The CARISMA project (Catchment Risk Assessments using Multi-Scale Data) aims to address issues of data scarcity in catchment water balance assessments through integrating independent sources of information from community-led monitoring (hydrological 'citizen science') and recent developments in remote sensing. The project focusses on two study areas in sub-Saharan Africa, in Ethiopia and Tanzania, where such problems are particularly acute, but aims to develop a more generally applicable methodology. Many catchments or river basins undergo water stress due to a combination of over-abstraction, changes in land-use, or climatic variability and changes. Water stresses may be felt as long-term trends over whole catchment areas, but more often are seen as episodic events such as seasonal or multi-annual drought, and may be spatially located in certain river tributaries or sub-catchments. It is then difficult to identify the underlying causes of water stress, and how particular groups of water users could contribute to potential solutions, if insufficient data are available in the right place and at the right time. Our recent research has demonstrated the viability of community-led monitoring to provide credible key hydrological information to improve understanding of surface and groundwater resources, and that multi-level governance approaches are a feasible way of addressing water management policy. Our work has shown that data-sharing platforms (such as 'Environmental Virtual Observatories' or 'Decision Theatres') have potential to help integrate and present information is ways that support decision making at all levels, but their design needs to be user-driven to facilitate their adoption. These emerging paradigms open up new opportunities for better environmental management, but require participatory development of open and transparent systems for integration of multiple sources of information to provide successful outcomes. This project aims to build on our previous research by co-developing with partners and stakeholders in two catchment in Ethiopia (Abay River Basin Authority) and Tanzania (Rufiji River Basin Authority) a prototype data integration and presentation platform that will quantify key hydrological indicators of catchment water balances at spatial scales appropriate to developing sustainable water management policies and practices in water-stressed catchments. Indicators relevant to stakeholder groups representing community, ecosystem, business, and governance interests will be identified using a participatory approach. The proposed platform uses publicly available remote sensing data for spatial assessments of key hydrological components, particularly rainfall and evapotranspiration, and community-led low-cost monitoring of ground-based variables (including rainfall, river levels and flows, and groundwater levels) to complement available formal monitoring networks. Evaluation of how uncertainty in each component can contribute to overall understanding of water balances will be assessed using a standardised water accounting modelling framework. This can then support better quantification of contributors to catchment water scarcity, to inform multi-stakeholder decision making. Understanding of uncertainty reduction from the different data sources will provide the basis for guidance on appropriate design of monitoring networks, and evidence to support a risk-based approach to water management. The project output of a prototype data platform will provide a key step towards our partner WWF-UK's strategy of working towards a generic capability for developing and sharing better hydrological data to underpin their global activities, particularly through their Water Stewardship Programme. Close involvement of WWF-Tanzania in this project will provide a tangible first step towards this goal.