Understanding persistent groundwater degradation in coastal communities through a participatory socio-hydrological framework

Groundwater is the lifeline for coastal rural communities in southern India, but it faces severe threats from complex interactions among hydrogeological processes, human behaviour and governance responses. This study adopts a socio-hydrological perspective to diagnose groundwater vulnerability in a coastal aquifer system along the Southwest coast of India. An integrated framework combining Participatory Rural Appraisal (PRA), Causal Loop Diagram (CLD), and the Driver-Pressure-State-Impact-Response (DPSIR) model was employed to translate community knowledge into explicit system-level understanding. PRA captured long-term community memory of groundwater change and was transformed into a causal representation of groundwater dynamics, enabling identification of dominant feedback mechanisms. Seasonal groundwater monitoring was used to contextualize community perception with hydrochemical parameters. Results reveal that persistent groundwater degradation is primarily driven by reinforcing feedbacks linking population, groundwater extraction, saline intrusion, and declining water quality. Even though shallow freshwater prevails seasonally, these feedbacks lock the system into a trajectory of increasing dependence on groundwater. The DPSIR synthesis highlights a temporal mismatch between rapidly evolving socio- hydrological pressures and slower institutional responses. This study demonstrates that operationalizing community-driven system knowledge can help in identifying social-ecological feedbacks governing groundwater degradation that are not captured through monitoring-based assessments alone. The proposed participatory socio-hydrological framework offers a potentially applicable approach for diagnosing groundwater stress and identifying leverage points for sustainable groundwater governance, thereby supporting water security, climate resilience, and broader sustainable development objectives in data-limited coastal regions.