic conditions, particularly during drought periods.

Objective

• To estimate and analyze surface motion in a regional peat area in Midden-Delfland, The Netherlands, using Sentinel-1 data and the SPAMS model in unsupervised contexts where ground truth data are unavailable.
• To assess the spatio-temporal variations of surface elevation changes by considering meteorological data, soil profiles, and elevation.

Study Configuration

• Spatial Scale: Midden-Delfland region, The Netherlands, covering approximately 160 square kilometres.
• Temporal Scale: From 1 January 2016 to 31 December 2024 (1675 Sentinel-1 acquisitions).

Methodology and Data

• Models used:
  • Interferometric Synthetic Aperture Radar (InSAR) time series analysis
  • SPAMS (Simple Parameterization for the Motion of Soils) model
  • DECADE (DElft Contextually Aided Distributed scatterer Environment) workflow
  • EMI (Eigendecomposition-based Maximum-likelihood-estimator of Interferometric phase)
  • Doris software for SAR image coregistration and geolocation
• Data sources:
  • Satellite: Sentinel-1A/B (C-band, Interferometric Wide Swath (IW) mode, VV polarization, 12-day repeat period, 6-day combined repeat cycle from two satellites, 1–3 day revisit from four orbit tracks).
  • Observation/Contextual:
    • Parcel data (geometry, crop type) from PDOK-BRP.
    • Soil data (shallow soil type < 1.2 m depth) from PDOK-BRO.
    • Water table data (water table zone) from PDOK-Peilgebied.
    • Meteorological data (daily precipitation, evapotranspiration) from Royal Netherlands Meteorological Institute (KNMI) stations.
    • Elevation data (terrain elevation, 0.5 m resolution) from AHN (airborne laser scanning, 2022).
    • Lithology data (soil class per 50 cm from surface down to 50 m depth) from DINOloket (TNO Geological Survey model).

Main Results

• The study area exhibits an average subsidence rate of −5.4 ± 0.7 mm/year, with rates ranging from −2.3 mm/year to −8.2 mm/year.
• Irreversible subsidence is strongly correlated with climatic conditions, with the most significant subsidence observed during prolonged dry periods in the summers of 2018 and 2022.
• Seasonal patterns show winter uplift due to precipitation surplus and reduced evapotranspiration, and summer subsidence due to water deficit, with oscillations typically ranging from 30 mm to 50 mm.
• Extreme drought events (e.g., 2018 and 2022) amplified these dynamics, causing subsidence peaks of 50 mm to 70 mm over six months.
• A Pearson correlation of 0.79 (p-value = 0.01) was found between mean annual irreversible subsidence and the annual Standardized Precipitation Evapotranspiration Index (SPEI).
• SPAMS model parameters (𝑥P, 𝑥E, 𝑥I) show spatial variability across the region, with medians of 0.070, 0.116, and −0.0273, respectively.
• The ̂𝜎2-statistic, indicating model adequacy, ranged from 0.75 to 1.15 for sample sites, with an overall range of 0.45 to 9.04 across the study area, suggesting good model performance for most parcels.

Contributions

• Provides a practical framework for interpreting InSAR-based surface displacement estimates in unsupervised contexts where ground truth data are unavailable.
• Advances the understanding of complex interactions between climate, soil composition, hydrological parameters, and elevation in driving peatland subsidence.
• Integrates InSAR and the SPAMS model as a valuable tool for monitoring peat surface elevation, informing water management, and reducing peatland degradation.
• Quantifies the distinct contributions of reversible and irreversible subsidence components and their link to climatic extremes.

Funding

• Indonesian Endowment Fund for Education (LPDP)
• European Union’s Copernicus program

Citation

@article{LumbanGaol2025Peat,
  author = {Lumban-Gaol, Yustisi and Conroy, Philip and Diepen, Simon A. N. van and Leijen, Freek van and Hanssen, Ramon F.},
  title = {Peat subsidence and dynamics in Midden-Delfland, the Netherlands, from time series InSAR analysis and the SPAMS model},
  journal = {Geoderma},
  year = {2025},
  doi = {10.1016/j.geoderma.2025.117551},
  url = {https://doi.org/10.1016/j.geoderma.2025.117551}
}