Quantitative Analysis of the Factors Influencing the Spatial Distribution of Benggang Landforms Based on a Geographical Detector

Kaitao Liao, Yuejun Song, Songhua Xie, Yichen Luo, Quan Liu, Hui LinISPRS Int. J. Geo Inf 2022 · DOI

TL;DR

This study investigates the complex factors driving "Benggang" erosion, a severe form of soil erosion in southern China, using a geographical detector. The research identifies rainfall erosivity, elevation, and land use as the most significant individual drivers, while also highlighting that the interaction between multiple factors, particularly rainfall and elevation, plays a crucial synergistic role in Benggang formation.

What Problem Does This Paper Address?

Benggang erosion is a unique and destructive landform phenomenon prevalent in the granite-red-soil hilly regions of southern China. It significantly impacts agricultural development and regional sustainability, leading to land degradation, sediment deposition in rivers, and threats to food security. While the general causes of Benggang erosion have been acknowledged (climate, topography, lithology, human activities), a comprehensive, quantitative analysis of the interactions between these driving factors at a regional scale has been lacking. This paper aims to fill this gap by identifying the key influencing factors and understanding how they combine to drive Benggang erosion in Ganzhou City, a representative area for this phenomenon.

Methods

The study employs a Geographical Detector model, a statistical tool designed to identify spatial differentiation and reveal driving factors. Specifically, it utilizes two detectors:

  • Factor Detector: This assesses the spatial heterogeneity of Benggang erosion and quantifies the explanatory power of individual factors using a "q-value." A higher q-value indicates a stronger influence of the factor on Benggang distribution.
  • Interaction Detector: This evaluates how pairs of factors interact to influence Benggang erosion, determining whether their combined effect is synergistic (enhanced), independent, or weakening. This is achieved by comparing the q-values of individual factors with the q-value of their interaction.

The researchers processed and classified 10 potential influencing factors (elevation, slope, aspect, slope type, lithology, soil type, rainfall erosivity, distance to river, vegetation coverage, and land use type) using ArcMap 10.2 before inputting them into the geographical detector.

Datasets

The study utilized the following datasets:

  • Benggang Erosion Data: Collected from surveys in South China (2004-2005) and Jiangxi Province (2015), including number, area, and geographical coordinates.
  • Digital Elevation Model (DEM): Advanced Spaceborne Thermal Emission and Reflection Radiometer Global Digital Elevation Model (ASTER GDEM 30).
  • Land Use and Land Cover Data: 30m resolution data from the Geospatial Data Cloud.
  • Geological Maps: 1:2.5 million scale maps from the China Geological Survey.
  • Daily Rainfall Data: From 26 stations in Ganzhou City and surrounding areas (1960-2016), used to calculate rainfall erosivity.
  • Soil Type Maps: 1:1 million scale maps from the Resource and Environment Science and Data Center.
  • Landsat Data: From the Aerospace Information Research Institute, used for vegetation coverage calculation.

Key Insights and Ideas

  • Benggang is a Multifaceted Problem: The research strongly emphasizes that Benggang erosion is not driven by a single factor but by a complex interplay of multiple elements.
  • Synergistic Interactions are Key: The interaction detector revealed that the combined effect of any two factors is always stronger than the effect of a single factor. This highlights the importance of considering these interactions for effective prevention and control.
  • Nonlinear Enhancement is Powerful: The study found that "nonlinear enhancement" interactions between factors exhibit the strongest synergistic effect on erosion.
  • Human Activities and Natural Factors are Intertwined: While natural factors like rainfall and elevation are critical, the paper repeatedly points to human activities (deforestation, land use changes) as significant accelerators of Benggang erosion, often in conjunction with favorable natural conditions.
  • Location Matters: The study underscores that the specific combination and interaction of factors leading to Benggang erosion are location-specific, necessitating tailored management strategies.

Key Results

  • Top Individual Factors: Rainfall erosivity (q=0.0113) and elevation (q=0.0111) were identified as the most influential individual factors. Land use (q=0.0035) also showed significant influence.
  • Optimal Conditions for Benggang: The study suggests that Benggang erosion is most likely to occur under conditions of:
  • Average annual rainfall erosivity of 6400–7000 MJ·mm/(hm²·ha).
  • Elevation between 100–400 m.
  • Vegetation coverage of 40-60%.
  • Slopes below 25° with a concave profile.
  • Proximity to rivers (within 500 m).
  • Presence of granite weathering crust and red soils.
  • Strongest Factor Interactions: The most significant interactions were found between:
  • Elevation and rainfall erosivity.
  • Elevation and distance to river.
  • Elevation and land use.
  • Rainfall erosivity and land use.
  • Dominant Drivers: Overall, rainfall, elevation, and land use were identified as the three most important factors, with their interactions being crucial.

Limitations and Future Work

  • Resolution of Factors: The authors acknowledge that the resolution of condition factors should be considered in future works, especially for regional-scale erosional analysis.
  • Generalizability: While the study provides a robust framework for Ganzhou City, the authors suggest that the methodological approach could be applied to other similar regions experiencing Benggang erosion in southern China.
  • Further Investigation: Future research could delve deeper into the specific mechanisms of these multi-factor interactions and refine the resolution of the input data for more precise modeling.

Summary updated

BibTeX
@article{liao2022quantitative,
  author = {Kaitao Liao and Yuejun Song and Songhua Xie and Yichen Luo and Quan Liu and Hui Lin},
  title = {{Quantitative Analysis of the Factors Influencing the Spatial Distribution of Benggang Landforms Based on a Geographical Detector}},
  journal = {ISPRS Int. J. Geo Inf.},
  year = {2022},
  doi = {10.3390/ijgi11060337},
}
Figure