The green water components of the water cycle - transpiration, surface soil moisture, and land precipitation - are critical for Earth system stability. However, complex and accelerating human pressures are altering the land surface and water cycle at vast spatial scales. These changes can erode the resilience of green water dynamics locally, and can also be transferred across space. Many approaches to water resilience treat water risk as a largely local problem, and lack tools for assessing how changes to green water dynamics may propagate to affect distant regions. This licentiate thesis addresses both dimensions, applying global quantitative analyses to estimate green water resilience, and the geographies of dependence water creates.
Paper I addresses the local dimension, assessing the resilience of the green water components of the terrestrial water cycle using an early warning signals (EWS) approach. The paper extends existing EWS analyses from vegetation to include key water cycle dynamics, and demonstrates the value of a plural and context-aware approach to resilience indicators that are generally applied in a system-agnostic way. The results show that no single indicator definition of resilience loss is adequate across all regions and variables: where critical slowing down was most informative for transpiration in semi-arid ecosystems, critical speeding up was most informative for precipitation in tropical and high-latitude forests. This suggests that the choice of EWS indicator may need to be conditioned on hydro-climatic and ecological context.
Paper II addresses the propagation dimension, analysing how atmospheric moisture transport and trade-mediated dependencies between countries can transmit and concentrate vulnerability to water resilience risks. By coupling a country-to-country moisture recycling network with a virtual water trade network, the paper traces how upwind hydro-climatic conditions can propagate, through agricultural production and export, to affect distant consumers. The results reveal patterns in which countries are either net structural sources or net sinks of moisture-attributed virtual water flows, and identify an uneven distribution of potential exposure to associated water-related risks.
Together, the papers offer a global diagnostic perspective on where green water resilience may be eroding, and how biophysical and socioeconomic connectivity concentrates and redistributes associated risks. The thesis demonstrates how green water resilience is both a local property of Earth system variables, and a property of connected social-ecological systems. Finally, it situates these contributions within the broader PhD project, which seeks to causally ground moisture transport processes, and interrogate how water resilience risks are constructed and governed in policy.
Stockholm: Stockholm University, 2026.