Introduction

Access to safe drinking water, a fundamental human right and the cornerstone of public health1, remains an elusive goal for a substantial portion of the global population despite decades of concerted efforts. The combined impacts of climate change, rapid population growth, and escalating pollution have transformed this longstanding challenge into an unprecedented crisis, threatening the foundation of sustainable development in the 21st century2. In response, the United Nations established Sustainable Development Goal (SDG) 6.1, targeting universal access to safely managed drinking water (SMDW) by 20303, as a catalyst for achieving broader sustainable development goals4. However, progress has been alarmingly slow-nearly 2.2 billion people still lack SMDW services as of 20205, necessitating a sixfold acceleration in implementation rates to meet the target6. This stark reality prompted the United Nations to launch the Global Acceleration Framework and Blueprint for SDG 67,8,9, emphasizing the urgent need for transformative solutions, with safe drinking water access being a critical priority.

Traditional approaches to providing SMDW, while effective in high-income countries, face insurmountable financial barriers in many parts of the world10. The estimated annual infrastructure and operational costs of $37.6 billion particularly hinder progress in low-income countries, where the need is most acute. This financial constraint, compounded by increasing water stress, calls for innovative, sustainable alternatives to conventional water supply methods11. Rainwater harvesting (RWH) emerges as a promising solution, offering widespread availability and relatively lower contamination risks12,13. The United Nations recognizes RWH as an improved water source (IWS)6,14, and the United Nations Environment Program highlights its critical role in ensuring global drinking water safety and promoting sustainability15.

Driven by the pressing need for safe drinking water supplies, practitioners around the world invested considerable effort into RWH for drinking water production16. Globally, the most prevalent and widely recognized form of RWH was rooftop collection, where precipitation was captured from the roofs of buildings. However, relying solely on rooftop collection presented fundamental limitations that varied considerably across different regional contexts. In arid and semi-arid areas with low-to-moderate rainfall, traditional rooftop harvesting often proved insufficient to meet basic demands, primarily because sparse precipitation required expansive catchment areas to gather adequate volumes for year-round drinking water needs. Additionally, in low- and middle-income regions, roof space constraints were far more severe than commonly recognized, as many households lacked adequate roofing infrastructure or sufficient roof area to fulfill their drinking water requirements; for instance, in rural South Africa, far fewer than 1% of households relied on conventional rooftop RWH setups17. To address these diverse challenges, extensive field implementations and engineering projects worldwide harnessed virtually all land surface types as effective catchments for drinking water production through RWH18. For example, in rural areas of northwest China, where annual precipitation ranged from only 200-300 mm, traditional roofing rainwater collection fell short of basic water demands, prompting local populations to expand collection systems beyond rooftops to include courtyards, road surfaces, and artificially hardened catchment surfaces, and more crucially, to leverage natural landscapes such as hillside slopes and other terrain to maximize every drop of available rainfall for drinking water supply19,20,21,22. Similarly, in resource-poor rural areas of South Africa, where infrastructure remained limited and rooftop collection coverage was minimal, rainwater harvesting catchment surfaces ranged from natural features to engineered structures. The most common surfaces included cleared land, terraces, earth dams, and rock formations, which were frequently directed into households via improvised gutters to provide a potable water supply17,23. In rural India, where seasonal monsoons provided erratic rainfall, populations adapted sand dams and forested areas as supplementary catchments to bolster rooftop limitations and secure potable water supplies24,25. Likewise, in arid rural Kenya, where precipitation was scarce and unevenly distributed, bare rock surfaces and rangelands served as vital collection points for runoff, directly supporting household drinking water needs26. Table S1 provides a detailed summary of these applications. These widespread field practices across diverse geographic and climatic conditions not only revealed the extensive potential of RWH to address safe drinking water challenges, but more importantly, established RWH as a globally proven and versatile drinking water solution that had already been implemented across nearly all available terrain (Table S1).

Despite these widespread efforts and acknowledged benefits, RWH remains underutilized and inadequately evaluated on a global scale. Previous studies primarily focused on regional RWH potential (RWHP) using area-specific methods, which failed to capture the complexities of global hydrological systems12,27,28. Developing standardized, globally applicable methods requires integrating multidimensional natural and social factors, accounting for geographic, climatic, and socio-economic differences. This challenge hinders the accurate quantification of global RWHP and limits cross-regional comparative studies. Additionally, a conceptual gap exists between RWHP metrics and SDG 6.1 indicators, exposing limitations in global drinking water safety assessment frameworks. This indicates the need for methodologies that transform environmental capacity indicators into quantitative indices aligned with SDG 6.1 principles.

In this work, to overcome these limitations, we develop an enhanced parametric approach to quantitatively assess the global potential of RWH in alignment with SDG 6.1 indicators. By refining parameter selection, normalizing processes, and eliminating population density constraints, this method ensures applicability across diverse geographic and climatic conditions. It integrates multidimensional natural and social factors as well as closely aligning with SDG 6.1 principles. Furthermore, by transforming environmental capacity indicators into quantitative metrics that reflect progress toward SMDW coverage, the methodology effectively bridges the gap between RWHP metrics and SDG indicators (Fig. S1). The results suggest that RWH offers a sustainable solution with considerable potential to address global drinking water challenges, especially in regions marked by inequality and high vulnerability. This research provides reliable scientific evidence to guide policymakers in incorporating RWH into comprehensive drinking water security frameworks, critical for alleviating pressures on SDG 6.1 and ensuring equitable access to safe drinking water.

