Kitui is a semi-arid and sparsely populated rural county, where low and unreliable rains create water insecurities for fragile cropping and livestock systems. Searching for and fetching water continue to dominate the daily lives of women and children, with households using around four different sources in a year. Rains drive a sharp shift in source choice from groundwater-based handpumps and piped schemes to free surface water sources, risking ill health. This, in turn, decrease revenues for water service providers, jeopardising operation and maintenance services in wet seasons. The Water Diaries reveal different expenditure groups, from those that incur no expenses throughout the year to those that pay more than 10 per cent of their annual expenditures for water. Yet daily consumption remains at only 20 litres per person. Donor investments in water security are fragile and fleeting with devolution transferring a legacy of past failures to newly elected county governments. The results of a professional service delivery model have illustrated how the government and donors can guarantee reliable drinking water services at lower costs, though action and uptake are slow. While hydroclimatic conditions are harsh, weak governance and opaque accountability compound challenges and waste investments.

Policy and investments based on assumptions of rational economic behaviour are often blind to the deeply ingrained social and cultural dispositions that govern choices. For instance, demand-driven ideologies backing community management assume that users will manage and pay for water infrastructure they need. Public awareness campaigns communicate water-related health risks assuming that information will change behaviour. However, extensive evidence across geographies and cultures have proven otherwise. To understand individuals’ and households’ daily water practices and how they vary across different environmental and institutional contexts, we designed and implemented the water diary method in Kenya and Bangladesh. The diaries captured household water source choices and expenditures every day for a whole year, complemented by interdisciplinary analysis of climate, infrastructure, and policy. With global and national monitoring efforts being largely based on aggregate snapshots generated through infrequent surveys, we argue how such granular behavioural dynamics can better inform policy and practice for an equitable water secure future.

Living in coastal Bangladesh is a good working definition of being water insecure. Cyclones and storm surges overwhelm the deltaic floodplains with high salinity in groundwater limiting safe drinking water. Decades of government, donor and household investments have created a portfolio of drinking water technologies – tube wells, pond sand filters, piped schemes, and rainwater harvesting – with varied water quality, costs and maintenance needs. Differences in local hydrogeology, infrastructure gaps, and seasonal variability create inequalities in water availability and cost burdens. Informal vendors source water from distant tube wells and reverse osmosis plants, selling it to places with no other alternatives. The Water Diaries chart households’ daily water source choices, facing uncertain health risks and high-cost burdens. Drawing on this research evidence, a new model for professional service delivery has been piloted in schools and healthcare facilities. Pilot results showed that the SafePani model can achieve water safety and reliability at less than USD 1 per person per year. The government has invested in scaling up the SafePani model through results-based funding, in recognition of the need for institutional and financial reforms for sustainable andsafe rural drinking water services.

With a legacy of sociopolitical and economic exclusion, ethnic conflicts, and an extremely arid climate, the Turkana people struggle to meet the two most basic necessities of life – food and water. One perennial river flowing from Uganda replenishes the groundwater on which the survival and growth of Lodwar depends. Lodwar’s water utility ranks poorly across national regulation metrics delivering an average of 24 litres per person per day to half of the town’s residents. Violent flash floods are known to wash away homes, people and animals almost every year. Relocating to safer grounds in peri-urban areas means losing income opportunities, education and healthcare services, and most importantly, water. A vibrant informal market has emerged, with tanker trucks and motorbikes selling water from the utility’s boreholes at high costs. For those who cannot afford vended water, scooping out dry riverbeds is the only option. With the discovery of crude oil and untapped groundwater reserves, coupled with construction of regional transport links, there is optimism for economic development in this marginalised county. Yet unless there is action to address the fundamental lack of capacity or coordination among multiple stakeholders, Lodwar’s water woes are likely to limit human and economic development.

Dhaka’s rivers are severely polluted by daily discharge of untreated industrial wastewater, coupled with raw sewage, and solid wastes from the city’s 20 million residents. The ready-made garment industry, a key player in the country’s economic growth since the 1990s, significantly contributes to this environmental degradation which disproportionately affects low-income riverbank communities, who rely on the contaminated water for daily activities. We explore these social inequalities through observation of people’s river use behaviour across the dry and wet seasons. In densely populated slums, women and girls turn to the rivers for washing clothes and dishes to avoid long queues at shared water points. Meanwhile, men engage in washing and bathing near bustling marketplaces and boat terminals. During the monsoon, recreational activities like swimming and fishing increase, as the rising water levels create a deceptive appearance of cleanliness. Supported by monthly river quality monitoring, household surveys, and regulatory analysis, these ‘river diaries’ paint the complexities of water-society dynamics. Findings identify priority responses for the most vulnerable today with analysis to guide the sequencing and prioritisation of major investments in water treatment infrastructure in the coming decades.

Despite the contrasts in history and hydroclimatic contexts, the water diaries of Bangladeshis and Kenyans reveal similar daily practices. Rains stand out as the most defining driver of water source choice. Rural populations in Khulna and Kitui shift to rainwater when available, whether harvested in containers from their own roof catchments or in rocks and dams. Whether in sarees or sarongs, in kolshis stacked on the waist or jerrycans balanced on the head, women are the primary drawers of water. When water needs to be transported via motorcycles or boats, a well is dug or a community tube well is installed, men come onto the scene. Individual practices are shaped by institutional behaviours and the quality of water governance. Regulation is missing or ineffective for rural drinking water services in Kitui and Khulna, while non-compliance is normalised in case of urban water pollution in Dhaka and unreliable piped supply in Lodwar. Our findings propose that policy and practice focus more attention on the interactions between rainfall and water use behaviours in a changing climate, and the need for better information on water risks for institutional accountability and sustainable finance. We finally chart where change is happening to improve water security in Bangladesh and the opportunities that exist in Kenya.