Community Science
Communities are the first and last to be affected by environmental stresses and pressures mediated by social-economic factors. Empowering communities most dependent on riparian natural resources are integral in developing systems for early warning and action for adaptation, mitigation and resilience.
Crowdsourced Knowledge
Crowdsourcing apps and in-person focus groups with vulnerable populations will be implemented to gather an inclusive response from a broad range of stakeholders in each basin to identify feasible portfolio of NBS and facilitate fund-raising from international agencies and social capital investors to implement the prioritized NBS.
AI-Augmented Sensor Neworks
We will deploy 1000 sensors in 1000 communities across focal river basins. Both satellite and community-managed sensor data will be harnessed to co-design and deploy basin-scale Artificial Intelligence augmented water security early warning early action systems. This monitoring system will generate high frequency sub-hourly data on 40 variables pertaining to assessment of water quantity and quality. This sensor data will be able to objectively and scientifically track access and availability of clean water, which is a key outcome variable for evaluating the success of this project.
Transparent and credible collection of watershed sensor monitoring data, enabled through community ownership and citizen science practices, promotes ecological cooperation and facilitates political compromise.
Early Warning and Early Action Systems (EWEAS)
Early warning and early action systems (EWEAS) provide critical information to anticipate and prevent crises before they occur. Effectively designed EWEAS are integrated with institutional mechanisms at multiple levels of governance to mitigate the impacts of hazards and build resilience against disasters and conflicts.
Particularly noteworthy is the integration of artificial intelligence (AI) technologies in the design of EWEAS. These systems, characterized by high spatial and temporal resolution, are not only automated but also capable of self-learning and enhancing forecast accuracy through the continual cross-validation of forecast data against actual hazard or conflict monitoring data. This evolution marks a significant leap in predictive capabilities for more accurate, reliable, and actionable insights.
