PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
June 4, 2026Land Degradation and Development0 citations

An Integrated Multi‐Source Factor Screening Strategy for Unraveling Response Mechanisms of Organic Matter Variability in the Coastal Saline Soil of China

View Full Paper
BHBingjie HanXLXin LiYLYujing Lei

Key Points

  • The aim is to develop an integrated strategy for accurately predicting soil organic matter (SOM) and understanding its response to environmental factors.
  • Used integrated strategy combining stepwise regression, Pearson correlation, ANOVA, geodetector, and random forest.
  • Employed TPE-XGB model for estimating SOM content.
  • Conducted SHapley Additive exPlanations analysis to identify key driving factors.
  • Integrated strategy showed R2 = 0.83, RMSE = 0.16 g kg−1, and RPD = 2.13 for estimating SOM content.
  • Climatic, soil, and vegetation factors consistently identified as key elements driving SOM variation.
  • Analysis revealed significant interaction effects between climatic and soil properties on SOM.

Abstract

ABSTRACT Accurately interpreting the response mechanisms of soil organic matter (SOM) with the assistance of environmental factors is crucial for promoting high‐quality agricultural development in coastal saline‐alkali land and maintaining ecosystem stability. However, a single screening method is often insufficient to comprehensively characterize the integrated effects of multiple complex environmental gradients on SOM. Here, we proposed an integrated screening strategy combining stepwise regression (SR), Pearson correlation (PC), analysis of variance (ANOVA), geodetector (GD), and random forest (RF). This strategy aimed to achieve high‐accuracy SOM prediction and systematically interpret its response mechanisms. The results showed that although the factor datasets obtained by different single methods differed significantly, they consistently identified climatic factors (CLI), SOIL, and vegetation factors as key elements. These factors also showed highly similar numerical change trends to SOM. The distribution characteristics of the C/N ratio (mean = 9.24, CV = 0.27) indicated that the carbon–nitrogen system was generally coupled under salinity stress, while carbon and nitrogen showed asynchronous responses at the process level. Compared with five single methods, the integrated screening strategy combined with the TPE‐XGB model demonstrated the best performance for estimating SOM content ( R 2 = 0.83, RMSE = 0.16 g kg −1 , RPD = 2.13). The SHapley Additive exPlanations (SHAP) analysis revealed that the selected SOIL and CLI showed strong driving effects on SOM (MSV > 85%). The variation partitioning (VP) and hierarchical partitioning (HP) further showed that the synergistic effect of climate and soil properties had a significant response to SOM variation. In summary, the integrated multi‐source factor screening strategy can accurately interpret the response mechanisms of SOM variation and is highly feasible, providing strong support for regional smart agriculture development and ecological restoration of saline‐alkali land.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Han et al. (2026) studied this question.

synapsesocial.com/papers/6a2116fad499ed480b16fd1chttps://doi.org/10.1002/ldr.70698
Ask AI
Helpful
Bookmark
Share
View Full Paper