Comparative analysis of soil organic carbon across different land types in plateau wetlands using Kriging interpolation based on spatial heterogeneity.
Wetlands, as an important global carbon reservoir, contribute significantly to the terrestrial carbon cycle. However, the high spatial heterogeneity of wetland ecosystems poses a considerable challenge to accurate estimation and mapping of soil organic carbon (SOC). In this study, we focused on Caoh...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science (PLoS)
2025-01-01
|
| Series: | PLoS ONE |
| Online Access: | https://doi.org/10.1371/journal.pone.0328246 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Wetlands, as an important global carbon reservoir, contribute significantly to the terrestrial carbon cycle. However, the high spatial heterogeneity of wetland ecosystems poses a considerable challenge to accurate estimation and mapping of soil organic carbon (SOC). In this study, we focused on Caohai Wetland in Guizhou Province, China, a typical plateau freshwater wetland, to evaluate the spatial variability of SOC across five land use types. A total of 122 surface soil samples were collected, and SOC content was analyzed using three Kriging interpolation methods-Ordinary Kriging (OK), Simple Kriging (SK), and Universal Kriging (UK)-in combination with four semi-variogram models (Gaussian, Hole effect, J-Bessel, and K-Bessel). The results indicated that SOC distribution varied significantly among different soil types. The spatial variability was highest in swamp and grassland soils and lowest in agricultural and forest soils. Among the semi-variogram models, the J-Bessel model showed the best performance in capturing local variation patterns. OK and SK yielded lower RMSE values (2.41) and higher R² (0.913 and 0.911, respectively) than UK (RMSE = 2.80; R² = 0.863). Principal component analysis revealed that SOC was positively correlated with total nitrogen, available nitrogen, Cd, Zn, DDT, and OCPs, and negatively correlated with pH. The cumulative variance explained by the two principal components was 81.3%. These findings demonstrate that Bessel-type models combined with Ordinary or Simple Kriging provide superior prediction accuracy in highly heterogeneous wetland soils. The methodology offers a scientific basis for SOC spatial modeling and targeted soil carbon management strategies in plateau wetland ecosystems. |
|---|---|
| ISSN: | 1932-6203 |