Results

Global rainwater harvesting paradox in drinking water

To explore the complex relationship between RWH and the global supply of safe drinking water, the current state of SMDW coverage worldwide was initially assessed (Fig. 1a) using the latest data from the WHO/UNICEF Joint Monitoring Programme (JMP) for Water Supply, Sanitation, and Hygiene. Additionally, 92.9% of the global population lacking SMDW (over 2.0 billion people) resides in three primary regions: Region A (most Latin American and Caribbean countries, excluding Uruguay, Chile, and Argentina), Region B (all sub-Saharan African countries, Yemen, and Sudan), and Region C (most Asian countries and Pacific Island nations, excluding West Asia). In Region A, 35.3% of the population lacks SMDW, while Region B exhibits the highest proportion of people without access to SMDW, at 72.1%. Region C accounts for the largest absolute number of people without SMDW services, representing 49.0% of the global population lacking access to safe drinking water. These regions are characterized by predominantly rural populations (Fig. S2), making substantial improvements in SMDW coverage through traditional, centralized water supply systems more difficult.

Fig. 1: Global status of safe drinking water supply, rainfall distribution, and rainwater utilization.
Fig. 1: Global status of safe drinking water supply, rainfall distribution, and rainwater utilization.
Full size image

a Proportion of the total population without SMDW in 2020, as reported by the WHO/UNICEF JMP. b Global distribution of precipitation in 2020, sourced from Version 4.05 of the CRUTS monthly high-resolution gridded multivariate climate dataset. c Proportion of the population using rainwater as an IWS in different countries, based primarily on 2020 data from the WHO/UNICEF JMP database (with nearest available values from 2017–2021 substituted for missing 2020 data to ensure temporal consistency and account for survey variability). Maps are produced in ArcGIS 10.8.

The global precipitation distribution was investigated using high-resolution gridded datasets from the Climatic Research Unit Timeseries (CRUTS 4.05)29. Considerable rainfall was observed in areas with low SMDW coverage (Fig. 1b), with 86.4% of countries in these regions receiving over 500 mm of annual rainfall and 78.4% receiving more than 800 mm. Among these three regions, 88.5% of the population lacking safe drinking water were living in areas receiving over 500 mm of annual rainfall, and 78.3% were residing in areas with more than 800 mm. Regions A and C, in particular, experience high levels of rainfall, averaging 1651 mm and 2017 mm annually, respectively. Even in Region B, 77% of countries receive more than 500 mm of rainfall, and 46% exceed 1000 mm. A positive correlation was identified between the proportion of the population lacking SMDW and annual rainfall (Fig. S2), highlighting a clear discrepancy between available water resources and access to safe drinking water.

Although RWH is implemented in 148 out of 208 countries, the proportion of the population relying on rainwater as an IWS remains low, with higher percentages observed primarily in Eastern and South-Eastern Asia and Oceania (Fig. 1c). Globally, over 80% of countries have less than 4.1% of their population using rainwater, and only 10% exceed 15.7% (Fig. S2). The global weighted average is a mere 1.37%, with the lowest reliance on rainwater observed in low-income countries (Fig. S3). In regions with lower SMDW coverage, an inconsistent pattern of rainwater utilization persists, despite abundant rainfall, particularly in Regions A and C (Fig. S3). Region B, composed mainly of low-income countries, shows particularly low levels of rainwater use for drinking purposes. Consequently, although 88.5% of the population lacking SMDW services in these three regions reside in areas that receive more than 500 mm of annual precipitation, the utilization of rainwater as a drinking water source remains remarkably low, with less than 2% of the population harvesting rainwater for drinking, with an average of only 1.26%. These observations reveal a paradox: regions with low SMDW coverage often experience sufficient rainfall but demonstrate minimal reliance on rainwater for drinking. This highlights the considerable potential to expand SMDW coverage through enhanced RWH practices.

The global potential of rainwater harvesting

To evaluate the global potential of RWH, a Geographic Information System (GIS) framework was developed to quantify the RWHP coefficient (RWHPC) globally. The analysis revealed a varied distribution of RWHPC values (Fig. 2a), with higher coefficients concentrated along the equatorial belt, driven by abundant rainfall and favorable climatic conditions, and lower coefficients in regions dominated by tropical deserts and temperate continental climates. Latitudinal analysis revealed a clear trend: RWHPC decreases with increasing distance from the equator and increases with greater rainfall depth. Notably, RWHPC peaks near 15° north and south latitudes, coinciding with subtropical high-pressure belts (Fig. 2b). These areas experience relatively uniform rainfall and substantial precipitation, enabling efficient and stable rainwater collection. In equatorial regions, characterized by consistent rainfall, the coefficient of variation for precipitation is notably low. However, the stability in rainfall patterns leads to frequent flooding, which severely hampers RWH system efficiency and performance. Intense and persistent precipitation can saturate the soil and increase runoff, reducing RWH systems ability to effectively capture and store rainwater30,31. Consequently, despite high theoretical RWHP due to abundant rainfall, flood risks often limit its practical implementation.

Fig. 2: Mapping the potential of rainwater harvesting using a parametric approach.
Fig. 2: Mapping the potential of rainwater harvesting using a parametric approach.
Full size image

a Global distribution of rainwater harvesting potential coefficients, with a zoom-in on Region A, Region B, and Region C. b Latitudinal distribution of averaged RWHPC and rainfall. Data are presented as latitudinal mean values. Shaded bands are the standard deviation intervals of the latitudinal averages, representing the variability within one standard deviation of the latitudinal averages. Maps are produced in ArcGIS 10.8.

