Water samples from 13 sites in the Oldonyosambu Ward in Northern Tanzania were collected and comprehensively analyzed for fluoride and other components. Local water resources are typically of sodium carbonate (Na-HCO3) type with low calcium (Ca2+) concentrations. Water sources contain excessive fluoride concentrations and are used as the principal water supply for drinking and cooking by the rural communities. The occurrence of fluoride has severe health impacts in the population from babies (pre-natal) to the elderly, resulting in severe dental and skeletal fluorosis in this area. Purification technologies suitable for such situations in very rough rural communities are scarce due to the absence of electricity supply, infrastructure and operational skills. The development of advanced, yet appropriate technologies is hence a novel separation and purification challenge in that somewhat established technologies need to be adapted to new environments. To this effect two types of experiments were conducted: (1) The water with the highest fluoride concentration (Site 4, 17.9 mg/L, WHO guideline 1.5 mg/L, Tanzanian guideline 4 mg/L) was treated with different nanofiltration (NF) and reverse osmosis (RO) membranes in batch-scale experiments. Results show that fluoride and other components including sodium, potassium, calcium, magnesium, carbonate, sulfate, silica, and organic matter were removed reliably from these natural waters by NF/RO. (2) A pilot-scale autonomous membrane system powered directly by solar energy was used to treat the water with the highest fluoride concentration and examine the impact of fluctuating energy on three different membranes (BW30, NF90, NF270). The results demonstrated that the BW30 and NF90 membranes achieved the WHO guideline, while the NF270 membrane bettered the Tanzanian fluoride guideline. The total permeate volume produced with these membranes was 1000, 2444 and 2656 L, respectively. The resilience of the system to solar fluctuations was good during the short term tests. The results suggest that such advanced membrane technologies powered directly by renewable energy are an attractive option for addressing the water problems in rural areas. While the technology is well established, the adaptation to villages such as those in the Oldonyosambu Ward is far from achieved. This requires further work in terms of integration, technology adaptation and operation and maintenance schemes. (C) 2015 Elsevier B.V. All rights reserved.