Integrating In-Situ and Earth Observation Data to Support Understanding of Functional Water Constraints in Small Reservoirs in Ghana

Small reservoirs are a rapidly expanding form of water infrastructure across sub-Saharan Africa, supporting irrigation, livestock watering, fishing, aquaculture, and domestic water supply. These systems are locally governed and highly multifunctional. However, they are rarely subject to regular hydrological or water quality monitoring due to their small size and large numbers. Earth observation (EO) provides a unique opportunity to complement ground data for systematic reservoir assessment across space and time. A previous EO-based study using a Sentinel-2 time series (2018–2024) identified 3,079 small reservoirs in northern Ghana with widespread vulnerability to seasonal drying¹. Understanding when and why reservoirs become functionally constrained requires an integrated perspective with information on water availability, but also on water quality and patterns of use, which motivates this research.

In a first step, measurements of turbidity, reflecting light availability as a relevant indicator of water quality, were collected across 103 small reservoirs in northern Ghana in December 2025. These data were analyzed together with information from a detailed reservoir user survey conducted by the International Water Management Institute (IWMI), and vulnerability to drying¹. Hierarchical cluster analysis showed three distinct types: 1. Small, moderate vulnerability to drying, high turbidity, mixed irrigation; 2. Large, low vulnerability to drying, low turbidity, fully irrigated; 3. Medium, low vulnerability to drying, moderate turbidity, non-irrigated. Across all reservoirs, turbidity was negatively correlated with reservoir size and positively associated with vulnerability to drying.

In a second step, Sentinel-2-derived turbidity estimates using the C2RCC processor² were validated using satellite-in-situ match-ups within a ±5-day window. The analysis focused on the dry season to capture early dry-season sediment accumulation following rainfall and late dry-season conditions shaped by aeolian inputs, while minimizing cloud contamination. The resulting turbidity time series (2017–2025) enabled scaling the analysis across space and time, supporting regional comparisons of quantity-quality-use interactions.

This study demonstrates how integrating in-situ observations and EO-derived indicators can support the understanding of functional water constraints in small reservoirs. By jointly considering feedbacks between water quantity, quality, and use, the approach reveals patterns that are not visible from single-variable assessments. While limitations remain, particularly regarding attribution of observed values to specific drivers or management decisions, the framework provides a scalable basis for interpreting vulnerability and emerging risk in small, human-managed water systems. It thus contributes to improved monitoring strategies for data-scarce environments and offers a foundation for informed, locally relevant water management under climatic and socio-economic pressures.

¹Siabi, Ebenezer K.; Akpoti, Komlavi; Zwart, Sander J. 2023. Small reservoirs in the northern regions of Ghana and their vulnerability to drying. Colombo, Sri Lanka: International Water Management Institute (IWMI). CGIAR Initiative on Aquatic Foods. 37p.

²Brockman, C., Doerffer, R., Peters, M., Stelzer, K., Embacher, S., & Ruescas, A. (2016). Evolution of the C2RCC Neural Network For Sentinel 2 and 3 for the Retrieval of Ocean Colour Products in Normal and Extreme Optically Complex Waters. Living Planet Symposium, Prague, Czech Republic.