Regions with lower SMDW coverage generally exhibit higher RWHPC values. Spatial analysis reveals distinct patterns in the global distribution of RWHPC, marked by pronounced regional variations and geographical dependencies. Region A shows notable RWHP, especially in western Mexico and the Amazon basin, where consistent precipitation and favorable topography create optimal conditions for RWH implementation. In Region B, Sub-Saharan Africa presents a nuanced RWHPC gradient, with higher coefficients in the northern and southern zones, while central areas exhibit lower potential due to frequent flooding in equatorial regions. Region C displays a clear north-south RWHPC gradient, with northern areas, including parts of India, Mongolia, northern China, Pakistan, and Afghanistan, showing considerably higher RWHPC values. Despite generally low and highly seasonal rainfall (resulting in lower active potential), these regions exhibit high passive RWHP. The strong demand for additional water sources makes RWH particularly beneficial, with rainwater contributing substantially to water sourcing. The distinct rainy seasons in these areas enhance the feasibility of efficient rainwater collection and storage, improving the viability of RWH systems in regions where traditional water resources are scarce.

The global potential contribution of RWH to SDG 6.1

To assess the global potential of RWH in achieving SDG 6.1, a spatially explicit methodology was employed to quantify the impact of various RWH implementation levels on SMDW coverage. This analysis considered the maximum harvestable rainwater per grid unit, varying daily drinking water requirements (20 and 40 liters per person per day, LPD)32, and actual population distribution. Seasonal precipitation variability profoundly influenced the efficacy of RWH and its potential contribution to SDG 6.1, amplifying water availability and SMDW coverage during wet periods while constraining them in dry phases, thereby introducing temporal disparities in access. Specifically, monthly surface runoff distributions revealed distinct seasonal patterns across major regions, with precipitation-driven variations from January to December (Fig. S4); monthly per-pixel populations accessing improved water sources via RWH (PIWS) showed comparable seasonal fluctuations (Fig. S5). These patterns collectively drove seasonal shifts in RWH’s global and regional contributions to SDG 6.1, including national SMDW coverage (Fig. S6). To counter seasonal variability as a key limitation on RWH potential for drinking water supply, extending rainwater storage durations served as an effective strategy. For instance, different storage technologies offer varying practical durations33,34. Household-scale rainwater tanks, for example, commonly support 2-4 months of storage depending on household size and usage patterns34,35,36. Community-scale cisterns can extend this to 4-8 months through shared larger capacities37,38. Large-scale RWH ponds or reservoirs enable 8-12 months or more, bridging dry seasons effectively34,39. Accordingly, three Maximum Seasonal Storage Potential (MSSP) conditions were incorporated into the contribution potential estimates, reflecting distinct storage durations: 3 months via household rainwater tanks, 6 months via community-scale facilities or cisterns, and 12 months via ponds or large cisterns.

Another key factor influencing contribution potential was the transition ratio from IWS to SMDW. According to SDG 6.1, SMDW is achieved when an improved water source is accessible on premises, available when needed, and free from contamination. This ratio is calculated as the minimum value among accessibility, availability, and quality, reflecting the limiting factor principle where performance is constrained by the weakest component. This construct provides a realistic assessment of RWH contribution to SDG 6.1 targets under varying local capacities (as detailed in the Methods Section). To account for varying implementation intensities of RWH deployment, four progressive scenarios were constructed to systematically simulate and quantify the potential enhancement of SMDW coverage through RWH, under different levels of implementation intensity. Scenario I involves the comprehensive global deployment of RWH, utilizing current transition ratios. Scenario II targets countries with transition ratios below 0.5, with the aim of raising them to 0.5. Scenario III focuses on nations with transition ratios below 0.7, striving to increase them to 0.7. Scenario IV promotes widespread RWH policies and initiatives, aiming to elevate transition ratios to 0.9 in countries currently below this threshold.

The analysis demonstrated that RWH can substantially enhance global SMDW coverage, with potential contributions ranging from 5.6–26.0% under 20 LPD conditions (Fig. 3) and 4.2–24.3% under 40 LPD (Fig. S7) conditions across different MSSP and transition ratios scenarios. Storage capacity strongly influenced this potential contribution. As the MSSP increased from 3-months to 12-months, RWH contribution potential expanded from 5.6% to 16.4% under Scenario I and from 9.5% to 26.0% under Scenario IV. In terms of equivalent MSSP conditions, improvements in transition ratios also substantially enhanced contribution potential: increasing transition ratios from current levels to 0.9 elevated potential from 5.6% to 9.5% under 3-month MSSP and from 16.4% to 26.0% under 12-months MSSP. These findings highlighted that both extended storage durations and improved transition ratios served as critical determinants of RWH contribution potential, with extended storage mitigating seasonal variability while higher transition ratios enabling greater utilization of harvested rainwater for safe drinking water supply. Additionally, recognizing that these findings hinge on our use of a composite IWS to SMDW transition ratio, we conducted a targeted validation to affirm our approach. As comprehensive, RWH-specific performance data is not yet available globally, we benchmarked our modeled potential contributions against those calculated using field-derived, RWH-specific data from 10 countries40. The comparison revealed a strong concordance between the two sets of results (r > 0.98), confirming that the composite ratio serves as a robust proxy for the national-level factors that govern the success of RWH interventions (Fig. S8).

