The modern Yellow River Delta (YRD) in China is sinking under the combined effect of natural consolidation and compaction of newly deposited river sediment and human activities such as underground brine extraction, oil exploitation, and groundwater drawdown. Liu et al. (2021) combined Small Baseline Subset Interferometric Synthetic Aperture Radar (SBAS-InSAR) measurements from 1992–2010 with a one-dimensional geotechnical consolidation model and concluded that long-term delta settlement is primarily controlled by natural compaction of recent river sediments, but they left open the quantitative attribution of natural versus human-induced subsidence across the delta, the settlement behavior of the very newest deposits where radar coherence is lost, and the refinement of simplifying geotechnical assumptions. A substantial 2021–2026 wave of Sentinel-1 time-series InSAR studies has since identified and statistically quantified the anthropogenic drivers — underground brine extraction above all, with local rates of several hundred mm/yr, followed by oil-field exploitation and groundwater drawdown — while natural compaction persists as a slow delta-wide background. What remains unresolved is a cause-resolved subsidence budget for the entire delta: direct measurement of deformation over the actively depositing zones, a physical partition of the observed subsidence field into natural and anthropogenic components with quantified uncertainty, and a layer-resolved consolidation model that relaxes the simplifying assumptions of the 2021 analysis.

Background

Many of the world's large deltas are sinking, and the modern Yellow River Delta (YRD) in China — built by the Yellow River since it shifted course in 1855 through successive distributary-channel sub-lobes, each leaving sediments of different age — is among the most dynamically subsiding. Two families of processes drive the sinking. The first is natural: newly deposited, water-rich silty and clayey sediment gradually expels pore water under its own weight and that of overlying deposits (consolidation and compaction), producing slow surface settlement that continues for decades after deposition; the amount of settlement depends on sediment compressibility, layer thickness, drainage paths, and the initial excess pore pressure (the pressure above hydrostatic trapped in the sediment at deposition). The second is anthropogenic: extraction of underground fluids — brine from coastal salt-field aquifers, oil from the Shengli oilfield, and fresh groundwater — lowers fluid pressure and compacts the strata, and urban or industrial loading adds surface settlement.

Surface motion is measured mainly by satellite radar interferometry. Interferometric Synthetic Aperture Radar (InSAR) compares phases of repeated radar images to map ground deformation; multi-temporal InSAR (MT-InSAR) methods such as the Small Baseline Subset (SBAS) and persistent-scatterer (PS) approaches combine many image pairs to retrieve displacement time series at coherent points. Persistent scatterers are stable radar reflectors (buildings, rocks), while distributed-scatterer techniques (e.g., SqueeSAR-type processing) recover points that are individually unstable but behave alike within a neighborhood, extending coverage into rural terrain. C-band radar (wavelength about 5.6 cm, e.g., ERS, Envisat, Sentinel-1) loses coherence over the YRD's active deposition zones — new delta sub-lobes and channel mouths — because the surface there changes between acquisitions, leaving spatial gaps in the measurements.

Prior work established both process families for the YRD. Zhang et al. (2017) modeled self-weight consolidation of the delta sediment with a Monte Carlo approach; Liu et al. (2019) estimated compaction rates delta-wide from 108 sediment cores and related them to the buried channel network; Zhang et al. (2020) compared soft-soil compaction and groundwater extraction as subsidence causes; Zhang et al. (2019) mapped YRD subsidence with PS plus distributed-scatterer InSAR, showing marked intensification between 2007–2010 and 2015–2018. Liu et al. (2021) then combined 75 ERS/Envisat SAR scenes (1992–2010) in three SBAS subsets calibrated with geodetic leveling (precision ground-survey height measurements) and a one-dimensional geotechnical consolidation model, finding annual settlement rates of 0 to more than 30 mm/yr, an average coastal-margin rate near 15 mm/yr, a seaward migration of the strongest subsidence with younger sub-lobe age, and strong agreement between InSAR time series and modeled settlement at representative boreholes — evidence that decadal- to century-scale settlement is primarily controlled by natural consolidation and compaction of recently deposited river sediment. They nevertheless noted that InSAR showed faster settlement than the model predicted at several sites during 2003–2010, that no measurement existed over the youngest depositional zone (their sedimentary zone I and the active channel mouths) because of coherence loss, and that their model assumed a uniform 2 m groundwater table, purely vertical one-dimensional compaction with a single drainage path per layer, laboratory-average geotechnical parameters, and a prescribed initial excess pore-pressure distribution.

