Patrick Thomson

Reducing risks to rural water security in Africa.

With increasing hydro-climatic risks in Africa all water resources will be affected (Vorosmarty et al. 2010; IPCC 2012). But the response to climate variability of groundwater will be slower than that of surface water. Groundwater resources are more resilient due to aquifer storage which acts as a buffer during periods of little or no rainfall. Despite the significant availability of groundwater resources across the major geological formations in Africa, the quantitative and qualitative distribution remains partly understood (MacDonald et al. 2012). Adapting to environmental change will likely require increasing use of groundwater for both domestic and productive uses. Information costs in recording longitudinal data on aquifer recharge and use are prohibitively high and point estimates from a limited sample of verifiable drilling sites provides partial information on whether shifting patterns of groundwater availability are attributable to environmental or operational factors. Sustainable use of aquifers is critical to inform and enforce appropriate policy responses to achieve rural water security for human development and environmental resilience in Africa (Bannerjee and Morella 2011; IPCC 2012). Smart handpumps that harness Africa’s expanding mobile network architecture provide a new technology to address the systemic informational deficit that limits accountability and increases risks of investments in and management of rural water supply (Thomson et al. 2012). Automatically measuring and transmitting low cost, reliable and scalable data on handpump performance allows three inter-connecting risks to rural water security to be simultaneously addressed: (a) environmental, (b) operational, and (c) governance. By explicitly linking groundwater resource risks with operational and governance risks the overlapping objectives of the Sustainable Development Goals and Millennium Development Goals (MDG) can be linked to the United Nations’ General Assembly recognition of the Human Right to Drinking Water and Sanitation.

Is silence golden? Of mobiles, monitoring, and rural water supplies

Reliable and cost-effective monitoring of rural water supply infrastructure has long been hampered by the geographical curse of dispersed and low-income populations, and weak institutional performance. Recent advances in monitoring technology combined with mobile network expansion into rural areas has created an opportunity to bypass these seemingly intractable challenges. Mobile-enhanced technologies have the potential to produce data that is orders of magnitude richer, faster, and cheaper than that provided by traditional monitoring methods, which require costly field visits. However, more data does not equate to better data; information generated by crowd-sourced and automated systems each has its respective limitations. We propose a framework for analysing monitoring and surveillance systems, which can help assess the strengths and weaknesses of different emerging approaches. We suggest that these advancements present an opportunity to fundamentally change the way we consider and conduct rural water s...

Risk factors associated with rural water supply failure: A 30-year retrospective study of handpumps on the south coast of Kenya

An improved understanding of failure risks for water supplies in rural sub-Saharan Africa will be critical to achieving the global goal of safe water for all by 2030. In the absence of longitudinal biophysical and operational data, investigations into water point failure risk factors have to date been limited to cross-sectional research designs. This retrospective cohort study applies survival analysis to identify factors that predict failure risks for handpumps installed on boreholes along the south coast of Kenya from the 1980s. The analysis is based on a unique dataset linking attributes of >300 water points at the time of installation with their operational lifespan over the following decades. Cox proportional hazards and accelerated failure time models suggest water point failure risks are higher and lifespans are shorter when water supplied is more saline, static water level is deeper, and groundwater is pumped from an unconsolidated sand aquifer. The risk of failure also appears to grow as distance to spare part suppliers increases. To bolster the sustainability of rural water services and ensure no community is left behind, post-construction support mechanisms will need to mitigate heterogeneous environmental and geographical challenges. Further studies are needed to better understand the causal pathways that underlie these risk factors in order to inform policies and practices that ensure water services are sustained even where unfavourable conditions prevail.

Accidental infrastructure for groundwater monitoring in Africa

A data deficit in shallow groundwater monitoring in Africa exists despite one million handpumps being used by 200 million people every day. Recent advances with smart handpumps have provided accelerometry data sent automatically by SMS from transmitters inserted in handles to estimate hourly water usage. Exploiting the high-frequency noise in handpump accelerometry data, we model high-rate wave forms using robust machine learning techniques sensitive to the subtle interaction between pumping action and groundwater depth. We compare three methods for representing accelerometry data (wavelets, splines, Gaussian processes) with two systems for estimating groundwater depth (support vector regression, Gaussian process regression), and apply three systems to evaluate the results (held-out periods, held-out recordings, balanced datasets). Results indicate that the method using splines and support vector regression provides the lowest overall errors. We discuss further testing and the potential of using Africas accidental infrastructure to harmonise groundwater monitoring systems with rural water-security goals. A data deficit exists in shallow groundwater monitoring in Africa.Our smart handpump has low-cost accelerometers mounted in the handle.We show that machine learning methods applied to the accelerometry can estimate aquifer depth.We demonstrate that we can use the accidental infrastructure of handpumps for estimating groundwater levels.

Tryptophan-like fluorescence as a measure of microbial contamination risk in groundwater.

Microbial water quality is frequently assessed with a risk indicator approach that relies on Escherichia coli. Relying exclusively on E. coli is limiting, particularly in low-resource settings, and we argue that risk assessments could be improved by a complementary parameter, tryptophan-like fluorescence (TLF). Over two campaigns (June 2016 and March 2017) we sampled 37 water points in rural Kwale County, Kenya for TLF, E. coli and thermotolerant coliforms (total n = 1082). Using three World Health Organization defined classes (very high, high, and low/intermediate), risk indicated by TLF was not significantly different from risk indicated by E. coli (p = 0.85). However, the TLF and E. coli risk classifications did show disagreement, with TLF indicating higher risk for 14% of samples and lower risk for 13% of samples. Comparisons of duplicate/replicate results demonstrated that precision is higher for TLF (average relative percent difference of duplicates = 14%) compared to culture-based methods (average RPD of duplicates ≥ 26%). Additionally, TLF sampling is more practical because it requires less time and resources. Precision and practicality make TLF well-suited to high-frequency sampling in low resource contexts. Interpretation and interference challenges are minimised when TLF is measured in groundwaters, which typically have low dissolved organic carbon, relatively consistent temperature, negligible turbidity and pH between 5 and 8. TLF cannot be used as a proxy for E. coli on an individual sample basis, but it can add value to groundwater risk assessments by improving prioritization of sampling and by increasing understanding of spatiotemporal variability.

Mobilizing Payments for Water Service Sustainability

The current model of rural water service delivery is broken. Money flows down from donors and governments to install infrastructure but little reliable information on performance flows back. Increased use of handpump mapping exercises by survey teams may usefully identify handpumps working one day of the year but this leaves the remaining 99.7 % of any year unknown. The continuous monitoring of services is increasingly important if we are to know the real level of water services being enjoyed by rural communities, with the growing consensus on the Human Right to Water adding further impetus. For governments and donors, knowing whether investments deliver verifiable impacts over time rather than simply knowing that budgets have been spent, is transforming established thinking. Mobile networks provide an inclusive architecture to reduce the information asymmetry between investments and outcomes. Information alone is insufficient to make progress but it is necessary to track and improve accountable service delivery. Information can improve institutional performance and help define appropriate roles and responsibilities between communities, governments and donors to close the loop between well-meaning investments and quantifiable outcomes. Donors can demonstrate value-for-money, government and water service regulators can align performance with measureable outcomes, and communities can contribute to financial sustainability through user payments that are contingent upon service delivery. Using unique observational data from monitoring handpump usage in rural Kenya, we evaluate how dramatic improvements in maintenance services influence payment preferences across institutional, operational and geographic factors. Public goods theory is applied to examine new institutional forms of handpump management. Results reveal steps to enhance rural water supply sustainability by pooling maintenance and financial risks at scale supported by advances in monitoring and payment technologies.