Fig. 3: Rainwater harvesting contribution to safe drinking water under varying scenarios and conditions.
Fig. 3: Rainwater harvesting contribution to safe drinking water under varying scenarios and conditions.
Full size image

Global spatial distribution of RWH’s potential contribution to SDG 6.1 (percentage increase in safely managed drinking water coverage) across four implementation scenarios (rows: Scenario I–IV with increasing transition ratios from current levels to 0.9) and three maximum seasonal storage potential conditions (columns: 3-, 6-, and 12-month duration). Colors represent potential contribution percentage ranges; global totals are labeled on each map. Gray indicates no data or 100% existing coverage. Produced in ArcGIS 10.8.

Regional analysis revealed distinctive patterns in RWH contribution potential across different MSSP conditions and implementation scenarios. Under 3-months MSSP, regional differences in baseline performance emerge clearly: Region B demonstrated the lowest initial potential contribution under the baseline scenario but the widest contribution range (6.0–19.5% for 20 LPD), while Regions A and C maintained relatively stable ranges of 8.5–11.8% and 6.0–8.8%, respectively. (Table. S2). As MSSP extends from 3-months to 12-months, all regions experience enhanced potential, but with markedly different magnitudes of improvement. Region B demonstrated the most pronounced responsiveness to increased MSSP, with contributions expanding dramatically from 6.0%-19.5% under 3-months to 20.3–61.7% under 12-months MSSP, indicating this region’s high vulnerability to seasonal rainfall variability and correspondingly substantial benefits from enhanced storage capacity. Regions A and C showed more moderate improvements, with Region A progressing from 8.5–11.8% to 15.4–22.5% and Region C from 6.0–8.8% to 20.1–25.1%, reflecting their relatively stable water service foundations that buffer against seasonal fluctuations. The differential MSSP sensitivity across regions revealed that over 82% of countries in Region A and more than 70% in Region C have transition ratios above 0.5, supported by strong economies and established RWH practices. Consequently, as RWH implementation expands from Scenario I to IV, these regions showed relatively limited incremental gains due to already high adoption rates, while Region B experienced the most substantial improvement as implementation progresses across scenarios. This pattern underlined the strategic imperative for differentiated, context-specific interventions that prioritize enhanced storage infrastructure in seasonally vulnerable regions to maximize RWH’s remarkable potential in achieving equitable water security. Similar patterns are observed under 40 LPD conditions (Fig. S7).

Socio-ecological interactions with RWH contribution

To elucidate regional variations and the multifaceted benefits of RWH, the global potential contribution of RWH to SDG 6.1 was analyzed in the context of various socio-ecological factors. The findings suggested that RWH may improve SMDW coverage and could act as a potential catalyst for progress across multiple interconnected SDGs by addressing socio-economic and environmental challenges.

A notable inverse relationship was observed between water productivity (a measure of water use efficiency) and RWH’s potential impact on SMDW coverage (Fig. 4a). Regions with lower water productivity demonstrated greater potential for RWH to enhance SMDW services, offering alternative water sources that alleviate water stress and promote efficient use, thereby potentially contributing to SDG 6.4, which aims to ensure sustainable freshwater withdrawals.

Fig. 4: Interaction relationships between environmental, socio-economic, and health factors and the potential contribution of rainwater harvesting to SDG 6.1 under Scenario IV and 12-months MSSP.
Fig. 4: Interaction relationships between environmental, socio-economic, and health factors and the potential contribution of rainwater harvesting to SDG 6.1 under Scenario IV and 12-months MSSP.
Full size image

a water productivity59. b Gross Domestic Product per capita60. c Proportion of rural population61. d Annual population growth rate62. e Prevalence of moderate or severe food insecurity in the total population63. f Gender Inequality Index64. g Climate change risk index65. h Proportion of deaths attributed to diarrhea among total deaths66. Lines represent linear regression fits. Grey shaded bands represent the 95% confidence intervals of the regression fits. Pearson correlation coefficients (r) and exact P values are displayed in each panel. Sample sizes (n) represent the number of countries included in each analysis. All statistical tests were two-sided. Statistical significance was determined at 0.05.

Additionally, a strong inverse correlation was identified between per capita GDP and RWH’s potential contribution to SDG 6.1 (Fig. 4b). Regions with lower per capita GDP showed a higher potential for RWH to improve SMDW coverage. In countries where per capita GDP is below USD 1000, RWH implementation could increase SMDW coverage by at least 50%. This pattern likely reflects the practical reality that economically constrained regions face greater barriers to establishing centralized water infrastructure, making decentralized, low-cost RWH systems a more viable alternative for achieving water security.

The analysis also revealed that regions with higher rural population ratios and greater population growth rates experienced stronger contributions from RWH to SDG 6.1 (Fig. 4c, d). In regions with substantial rural populations, centralized water supply systems often fail to reach dispersed communities, resulting in persistent water supply gaps41,42. In countries where rural populations exceed 50%, implementing RWH could increase SMDW coverage by over 25%. Furthermore, in areas with annual population growth rates above 2.0%, RWH could boost SMDW coverage by an average of 30%, addressing the growing water demand that may outpace the capacity of traditional systems.

