Analysis of the drivers of tropical soil acidification in Hainan Island based on explainable machine learning

Read full text Summarise with VIRAT
Authors: Yichen Wang, Lingbo Yang, Dongming Zhang, Limin Wang, Zhonghe Zhao, Zhenyu Lin, Lei Wen, Wen Zhang
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Soil acidification poses a severe threat to tropical agriculture. This study establishes an explainable machine learning framework to quantify drivers of soil acidification in Hainan Island, China. Total runoff and nitrogen fertilizer application emerged as the dominant natural and anthropogenic drivers. Hydrological factors exert contrasting effects: total runoff intensifies acidification, whereas subsurface runoff buffers it. A critical interaction was identified, with nitrogen fertilizer exerting its strongest acidifying effect under medium runoff conditions but weakening under extreme conditions. Spatial SHAP mapping revealed acidification hotspots concentrated in the central-western region. The findings highlight that effective mitigation requires integrating hydrothermal conditions with human inputs and tailoring fertilization strategies to local hydrological regimes, providing a scientific basis for precision soil management.

Fields of Study & Subjects

Soil Geostatistics and Mapping Soil Management and Crop Yield Soil erosion and sediment transport

Publication details

Journal / source
Figshare
Publication date
30 March 2026
Language
EN
Document type
Dataset