While the impact of climate change is exacerbating issues of pollution, land degradation, poor ecosystem services, and unsustainable use of water resources across the Africa Continent, the need for more water infrastructure is becoming more eminent. Lack of water infrastructure investment coupled with slow uptake of No/Low regret investment projects is gradually inhibiting the continent’s economic development. This situation calls for a better planning of new water infrastructure and an efficient management of existing ones. This is easier said than done especially where most reservoirs are not (properly) gauged. The small reservoirs project (www.smallreservoirs.org) has shown over the past decade that it is possible to monitor the surface extent of water bodies from space (using satellite imagery) and to combine this with the areas covered by the water and their volumes using a field correlation developed from bathymetric surveys (Annor et al., 2009, Liebe et al., 2005). The area alone explained over 98% of the variance in the Area-Volume correlation giving a strong indication that monitoring stored water in reservoirs by just their surface areas from space was doable and quite accurate. Long time series of both optical (landsat, spot, rapideye, quickbird) and radar (ASAR, ESR-1/2, Radarsat-2) images have been used for these analyses with all showing very promising results. Recently it has been shown that a combination of both optical and radar satellite images yield optimal results especially with clouds affecting optical images in the rainy season, and radar images sometimes requiring more polarizations modes to reduce the effect of Bragg scattering. This algorithm was developed using a Bayesian classification system (Eilander et al., 2014) which creates more room for seamless integration of in-situ data and satellite data for improved classification of water bodies for flood mapping, runoff estimation and water balance assessment. The surface area-volume correlation was first developed in 2002 and re-evaluated after 10 years. It turned out that the correlation has not been affected much by sedimentation at least for the Volta basin. This cannot be said to be true for all basins, which points to the continued need for a dedicated effort of using in situ data combined with new observation systems, such as the Sentinels, to develop and update these correlations. We show in this research that the sentinels are a great addition to the satellite constellations available for near-real-time water monitoring combined with cost-effective TAHMO ground stations for operational water management in the Volta basin to support decision making in agriculture to alleviate poverty.

Keywords