From a social perspective, positive correlations were observed between food insecurity indicators and the Gender Inequality Index with RWH’s potential contribution (Fig. 4e, f). Regions with higher food insecurity and gender inequality demonstrated greater potential for RWH to improve SMDW coverage. This suggests RWH’s ability to help address complex social challenges by providing reliable drinking water access, potentially reducing household productivity constraints, easing women’s water collection burdens43, and creating opportunities for economic development44. By potentially alleviating water-related barriers and decreasing the time spent on water procurement, RWH may create enabling conditions that could contribute to food security (SDG 2) and gender equality (SDG 5).

A positive correlation was also identified between the Climate Change Risk Index and RWH’s potential contribution (Fig. 4g), suggesting RWH’s potential role in climate adaptation strategies. Regions most vulnerable to climate change impacts exhibit a heightened potential for RWH to enhance SMDW coverage. By capturing and storing rainwater during periods of surplus, RWH systems may help communities better cope with dry spells and adapt to climate variability45, potentially contributing to SDG 13 (Climate Action).

The study further reveals RWH’s potential impact on public health through improved SMDW coverage, particularly in regions with high mortality rates due to diarrhea. Analysis of the latest Global Burden of Disease data from the Institute for Health Metrics and Evaluation indicates that regions with higher diarrhea-related mortality rates tend to exhibit greater potential for RWH to contribute to SMDW access (Fig. 4h). By enhancing SMDW coverage, RWH can help reduce the risks associated with unsafe drinking water, which is a major contributor to diarrhea46. The widespread adoption of RWH may therefore contribute to mitigating the risk of diarrhea-related mortality. Using the comparative risk assessment method47, the reduction in annual diarrhea-related deaths from unsafe water sources achievable through SMDW access was estimated via RWH implementation (Fig. S9). The results suggest that enhancing SMDW coverage through RWH could reduce global diarrhea-related deaths by 91.0% in Scenario IV, with reductions of 64.5% in Region A, 86.5% in Region B, and 95.5% in Region C. These results suggest the potential of RWH as a public health intervention, particularly in vulnerable regions, prospectively contributing to SDG 3 (Good health).

In summary, these interactions suggest that RWH holds substantial, context-dependent potential to enhance SMDW coverage and contribute to multiple SDGs. Regions with low SMDW coverage not only face challenges in accessing safe drinking water but also grapple with low economic development, high rural populations, and environmental vulnerabilities. Under optimal conditions, RWH can have a substantial impact in these regions. By improving safe drinking water access, RWH has the potential to contribute to multiple development goals, including food security, gender equality, climate resilience, and public health improvements. A deeper understanding of these socio-ecological dynamics may provide a framework for targeted interventions, potentially enabling policymakers to optimize resources and design more effective, context-specific sustainable development strategies that prioritize high-efficiency RWH in high-vulnerability regions. However, these findings are based on aggregated country-level data, which capture broad, generalizable patterns at the macro level but do not necessarily apply directly to individuals or communities. Moreover, observed correlations do not imply causation, as unaccounted factors could influence results. Future research could employ finer-scale data and causal methods to further explore the interactions between the global potential contribution of RWH and socio-ecological factors.

Discussion

Global research on RWH for drinking water has progressed markedly in recent decades, documenting its technical feasibility, water quality requirements, and implementation challenges in diverse contexts16. However, prior studies often focused on localized assessments, leaving a critical implementation gap and the full potential of RWH for scalable integration with global sustainability goals largely unexplored. Building on this foundation, the present analysis quantifies a striking paradox: 88.5% of people without access to SMDW services reside in areas with over 500 mm of annual rainfall, yet only 1.26% rely on rainwater as their primary source. Quantitative modeling demonstrates that optimized RWH, when embedded in broader water management strategies, could elevate global SMDW coverage from 72.25% to 77.85–98.25%. This shift would provide 20 liters of safe water daily to 0.45–2.08 billion people, accelerating progress toward SDG 6.1.

A key contribution of this study lay in developing a stepwise actionable roadmap that tailored RWH interventions to regional conditions. We identified two synergistic pathways that formed the core of this roadmap. The first pathway involved extending the maximum seasonal storage potential, a metric we developed to assess how scaled-up storage could mitigate rainfall seasonality. For instance, elevating the storage duration from a baseline of 3 months to a more resilient 12 months nearly tripled the potential contribution of RWH in arid or seasonally dry regions by bridging prolonged dry spells. The second, complementary pathway entailed improving transition ratios from harvested rainwater to SMDW, encompassing enhancements in accessibility, availability, and quality to ensure comprehensive adherence to the requirement for SMDW.

A comparison of the pathways revealed their relative impacts. Increasing transition ratios across our scenarios raised the global potential contribution from 5.6% to 9.5%, while extending MSSP from 3 to 12 months elevated it from 5.6% to 16.4%. This disparity suggested that MSSP extension offered greater expansion potential overall, and in certain countries, prioritizing it could yield more pronounced short-term gains in SMDW coverage. Nevertheless, relying solely on one pathway was insufficient, as water quality remained a critical limiting factor. The synergy between the pathways was particularly potent. For example, progressing from Scenario I with 3-month storage (a 5.6% global contribution) to Scenario IV with 12-month storage (26.0%) amplified the potential contribution by approximately 364%, relative to the lowest baseline, far exceeding gains from isolated improvements in either storage or transition ratios alone. This combined effect arises because extended storage provides a stable volume of water that can then be more effectively upgraded to safe standards, thereby enhancing overall efficiency. Such integration not only boosted water security but also fostered community resilience and delivered socioeconomic co-benefits, such as improved health and gender equity.

