EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Soil Organic Matter Differences Among Land-Use Types and Spatial Prediction in an Urbanizing Landscape

Hongzhu Zou, Xiaoxue Guo, Danwei Chen, Hongfei Zhao, Haifei Lu, Tailin Zhong, Kexin Zhang

Forests · 14.11.2026 · 17 (11) · 1355

DOI: 10.3390/f17111355 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Forests, online erschienen am 14.11.2026. Verfasst von Hongzhu Zou und 6 weiteren Autor:innen. Die Arbeit nennt 66 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Zhejiang Shuren University, Zhejiang Shuren University Basic Scientific Research Special Funds, Key Specialty Project of Ordinary Colleges and Universities in Shaoxing City, Zhejiang Province, “Landscape Architecture”.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In rapidly urbanizing landscapes characterized by pronounced differences in vegetation, management, and construction disturbance, land-use type can significantly influence the spatial variability of soil organic matter (SOM). Using observations from 89 soil sampling sites in Xiong’an New Area, China, this study quantified differences associated with land-use type and soil depth through mixed-effects modelling. Leave-one-out cross-validation (LOOCV), five-fold cross-validation, and spatial validation were used to evaluate SOM spatial predictions from ordinary kriging (OK), global principal component regression kriging (RK), and land-use-specific regression kriging (LU-RK). In the 0–30 cm layer, mean SOM content was significantly higher in forest land than in farmland and construction land, with values of 15.52, 13.70, and 9.56 g·kg−1, respectively. SOM content decreased with increasing soil depth, and the difference between the surface and deeper layers varied among land-use types. RK outperformed the other two methods at both depths, with RMSE values of 3.755 and 2.593 g·kg−1 and predictive R2 values of 0.379 and 0.528, respectively. Five-fold cross-validation and spatial validation further supported the relative advantage of RK, although LU-RK showed a slight improvement in its relative performance with a 7.5 km buffer. Overall, RK integrating soil and environmental covariates provided comparatively good predictions of SOM variation across soil depths within a mosaic of land-use types. Its deeper-layer predictions retained some predictive ability under greater spatial separation, providing evidence to inform SOM prediction model selection in rapidly urbanizing landscapes.

Förderung

  • Zhejiang Shuren University
  • Zhejiang Shuren University Basic Scientific Research Special Funds
  • Key Specialty Project of Ordinary Colleges and Universities in Shaoxing City, Zhejiang Province, “Landscape Architecture”

Zitieren

APA

Zou, H., Guo, X., Chen, D., Zhao, H., Lu, H., Zhong, T., & Zhang, K. (2026). Soil Organic Matter Differences Among Land-Use Types and Spatial Prediction in an Urbanizing Landscape. Forests, 17(11), 1355. https://doi.org/10.3390/f17111355

Vancouver

Zou H, Guo X, Chen D, Zhao H, Lu H, Zhong T, et al. Soil Organic Matter Differences Among Land-Use Types and Spatial Prediction in an Urbanizing Landscape. Forests. 2026;17(11):1355. doi:10.3390/f17111355

BibTeX
@article{Zou2026,
  author = {Zou, Hongzhu and Guo, Xiaoxue and Chen, Danwei and Zhao, Hongfei and Lu, Haifei and Zhong, Tailin and Zhang, Kexin},
  title = {Soil Organic Matter Differences Among Land-Use Types and Spatial Prediction in an Urbanizing Landscape},
  journal = {Forests},
  year = {2026},
  volume = {17},
  number = {11},
  pages = {1355},
  doi = {10.3390/f17111355}
}
Kostenlos starten →