After 2021, Sentinel-1 time-series studies transformed the attribution picture. Zhang et al. (2021) correlated 30 years of land-use change with InSAR deformation, finding salt-field subsidence rising from 13 mm/yr (1992–2000) to 453 mm/yr (2016–2017); Wang et al. (2022) isolated brine-mining subsidence funnels with InSAR and principal component analysis; Li et al. (2023) mapped funnels exceeding 50 mm/yr (maximum over 300 mm/yr) in brine-mining clusters and showed land subsidence will contribute at least three times as much as eustatic sea-level rise (the global, ocean-volume component of sea-level change) to 2050 flood risk; Wang et al. (2024) combined InSAR with machine learning to characterize coastal subsidence risk across the delta; Yu and Ma (2024, 2025) used ascending and descending Sentinel-1 stacks (2019–2022) to resolve spatiotemporal patterns of human-induced subsidence by activity type (maximum about −255 mm/yr); Fan et al. (2025) quantified how precipitation and runoff seasonally modulate human-induced subsidence; Liu et al. (2025), from the same group as the 2021 study, combined a 20-year SBAS record with Random Forest (a machine-learning ensemble-classification method) land-use analysis and a contraction source model (maximum vertical contraction 18 cm/yr) implicating brine extraction; Ge et al. (2026) used PS plus SBAS InSAR (2020–2021) with Random Forest attribution, ranking brine mining as the main contributor; Ma et al. (2026) integrated SAR, extensometers (borehole instruments that measure deformation of individual soil layers), and groundwater levels, identifying deep strata as the main subsiding layers and brine and deep groundwater exploitation as the key drivers; and Xu et al. (2026) used SBAS over 2017–2024 and explicitly quantified natural and human factors with a geodetector analysis (a spatial-statistical method that scores how strongly candidate factors explain the spatial variation of a response variable; explanatory power q = 0.21 for salt-field distribution, 0.14 for nearshore distance, 0.07 for GDP), with peak rates of 325 mm/yr in brine-mining zones. These studies establish that the fast subsidence funnels are anthropogenic and that natural compaction forms a slow background, but they quantify drivers statistically — by spatial association, machine-learning variable importance, or geodetector q-values — rather than decomposing the observed deformation field itself.

Problem Statement

For the modern Yellow River Delta, determine the cause-resolved subsidence budget: quantitatively partition the observed land-subsidence field — from multi-temporal InSAR and independent geodetic observations — into (i) natural consolidation and compaction of newly deposited river sediment and (ii) anthropogenic components (underground brine extraction, oil exploitation, groundwater drawdown, surface loading), with per-component rates and uncertainties, across the entire delta including the actively depositing sedimentary zone I and the active channel mouths where C-band InSAR loses coherence. The partition must be reconciled with a layer-resolved geotechnical consolidation model that relaxes the simplifying assumptions of Liu et al. (2021) — replacing the uniform 2 m groundwater table, one-dimensional single-drainage compaction, laboratory-average geotechnical parameters, and prescribed initial excess pore-pressure distribution with spatially variable, data-constrained formulations — and validated against independent observations (leveling, GNSS (Global Navigation Satellite System) positioning, extensometers, groundwater-level records). The settlement behavior of the youngest depositional zones must be measured where coherent observation is possible and bounded otherwise.