To implement these pathways, targeted strategies are essential. Enhancing storage capacity began with scaling durations to address rainfall mismatches. This ranged from deploying 3-month household rainwater tanks to 6-month community-scale facilities or cisterns, and larger 12-month catchment ponds or reservoirs, as outlined in our scenarios. These options maintained water integrity through sealed, ventilated designs that minimized evaporation and ensured year-round availability. For improving transition ratios, strategies focused on comprehensive quality management. Most populations lacking SMDW reside in rainfall-abundant regions, and cross-sectional evidence from 27 surveys in low- and middle-income countries demonstrated that rainwater performed well on accessibility and availability dimensions due to its natural characteristics as a decentralized resource accessible at the point of use40. However, for most regions, rainwater quality emerged as the dominant factor determining transition ratio performance, creating the primary bottleneck in RWH effectiveness. Addressing this quality gap, which requires a multi-barrier approach from collection to consumption, was therefore as critical as extending storage for realizing the full potential of RWH. This started with contaminant control during collection, such as using first-flush diversion systems to discard initial, more polluted runoff. Subsequent low-cost, low-energy purification methods were particularly effective in resource-poor settings. These included solar disinfection to inactivate pathogens, gravity-driven membrane filtration for particle removal, and bio-sand or ceramic filters for biological treatment, as well as community-scale systems like natural filtration ponds or other biological treatment setups. These approaches collectively ensured adherence to WHO standards using low-maintenance, sustainable technologies48,49.

Different regions required tailored combinations of these strategies. In seasonally variable zones like Sub-Saharan Africa (e.g., Kenya and Tanzania), investments in 6-month MSSP via durable reservoirs, combined with bio-sand filtration, could substantially increase SMDW access. South Asia (e.g., India and Bangladesh), marked by abundant monsoon rainfall but high contamination risks in densely populated urban-rural interfaces, would benefit from quality-focused interventions like solar disinfection or ceramic filters. Arid regions, including parts of the Middle East and North Africa (e.g., Egypt and Iran), should prioritize MSSP through shaded cisterns paired with affordable purification. In areas with high soil permeability or challenging geological features, such as the sandy soils of the Sahel region or southwestern China’s karst regions, collection systems must be specifically engineered to minimize infiltration losses. For zones with persistently high water quality risks from pollution and flooding, like parts of Southeast Asia (e.g., Cambodia and Laos), multi-barrier strategies combining diversion, extended storage, robust filtration50,51, and community training for maintenance were optimal52. Ultimately, these examples demonstrate the inherent flexibility of the stepwise actionable roadmap. Its power lies not in prescribing a universal solution, but in providing a framework to assemble bespoke combinations of storage and quality-enhancement strategies, precisely tailored to a region’s unique environmental and socioeconomic landscape.

In summary, this study demonstrated that optimized RWH could make a considerable contribution toward SDG 6.1, with the potential contribution of 5.6–26.0% on SDG 6.1 progress. The stepwise, actionable roadmap developed here, anchored in the two synergistic pathways of extending MSSP and improving transition ratios, provides a clear, evidence-based strategy for realizing this potential. While our global framework provides valuable insights into RWH potential worldwide, its broad scope necessitates country-specific adaptations that account for local hydrogeological conditions, policy frameworks, and socioeconomic contexts, which represent important areas for future research. It is also important to recognize that RWH, despite its advantages, was not a panacea. Effective water security strategies will require integrating RWH with conventional water supplies, such as piped networks and groundwater, to create a diversified and resilient portfolio. In certain contexts, this portfolio could be further enhanced by incorporating unconventional sources like desalination, solar distillation, or atmospheric water harvesting.

Methods

Extracting the water access data

This study utilized population coverage data for SMDW and IWS from the JMP dataset (https://washdata.org/data) for the year 2020, updated in July 2023. The JMP consolidates global Water Supply, Sanitation, and Hygiene data from administrative sources, national censuses, and field surveys3,6,53. To analyze the relationship between SMDW and IWS at the national level, the transition ratio was calculated using the following equation:

$${Transition\; ratio}=\frac{{{SMDW}}_{{national}}}{{{IWS}}_{{national}}}$$
(1)

In this equation, SMDWnational and IWSnational represent the national coverage of populations with access to SMDW and IWS, respectively. The proportion of the population with SMDW services was estimated based on three criteria: accessibility on premises, availability when needed, and freedom from contamination. To address missing data, the SDG 6.1 indicator metadata treatment guidelines54 and the estimation method by Jackson et al. 55 were followed, ensuring robust estimates for 208 countries with sufficient data.

Data on the proportion of populations using harvested rainwater as a primary drinking source were extracted from the WHO/UNICEF Joint Monitoring Programme 2022 report country files (https://washdata.org/data/downloads). To ensure maximum temporal consistency with the other 2020-centric datasets used in our model, we established a standardized protocol for selecting this baseline data. Recognizing that JMP survey years vary by country, 2020 estimates for all nations were identified first. For any country where a 2020 datapoint was unavailable, we utilized the most recent, nationally representative estimate within a five-year window (2017-2021). This pragmatic approach ensured that our baseline for the current RWH-using population is as temporally aligned as possible across all nations, providing a robust foundation for our analysis. This process enabled us to quantify the most relevant population coverage of rainwater as an IWS for this study.

