EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Fixed-Budget, Long-Tail-Aware Adaptive Quadtree Framework for Spatial Training Data Representation in Land-Subsidence Modeling

Yunxiao Liu, Mingliang Gao, Huili Gong, Min Shi, Lin Zhu, Rong Shen, Huayu Guan, Jie Wang, Zheng Chen

ISPRS International Journal of Geo-Information · 18.11.2026 · 15 (11) · 515

DOI: 10.3390/ijgi15110515 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ISPRS International Journal of Geo-Information, online erschienen am 18.11.2026. Verfasst von Yunxiao Liu und 8 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Basic Research Program of Jiangsu, Major Science and Technology Innovation Demonstration Special Project of the Inner Mongolia Autonomous Region.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Dense georeferenced training data often combine local redundancy with rare responses concentrated in small areas. This study proposes a fixed-budget, long-tail-aware adaptive quadtree framework that integrates response stratification, severe-cluster protection, projected-coordinate partitioning, and observed spatial medoids. The analysis used 28,587 one-kilometer land-subsidence records from California’s San Joaquin Valley and repeated blocked validation over five grid origins and five folds per origin. At a 35% output budget, the adaptive representation achieved a severe-tail RMSE of 56.36 mm/yr, compared with 68.60 for unweighted full-data training, 56.59 for response-weighted full-data training, 55.88 for response-stratified random sampling, and 56.29 for response-stratified local pivotal sampling. Its non-tail RMSE and bias were 28.73 and −11.14 mm/yr, versus 23.51 and −6.30 mm/yr for unweighted full-data training. A coordinate-only k-nearest-neighbor reference achieved a tail RMSE of 29.42 mm/yr and hotspot IoU of 0.643 in the primary within-footprint setting. Across 30 prespecified sampling seeds, both response-matched baselines showed narrow severe-tail RMSE distributions that overlapped the deterministic quadtree result. Across 21 model seeds and the 5- and 10-km buffer analyses, the proposed framework consistently achieved lower severe-tail RMSE than unweighted full-data training. The proposed framework provides deterministic, spatially structured training data reduction under a prespecified output budget.

Förderung

  • National Natural Science Foundation of China
  • Basic Research Program of Jiangsu
  • Major Science and Technology Innovation Demonstration Special Project of the Inner Mongolia Autonomous Region
  • Central Funds for Guiding Local Science and Technology Development
  • Open Fund of National Field Observation and Research Station (Hebei Cangzhou) for Groundwater and Land Subsidence

Zitieren

APA

Liu, Y., Gao, M., Gong, H., Shi, M., Zhu, L., Shen, R., Guan, H., Wang, J., & Chen, Z. (2026). A Fixed-Budget, Long-Tail-Aware Adaptive Quadtree Framework for Spatial Training Data Representation in Land-Subsidence Modeling. ISPRS International Journal of Geo-Information, 15(11), 515. https://doi.org/10.3390/ijgi15110515

Vancouver

Liu Y, Gao M, Gong H, Shi M, Zhu L, Shen R, et al. A Fixed-Budget, Long-Tail-Aware Adaptive Quadtree Framework for Spatial Training Data Representation in Land-Subsidence Modeling. ISPRS International Journal of Geo-Information. 2026;15(11):515. doi:10.3390/ijgi15110515

BibTeX
@article{Liu2026,
  author = {Liu, Yunxiao and Gao, Mingliang and Gong, Huili and Shi, Min and Zhu, Lin and Shen, Rong and Guan, Huayu and Wang, Jie and Chen, Zheng},
  title = {A Fixed-Budget, Long-Tail-Aware Adaptive Quadtree Framework for Spatial Training Data Representation in Land-Subsidence Modeling},
  journal = {ISPRS International Journal of Geo-Information},
  year = {2026},
  volume = {15},
  number = {11},
  pages = {515},
  doi = {10.3390/ijgi15110515}
}
Kostenlos starten →