References

  1. Liu Yilin, Liu Jinqing, Xia Xingfan, Bi Haibo, Huang Haijun, Ding Renwei, Zhao Lihong (2021). Land subsidence of the Yellow River Delta in China driven by river sediment compaction. Science of The Total Environment 750:142165. DOI: 10.1016/j.scitotenv.2020.142165. https://doi.org/10.1016/j.scitotenv.2020.142165
  2. Zhang Yi, Huang Haijun, Liu Yanxia, Liu Yilin (2017). Self-weight consolidation and compaction of sediment in the Yellow River Delta, China. Physical Geography 39(1):84–98. DOI: 10.1080/02723646.2017.1347420. https://doi.org/10.1080/02723646.2017.1347420
  3. Liu Yanxia, Du Tingqin, Huang Haijun, Liu Yilin, Zhang Yi (2019). Estimation of sediment compaction and its relationship with river channel distributions in the Yellow River delta, China. CATENA 182:104113. DOI: 10.1016/j.catena.2019.104113. https://doi.org/10.1016/j.catena.2019.104113
  4. Zhang Bowen, Wang Robert, Deng Yunkai, Ma Peifeng, Lin Hui, Wang Jili (2019). Mapping the Yellow River Delta land subsidence with multitemporal SAR interferometry by exploiting both persistent and distributed scatterers. ISPRS Journal of Photogrammetry and Remote Sensing 148:157–173. DOI: 10.1016/j.isprsjprs.2018.12.008. https://doi.org/10.1016/j.isprsjprs.2018.12.008
  5. Zhang Yi, Huang Haijun, Liu Yanxia, Bi Haibo, Zhang Zehua, Wang Kuifeng, Yan Liwen (2020). Impacts of soft soil compaction and groundwater extraction on subsidence in the Yellow River Delta. Marine Georesources & Geotechnology 39(12):1497–1504. DOI: 10.1080/1064119X.2020.1854909. https://doi.org/10.1080/1064119X.2020.1854909
  6. Zhang Yi, Liu Yilin, Zhang Xinyuan, Huang Haijun, Qin Keyu, Bai Zechao, Zhou Xinghua (2021). Correlation Analysis between Land-Use/Cover Change and Coastal Subsidence in the Yellow River Delta, China: Reviewing the Past and Prospecting the Future. Remote Sensing 13(22):4563. DOI: 10.3390/rs13224563. https://doi.org/10.3390/rs13224563
  7. Wang Guoyang, Li Peng, Li Zhenhong, Liang Cunren, Wang Houjie (2022). Coastal subsidence detection and characterization caused by brine mining over the Yellow River Delta using time series InSAR and PCA. International Journal of Applied Earth Observation and Geoinformation 115:103077. DOI: 10.1016/j.jag.2022.103077. https://doi.org/10.1016/j.jag.2022.103077
  8. Li Peng, Wang Guoyang, Liang Cunren, Wang Houjie, Li Zhenhong (2023). InSAR-Derived Coastal Subsidence Reveals New Inundation Scenarios Over the Yellow River Delta. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 16:8431–8441. DOI: 10.1109/JSTARS.2023.3272782. https://doi.org/10.1109/JSTARS.2023.3272782
  9. Wang Guoyang, Li Peng, Li Zhenhong, Liu Jie, Zhang Yi, Wang Houjie (2024). InSAR and machine learning reveal new understanding of coastal subsidence risk in the Yellow River Delta, China. Science of The Total Environment 915:170203. DOI: 10.1016/j.scitotenv.2024.170203. https://doi.org/10.1016/j.scitotenv.2024.170203
  10. Yu Bing, Ma Deying (2024). Spatiotemporal Patterns of Human-Induced Subsidence in the Yellow River Delta Region revealed by Ascending and Descending Orbit Time-Series InSAR. EGU General Assembly 2024, abstract EGU24-20003. DOI: 10.5194/egusphere-egu24-20003. https://doi.org/10.5194/egusphere-egu24-20003
  11. Liu Yilin, Yan Bing, Huang Jue, Zhang Naipeng, Niu Yufen, Peng Mimi, Qu Feifei, Huang Haijun, Zhao Lihong (2025). Accelerated coastal subsidence variability across the Yellow River Delta due to anthropogenic land use and land cover changes over the last two decades. Natural Hazards 121:21583–21610. DOI: 10.1007/s11069-025-07655-9. https://doi.org/10.1007/s11069-025-07655-9
  12. Yu Bing, Ni Jialong, Liu Guoxiang, Liu Hengchong, Zhang Rui, Dai Keren, Wang Xiaowen, Zhang Bo (2025). Spatiotemporal patterns of different human activity induced subsidence over the Yellow River Delta revealed by MT-InSAR. Physical Geography 46(5):367–406. DOI: 10.1080/02723646.2025.2542903. https://doi.org/10.1080/02723646.2025.2542903
  13. Fan Yaoshen, Wang Guangzhou, Dou Shentang, Jiang Chao, Ji Hongyu, Chen Shenliang, Du Xiaokang, Yu Shoubing, Wu Yan, Zhang Shaohua (2025). Influence of precipitation and runoff on human-induced land subsidence in the Yellow River Delta. Quaternary Science Advances 19:100288. DOI: 10.1016/j.qsa.2025.100288. https://doi.org/10.1016/j.qsa.2025.100288
  14. Ge Pengfei, Chen Mi, Tomás Roberto, Liu Hui, Fan Kailun, Cheng Xi, Fu Xingyuan, Wang Shuang (2026). Insights into land deformation processes in the Yellow River Delta (China) from synthetic aperture radar interferometry time series and machine learning. Advances in Space Research 77(4):4527–4551. DOI: 10.1016/j.asr.2025.12.029. https://doi.org/10.1016/j.asr.2025.12.029
  15. Ma Ziyao, Chen Beibei, Gong Huili, Zhou Chaofan, Xu Lewei, Wang Xincheng, Li Xiaojuan, Wang Haigang, Lei Kunchao, Kang Fengxin (2026). Impact of land subsidence on relative sea level rise and coastal inundation in the Yellow River Delta. Journal of Hydrology: Regional Studies 63:103020. DOI: 10.1016/j.ejrh.2025.103020. https://doi.org/10.1016/j.ejrh.2025.103020
  16. Xu Hui, Wang De, Ning Rongrong, Li Xiaoyong, Tian Xinpeng, Bi Xiaoli, Zhou Zixiang, Zhang Yongwei (2026). Human activity intensifies land subsidence in the Yellow River Delta. Estuarine, Coastal and Shelf Science 330:109721. DOI: 10.1016/j.ecss.2026.109721. https://doi.org/10.1016/j.ecss.2026.109721