Quantifying rainwater harvesting potential distribution

The RWHP is highly influenced by a region’s physical characteristics and the socio-economic context of its population. Previous assessments have struggled with integrating diverse datasets globally, due to the complexity of factors such as climatic conditions, soil properties, hydrological dynamics, and population distribution56. To address these challenges, an enhanced parametric approach was developed, aligning with SDG principles and the SDG 6.1 indicator.

This approach refines parameter selection and normalization processes, eliminating constraints related to population density to ensure broader applicability across varying geographic and climatic conditions. Six key factors are considered: annual precipitation, rainfall coefficient of variation, soil drainage, aridity index, water quantity risk57, and land cover. The first three assess the feasibility of RWHP based on environmental and geological conditions, while the latter three focus on regional demand. This integration of both natural and socio-economic factors extends previous models, offering a more holistic assessment of RWHP (Table S3 for factor effects).

Global raster data for these variables were sourced from international databases, resampled to a consistent 30-arc second resolution, and normalized to a 0-1 scale using updated criteria outlined in Table S3 (Fig. S10). In line with SDG 6.1 indicators, population density was excluded from the RWHP estimation, recognizing the essential role of rainwater as a drinking water source in sparsely populated and remote areas, thus supporting the goal of universal access.

After standardizing the six factors, a multiplicative integration approach using ArcGIS (performed by ArcGIS 10.8) was applied to calculate the RWHPC, a quantitative indicator reflecting the maximum inherent capacity for rainwater collection within a defined unit area. This multiplicative method distinguishes the approach by effectively capturing the interaction between demand and feasibility factors. In alignment with decision science and resource management principles, it accounts for interactions among variables to reflect cumulative impacts on a global scale. Maximum RWHP occurs when both demand and feasibility are high, as their product represents synergistic effects. In contrast, high demand coupled with low feasibility, or vice versa, leads to reduced potential. For example, regions with limited rainfall typically yield lower RWHPC values, while areas with abundant rainfall may still experience reduced RWHPC due to flood risks and other adverse conditions indicated by the land cover factor, thereby lowering overall RWHP.

By fine-tuning the integration of these factors in the methodological framework, a more detailed and accurate evaluation of global RWH potential is achieved. Central to this is the RWHPC, which captures the maximum runoff that can realistically be collected in each region and thus provides a practical limit on potential yields. This not only overcomes earlier shortcomings but also ties directly into SDG 6.1’s aim of fair access to safe drinking water, helping shape effective policies around the world.

Global surface runoff dataset integration

To estimate the contribution potential of RWH to SDG 6.1, accurate quantification of global surface runoff was essential, as it directly influences the volume of harvestable rainwater and thus the potential coverage of SMDW. The methodology recognizes that RWH capacity differs markedly across different catchment surface types, with collection efficiency constrained by the RWHPC, which accounts for surface-specific efficiencies and practical limitations in runoff capture. Traditional approaches for estimating global runoff, such as the Soil Conservation Service Curve Number (SCS-CN) method, often overlook critical physical processes, including inter-event soil infiltration, evapotranspiration dynamics, and seasonal variability between rainfall events, leading to potential inaccuracies in global-scale assessments.

To address these limitations, this study incorporated a 2020 monthly surface runoff dataset at 1-km resolution, provided by the Geographic Remote Sensing Ecological Network platform (www.gisrs.cn). This dataset integrates hydrological monitoring data from regional stations and applies Anusplin software for spatial interpolation of climate data to predict global runoff distributions. Compared to the SCS-CN method, this approach offers distinct advantages in capturing temporal variability, incorporating site-specific empirical observations, and accounting for complex hydrological interactions, thereby providing a more robust foundation for evaluating RWH potential across diverse geographic regions and seasonal periods.

Quantifying population access to IWS through RWH

To assess the contribution of RWH to SDG 6.1, the proportion of the national population with access to IWS through RWH was quantified. Using the RWHPC and direct annual runoff maps, the population capable of accessing IWS per pixel (PIWS) was calculated with the following equation:

$${PIWS}=\frac{A\times Q\times {RWHPC}\times {CE}}{{DRC}\times {MSSP}}$$
(2)

In this equation, A represents the pixel area (1-km resolution); Q denotes the cumulative surface runoff per pixel (mm) over 3-month, 6-month, and 12-month periods, obtained through raster calculations on the global surface runoff dataset; RWHPC is the coefficient of RWH potential per pixel; CE stands for collection efficiency (%); DRC refers to the daily required consumption of drinking water (Liter/person/day, LPD); MSSP, represent the maximum seasonal storage potential, including 3-months, 6-months and 12-months periods (days). Collection efficiency is defined as the percentage of rainwater lost during the harvesting and storage process due to overflow, leakage, and evaporation. Typically, the collection efficiency of a well-designed RWH system can reach 80–90%, with roofing systems achieving relatively higher efficiencies58. Given global disparities in RWH development, the collection coefficient is set at 80%. DRC is defined according to WHO guidelines32, with daily required consumption values of 20 and 40 LPD. For 3-month and 6-month MSSP conditions, based on SDG 6.1 safety managed drinking water standards (accessibility, availability, and quality), the final result of PIWS was determined by the minimum population that could be sustained across all seasonal periods, ensuring continuous year-round water security.

