Spatial changes of nutrients and metallic contaminants in topsoil with multi-geostatistical approaches in a large-size watershed
Appropriate assessment on concerned soil contaminants spatially is of importance for decision-makers and stakeholders to make efficient mitigation countermeasures. In this study, we applied multiple geostatistical approaches to explore soil nutrient and metallic contaminant distributions in a large...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Article |
| Published: |
2023
|
| Subjects: | |
| Online Access: | https://repository.li.mahidol.ac.th/handle/123456789/84703 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Mahidol University |
| Summary: | Appropriate assessment on concerned soil contaminants spatially is of importance for decision-makers and stakeholders to make efficient mitigation countermeasures. In this study, we applied multiple geostatistical approaches to explore soil nutrient and metallic contaminant distributions in a large river watershed in Thailand, and to compare their performances in predicting spatial distribution patterns of the concerned soil contaminants under suitable application scenarios. The total carbon, nitrogen and phosphorous in surface soils over the whole watershed were measured with their maximum concentrations up to 131.47, 9.24, 5.33 g·kg−1, respectively, while the concentrations of eight metallic elements (Cu, Zn, Pb, Cd, Hg, As, Cr, and Ni) were 933.00, 6862.50, 373.00, 6.22, 1.15, 178.53, 761.11, and 372.44 mg·kg−1, respectively. It was found that the conditional interpolation approaches such as land use stratified inverse distance weighted and land use stratified original kriging provided better boundary details than original interpolations, with substantially reduced root mean square errors (up to 28% for nutrients and 54% for specific metals) and mean relative errors (up to 38% for nutrients and specific metals respectively) in predicting the spatial patterns of soil nutrients and several land use specific metals (Cu, Zn, Cd, and Pb). The global accuracies were equivalent or higher than those of geographically weighted regression. Nonetheless, the prediction accuracy for Cr, Ni, As, and Hg could not be improved using the land use stratified interpolation because their sources and pathways were not significantly correlated with land use types in the watershed, as reflected by the results of analysis of variance with post hoc test (p ≤ 0.05) and principal component analysis. |
|---|