Reengineering of sand dams to reduce water scarcity and meet Millennium Development Goals

There are 748 million people in the world who do not have access to improved sources of drinking water and 43% of these live in sub Saharan Africa (WHO and UNICEF 2012). They face long walks to fetch water, during which time they are not in school or earning money. The water they drink may be so contaminated it makes them ill. There are lots of technologies that can bring clean water close to people's homes. One option, particularly in Kenya, is sand dams. Sand dams are impermeable concrete structures constructed across seasonal rivers in order to trap both water and sediment (sand) behind them during rain storms. The water is stored in the spaces between the sand grains for abstraction using a well during the dry season. With no standing water, there are fewer threats from water-borne vectors like mosquitoes (Hut et al 2008, Lasage et al 2008).

However, there are to date no studies on the quality of water removed from sand dams. There is an assumption that the water quality is protected by the sand, but this has not been tested. We know that pathogens are removed by biological processes when water is passed through clean sand - this principle of slow sand filtration (SSF) is used in conventional water treatment. So in this research we are testing the hypothesis that water in a sand dam is not only protected from contamination, but its quality is improved as it passes through the sand. This hypothesis will be tested through a combined programme of field measurements, laboratory experiments and computer modelling. If sand dam water quality can be demonstrated to be of acceptable microbiological quality then it can be drunk safely without any further treatment and it is a more attractive water supply option to communities and agencies serving them.

There are principle concerns about how well sand dams will remove contaminants:
1. We know that more pathogens are removed as water passes through fine sand than coarse sand. However, sand dams trap coarse sand.
2. The biological processes that remove pathogens in engineered SSFs happen mainly in the upper 2cm of the sand. However, this layer in sand dams is disturbed during the rain storms so the biological processes may not work effectively.
3. During rain events the water trapped by the dam has high turbidity. This means it has a lot of fine particles. These may change biological processes.
Overall, through this research we will comprehensively measure the quality of water trapped behind sand dams, determine whether grain size or turbidity affect water quality and thus propose how sand dams can be better designed to maximise water quality.

The field measurements will be supported by Excellent Development, a UK based charity who have built 325 sand dams in Kenya since 1985. We will randomly select three of their sand dams and install piezometers after the seasonal rain storms. A piezometer is a tube with holes in it such that the water level in the tube reflects the water level in the trapped sand. Water levels will be measured daily. Samples for water quality testing will be collected from these tubes weekly. A range of water quality parameters will be measured including coliforms, which are the bacteria that cause diarrhoea, and turbidity.
The next stage will be to do lab experiments that replicate conditions within sand dams from the end of the flood events. Columns will be filled with sand of different grain sizes, and then saturated with water with similar coliform counts and turbidity to that measured in the field. The coliform count of water removed gradually from the bottom of the column over a four month period will be measured to establish the efficiency of pathogen removal.

The final stage will be to create computer models of the sand dam system using the results of the lab and field work. These can be used to understand the effects of factors like grain size, sand dam size and input water quality so that recommendations on sand dam design can be made.

The major impact of this project will be to inform development practice by improving the scientific understanding of sand dams so as to provide donors with the confidence to invest in this low cost technology and thereby improve welfare. Sand dams are the cheapest way to store rainwater in low income rural areas (Batchelor et al 2011) but there is no evidence to demonstrate drinking water quality. By providing the evidence base for cheaper technologies donor's funds can stretch further, or communities will be able to afford to invest in improving their own water supplies and realise the benefits.

These benefits include less time spent collecting water. Foster and Tuinhof (2004) found that sand dams reduced household water collection time from 5 hours per day to 1 hour per day averaged over 200 000 households in the Kitui region. This means more time can be dedicated to education and income generating activities, which helps release households from the cycle of poverty. These benefits are particularly experienced by women and children who are the principal water collectors.

A further benefit of a clean, local water supply is increased water consumption by households. Lasage et al (2008) found that sand dams increased domestic water use by 50%. This has been widely demonstrated (e.g. Curtis et al 1995) to improve hygiene.

It is estimated that they catch only 2-3% of rainfall (Hut et al 2008, Ertsen and Hut 2009), so the ultimate result is that communities have a local water supply but the impact on downstream water users is minimal.