Using the ArcGIS Raster Calculator, PIWS values were computed globally and compared with actual population data from the WorldPoP dataset (2020) at the same resolution (https://doi.org/10.5258/SOTON/WP00647). To ensure accuracy, PIWS values exceeding the actual population in any pixel were adjusted to align with the observed population numbers. Zonal statistics were then applied to aggregate the corrected PIWS and population counts at national and regional levels, enabling the calculation of the percentage of national populations with access to IWS via RWH (IWSRWH):

$${{IWS}}_{{RWH}}=\left(\frac{{Total\; PIWS}}{{Total\; population}}\right)\times 100\%$$
(3)

To prevent double-counting, the existing proportion of the population already utilizing rainwater as IWS, as reported by the WHO/UNICEF JMP in 2020, was subtracted from the calculated IWSRWH for each country. This adjustment ensured that the estimates accurately reflected the additional population benefiting from expanded RWH practices.

By converting environmental capacity indicators into quantitative metrics of IWSRWH, this methodology establishes a foundation for linking these metrics with SMDW, thereby enabling an accurate assessment of RWH’s potential contribution to SDG 6.1. This robust framework facilitates a comprehensive global evaluation of RWH’s capacity to enhance SMDW coverage.

Mapping RWH potential contribution to SDG 6.1

To evaluate the potential contribution of RWH to SDG 6.1, the proportion of the population with potential access to SMDW through RWH (SMDWRWH) was quantified using the following equation:

$${{SMDW}}_{{RWH}}={{IWS}}_{{RWH}}\times {Transition\; ratio}$$
(4)

In this equation, IWSRWH represents the national coverage of the population using harvested rainwater as IWS. In determining SMDW, three factors are considered: accessibility, availability, and quality. The transition ratio, defined as the minimum value among these factors and reflecting the efficiency of converting IWS into SMDW, varies by country based on existing SMDW levels reported by JMP (2017)54. SMDW coverage is calculated by multiplying the transition ratio by IWS coverage. Our analysis identified a global transition ratio of 0.70, with Regions A, B, and C having transition ratios of 0.71, 0.39, and 0.63, respectively. Quality was identified as the primary determinant of the transition ratio globally, while both accessibility and quality were critical in Sub-Saharan Africa. Considering the variability in transition ratios across different countries, four distinct scenarios were developed:

  1. 1.

    Scenario I: Models the maximum attainable SMDW coverage under current conditions via RWH, maintaining existing transition ratios.

  2. 2.

    Scenario II: Targets countries with transition ratios below 0.5 (e.g., Sub-Saharan Africa), enhancing RWH implementation to achieve a minimum transition ratio of 0.5, while leaving countries with transition ratios already at or above 0.5 unchanged.

  3. 3.

    Scenario III: Promotes RWH technologies and infrastructure in nations with transition ratios below 0.7, raising their transition ratio to 0.7, while countries with ratios already at or above 0.7 remain unchanged.

  4. 4.

    Scenario IV: Advocates for global RWH policies and initiatives, aiming to elevate transition ratios to 0.9 in countries where the ratio is below 0.9 while maintaining unchanged ratios in countries already exceeding 0.9.

The potential contribution of RWH to SDG 6.1 at the national level was then calculated using these scenarios. The potential contribution (PCRWHnational) represents the proportion of the population without current access to SMDW that could gain access through RWH, with adjustments made to prevent overestimation:

$${{PCRWH}}_{{national}}=\left\{\begin{array}{c}{{SMDW}}_{{RWH}},{{SMDW}}_{{RWH}}\le 100\%-{{SMDW}}_{{national}} \hfill \\ 100\%-{{SMDW}}_{{national}},{{SMDW}}_{{RWH}} > 100\%-{{SMDW}}_{{national}}\end{array}\right.$$
(5)

This adjustment ensures that the potential contribution does not exceed the portion of the population still lacking access to SMDW.

ArcGIS Raster Calculator was employed to map the proportion of populations in different countries that could achieve SMDW through RWH under each scenario and MSSP condition. This spatially explicit approach integrates RWHP metrics with population distribution data, enabling a precise, context-specific national-level assessment. By converting environmental capacity indicators into quantitative metrics of SDG, the methodology bridges the gap between RWHP and SDG 6.1 indicators, offering policymakers robust scientific evidence to formulate targeted, sustainable water policies.

To address potential biases in using composite transition ratios from IWS to SMDW, which may not fully reflect RWH-specific performance in quality, accessibility, and availability, we conducted a validation analysis using data from Bain et al. (2021)40. Due to global data scarcity for RWH systems in monitoring frameworks like WHO/UNICEF JMP, we extracted RWH-specific quality compliance rates from 10 countries in that study to calculate targeted transition coefficients. Comparing these with our composite method under 20 LPD and 40 LPD standards (6-month MSSP, Scenario I) revealed strong consistency, with no significant differences in 9 of 10 countries (r = 0.9865 and p = 0.00018 for 20 LPD; r = 0.9865 and p = 0.0001 for 40 LPD), as shown in Fig. S8. This alignment validates our approach, as composite ratios proxy national management capacities that can mitigate RWH quality challenges through feasible interventions like improved storage and basic treatment. For assessments in specific regions or countries, we recommend incorporating RWH-specific transition ratios obtained from field surveys or localized studies to enhance accuracy and context-specific applicability. Statistical analyses described above and throughout this study were performed using Origin 2022 (OriginLab Corporation, Northampton, MA, USA).