Social network shapes farmers’ non-point source pollution governance behavior – A case study in the Lijiang River Basin, China

Abstract

Agricultural non-point source (ANPS) pollution increasingly threatens China’s aquatic ecosystems. Intercropping grain crops (GC) and cash crops (CC) increases landscape fragmentation, making pollution control harder. As producers of ANPS pollution and beneficiaries of its governance, farmers’ participation is crucial for improving water environments. However, the impact of social networks, closely related to farmers’ economic behavior in rural China, on different cropping types of farmers’ ANPS pollution governance behavior remains unclear. Based on survey data from 305 farmers in a typical village in the Lijiang River Basin, social networks of GC and CC farmers were constructed. By introducing network embedding theory, we examined how social networks influence governance resource allocation and collective action among farmers. Combining social network theory with technology acceptance model (TAM), an extended TAM was proposed to discuss the influence of farmers’ social network structural position (SNSP) on their willingness to participate in governance (WP), considering the role of cognition. Results showed that GC farmers’ social network have small-world characteristics (Small-world quotient=2.153) with a network density of only 0.016, lacking bridging ties and showing low trust among actors. CC farmers’ network had a density of 0.029, a clearer core-periphery structure (Core-periphery index=0.267), key farmers showed stronger bridging capabilities with average betweenness centrality of 4.234 %. CC farmers’ networks had diverse ties and higher trust among actors. CC farmers’ network structure improved information diffusion and is more effective in acquiring resources and collaborative governance. Structural equation modeling showed that SNSP positively affect WP for GC and CC farmers, with path coefficients of 0.245 and 0.294. Mediating analysis showed that GC farmers’ perceived usefulness and CC farmers’ perceived ease of use had the largest mediating effects between SNSP and WP, at 20.9 % and 26.8 %, highlighting cognition’s different roles. Social networks variably impact governance behavior among different farmers, and strategies considering these differences can enhance governance efficiency.

Keywords

• Social network analysis
• Resource allocation
• Collaborative governance
• Technology acceptance model
• Mediating effects
• Willingness