Dynamic Changes in Agroecosystem Landscape Patterns and Their Driving Mechanisms in Karst Mountainous Areas of Southwest China: The Case of Central Guizhou

Limin Yu, Yangbing Li, Meng Yu, Mei Chen, Linyu YangSustainability 2023 · DOI

TL;DR

This study investigates the dynamic changes in agroecosystem landscape patterns in the karst mountainous region of Central Guizhou, China, from 2004 to 2020. It reveals a trend towards increased complexity and diversification, driven primarily by socio-economic factors, with distinct transformation patterns emerging across different topographic gradients.

What Problem Does This Paper Address?

Karst mountainous areas, particularly in Southwest China, are ecologically vulnerable and face significant land degradation challenges. Agroecosystems (AESs) in these regions are undergoing rapid transformation due to urbanization, socio-economic development, and policy interventions. Understanding the trajectory and driving forces behind these changes is crucial for promoting sustainable land use and ecological restoration. This paper addresses the need to analyze the dynamic changes in AES landscape patterns and their underlying mechanisms in these sensitive environments, moving beyond simply looking at agricultural output to consider the integrated social and ecological systems.

Methods

The researchers employed a multi-faceted approach:

  • Remote Sensing and GIS: High-resolution satellite imagery from 2004, 2015, and 2020 was used to classify land use types (woodland/grassland, slope cultivated land, construction land, economic/fruit forests, etc.).
  • Landscape Pattern Indices: Standard indices like Landscape Percentage (PLAND), Patch Density (PD), Edge Density (ED), Shape Index (LSI), Contagion Index (CONTAG), and Shannon's Diversity Index (SHDI) were calculated to quantify landscape characteristics and changes.
  • Moving Window Method: This technique was used to analyze landscape index values at different spatial scales, helping to determine the optimal granularity for analysis.
  • Geodetector Model: This statistical tool was employed to identify and quantify the influence of various natural (slope, elevation, lithology, soil organic matter, landform) and socio-economic (resident population, urbanization rate, population sex ratio) factors on AES transformation.

Datasets

  • Remote Sensing Data: ASTER images (15m resolution), SPOT images (2.5m resolution), and Google Earth high-resolution images (0.54m resolution) for the years 2004, 2015, and 2020.
  • Digital Elevation Model (DEM): ALOS-12m DEM data (12.5m resolution).
  • Socio-economic and Population Data: Obtained from the China County Statistical Yearbook (Township Volume) and the Anshun Statistical Yearbook.

Key Insights and Ideas

  • Karst AESs are Dynamic and Complex: The study highlights that AESs in karst regions are not static but are undergoing significant, complex transformations driven by a interplay of natural and human factors.
  • Topography as a Key Differentiator: The research clearly demonstrates how topographic gradients (high mountains vs. valleys/hills) lead to distinct AES transformation patterns and dominant functions. This is a crucial insight for targeted land management.
  • Socio-economic Factors Dominate: While natural factors play a role, the study strongly emphasizes that socio-economic drivers, particularly urbanization and population dynamics, are the primary forces shaping AES transformation in these areas.
  • Emergence of Eco-Economic Landscapes: The shift from traditional food crop-focused agriculture towards economic and ecological forests (EFF) signifies a move towards more sustainable and diversified land use practices.

Key Results

  • Overall Trend: From 2004 to 2020, the AES landscape became more complex and diversified. Woodland and grassland (WGL) increased, slope cultivated land (SCL) decreased, construction land (CL) expanded, and economic and fruit forests (EFF) significantly increased.
  • Three Typical Transformation Patterns:
    • WGL Restoration Type: Dominant in middle and high mountains, characterized by increased ecological functions.
  • CL Growth Type: Dominant in low mountain areas, characterized by enhanced living functions.
  • EFF Growth Type: Dominant in valleys and hills, characterized by enhanced ecological and economic functions.
  • Driving Factors:
    • WGL restoration is primarily driven by slope and elevation.
  • CL growth is influenced by slope and urbanization rate.
  • EFF growth is influenced by soil organic matter, slope, urbanization rate, and resident population.
  • Interaction Effects: The study found that interactions between driving factors (e.g., slope and elevation) often have a greater explanatory power than individual factors.

Limitations and Future Work

  • Image Resolution Inconsistency: The authors acknowledge that the resolution of remote sensing images used was inconsistent but state that the use of high-resolution data with small resolution gaps did not significantly affect their findings.
  • Classification of Abandoned Land: Abandoned land was not classified individually but was included within WGL as ecological land. Future research could further subdivide land types and explore de-agriculturalization in more detail.
  • Rural-Urban Distinction: The study focused on mountainous rural areas. Future research could further distinguish between rural and urban areas and strengthen the guidance of research results for rural development.
  • Further Subdividing Land Types: The authors suggest future research could further subdivide land types and explore the de-agriculturalization of agricultural land in depth from different perspectives.
  • Focus on Rural Areas: Future work could also differentiate between rural and urban areas more distinctly and focus research on mountainous rural areas to strengthen the practical applicability of findings.

Summary updated

Abstract

Puding County in central Guizhou is a typical karst ecologically vulnerable area integrating rural, mountainous, and ecological migration areas. It is essential to study the dynamic trajectory and direction of its agroecosystems (AESs) transformation to promote sustainable land use development in karst mountain areas. Based on high-resolution remote sensing images from 2004 to 2020, this study analyzes the transformation characteristics, typical landscape patterns, and their driving forces of AESs using the landscape pattern index, moving window method, and Geodetector model. The results show that: (i) The overall landscape pattern of AESs in the study area developed towards complexity and diversification from 2004 to 2020. The overall trend of woodland and grassland (WGL) is increasing, the slope cultivated land (SCL) is shrinking, the construction land (CL) is spreading and expanding along the vicinity of the town, and the economic and fruit forests (EFF) are increasing significantly. (ii) Three typical patterns are concluded according to the main transformation directions of AESs: WGL restoration type, CL growth type, and EFF growth type; middle and high mountains are dominated by the WGL restoration type with increased ecological functions, and the concentration of the new WGL increases with the increase in elevation; low mountain areas are dominated by the CL growth type with enhanced living functions, and the concentration of the new CL decreases with the increase in slope and elevation; valleys and hills are dominated by the EFF growth type with enhanced ecological and economic functions, and the concentration of the new EFF increases with the decrease in slope and elevation. (iii) Socio-economic factors are the dominant drivers of AES transformation. The WGL restoration type is dominated by slope and elevation, the CL growth type is mainly influenced by slope and urbanization rate, and the EFF growth type is primarily influenced by soil organic matter, slope, urbanization rate, and resident population. The study results have implications for rural land use, rural development, and ecological restoration of rocky desertification in karst mountain areas and other similar mountainous areas.
BibTeX
@inproceedings{yu2023dynamic,
  author = {Limin Yu and Yangbing Li and Meng Yu and Mei Chen and Linyu Yang},
  title = {{Dynamic Changes in Agroecosystem Landscape Patterns and Their Driving Mechanisms in Karst Mountainous Areas of Southwest China: The Case of Central Guizhou}},
  booktitle = {Sustainability},
  year = {2023},
  doi = {10.3390/su15129160},
}
Figure