Regional probabilistic threats: assessing the impact of glacier melting on mental health using dynamic Bayesian networks

Abstract

Climate change is causing glaciers in Pakistan’s Karakoram, Himalayan, and Hindu Kush mountain ranges to melt at unprecedented rates. This melting significantly affects agriculture, vulnerable populations, and regional water resources, resulting in psychological distress and environmental insecurity. This study explores the effect of glacier melting on mental health and investigates traditional community solutions to psychological resilience. The study models the dynamic Bayesian network to describe the probabilistic interlinks between three nodes: glacier melting (X), mental health concerns (Y), and traditional coping mechanisms (Z). Data were collected through 40 semi-structured interviews with residents of Gilgit-Baltistan, including youths, farmers, elders, and women. The findings reveal that glacier melting triggers significant environmental changes, resulting in anxiety, trauma, and fear of displacement. Despite that, traditional practices such as spiritual rituals, collective rebuilding, and ancestral knowledge play a crucial role in improving mental health.