Nicco Plamonia*
| Rizki Arizal Purnama | Yeni Novitasari | Elshedevika Rosya | Heru Mulyono | Rizki Firmansyah | Wahyu Purwanta | Dwi Budiyanto Trisnoharjono | Ahmad Nuridha | Indri Mardiyana | Suwarti Wati | Agus Setiawan | Susilo Raharjo | Revina Devitani Putri
© 2026 The authors. This article is published by IIETA and is licensed under the CC BY 4.0 license (http://creativecommons.org/licenses/by/4.0/).
OPEN ACCESS
Small inhabited islands in archipelagic states experience persistent drinking water insecurity driven by structural hydrogeological constraints, extreme scale limitations, and institutional mismatch with centralized utility systems. In Indonesia, thousands of micro-islands host long-established communities but lack permanent freshwater due to limited catchment areas, saline groundwater intrusion, and high evapotranspiration. Despite small populations, households face disproportionately high water costs through reliance on rainwater harvesting, purchased water, and transported supplies. This study analyses drinking water provision across very small inhabited islands in Indonesia using a comparative mixed-evidence approach cantered on secondary datasets with limited field verification. Physical island characteristics, household water access conditions, cost structures, and governance arrangements are comparatively assessed across islands from Indonesia’s five major island regions. All micro-island cases (≤ 200 inhabitants) remain below 15 m³/day, while two comparator islands exceed this threshold. These demand levels remain substantially below the operational scale typically associated with centralized water utility systems. Household water prices range from Rp300–600 per liter, equivalent to 50–200 times mainland tariffs, with water expenditures accounting for approximately 10–20% of monthly household spending. Indicative capital–operational cost analysis shows that although per-capita capital costs of micro-scale systems are relatively high, total investment requirements remain modest in absolute terms, while operating costs remain lower than prevailing purchased-water prices. Across all cases, centralized on-grid systems and population relocation are empirically unviable. Scale-appropriate off-grid systems-including micro reverse osmosis units, rainwater harvesting with basic disinfection, and hybrid configurations-emerge as the most technically feasible, economically rational, and socially equitable approaches for small-island drinking water provision. The findings indicate that drinking water insecurity on very small inhabited islands is fundamentally a structural and scale-sensitive problem requiring island-category-based planning, decentralized system design, and sustained public investment rather than uniform utility expansion.
small islands, drinking water supply, off-grid systems, scale mismatch, archipelagic states
Indonesia comprises approximately 17,000–17,500 islands, of which about 5,000–6,000 are permanently inhabited [1, 2]. Most inhabited islands fall into the micro and small categories, characterized by limited land area, narrow catchments, shallow aquifers, and high vulnerability to seawater intrusion. Empirical evidence from Miangas Island shows that freshwater availability is constrained primarily by hydrogeological conditions, particularly high evapotranspiration and extremely limited catchment capacity, rather than by rainfall levels. Despite these constraints and relatively small populations, many of these islands sustain long-established livelihood systems, particularly small-scale fisheries, making water security both an economic and equity concern.
Within Indonesia’s national drinking water system, centralized utility services remain concentrated in mainland and urban growth centres, leaving most small islands outside piped network coverage. In practice, communities rely on rainwater harvesting, shallow saline groundwater, transported water, or high-cost bottled water. Structural limitations-limited raw water sources, geographic fragmentation, and insufficient demand to achieve economies of scale-restrict the feasibility of conventional centralized systems.
Approximately 2,000–3,000 islands are inhabited by fewer than 200 people, while only 100–200 islands exceed 10,000 inhabitants [1, 2]. Based on this distribution, micro islands cumulatively host an estimated 0.2–0.6 million people, and small islands (200–1,000 inhabitants) approximately 0.6–1.6 million people. In total, around 0.8–2.2 million people reside on micro and small islands, representing a demographically modest yet nationally significant population segment that falls structurally outside the technical and economic design envelope of centralized utility-based systems (see Table 1).
Table 1. Distribution of Indonesian islands by population size and raw water availability constraints
|
Island Category |
Population |
Estimated Number of Islands |
Key Characteristics |
Technical Implication |
|
Uninhabited |
0 |
± 11,000–11,500 |
No permanent settlement |
No water demand |
|
Micro |
≤ 200 |
± 2,000–3,000 |
Very small settlements, no permanent freshwater |
No local raw water |
|
Small |
200–1,000 |
± 1,500–2,000 |
Limited groundwater, high intrusion |
Severely limited raw water |
|
Medium |
1,000–10,000 |
± 600–800 |
Permanent settlements, limited services |
Limited local sources |
|
Large |
> 10,000 |
± 100–200 |
Urban-like islands, coastal towns |
Regional raw water available |
Although decentralized options-such as rainwater harvesting and micro-scale desalination (< 5 m³/day)—are generally more suitable for settlements below 300–500 inhabitants [3-6], a comprehensive framework integrating demographic thresholds, hydrogeological constraints, and institutional feasibility remains underdeveloped.
Addressing this gap, this study develops a scale-sensitive drinking water supply framework for very small inhabited islands in Indonesia by identifying structural freshwater constraints, examining demographic and institutional limits of centralized utilities, and formulating governance and technological pathways for equitable off-grid provision. In doing so, the study repositions island water planning from infrastructure expansion toward threshold-based, scale-appropriate system design in archipelagic states.
The materials and methods used in this study are organized into eight interrelated subsections. Section 2.1 outlines the research design and analytical approach; Section 2.2 presents the analytical perspective; Section 2.3 describes case selection and study areas; Section 2.4 details data sources and materials; Section 2.5 defines analytical dimensions and measurement; Section 2.6 introduces the evaluative framework; Section 2.7 explains the analytical methods; and Section 2.8 presents the comparative analytical strategy. Together, these subsections establish a coherent framework for comparative assessment of drinking water conditions and scale-appropriate supply strategies on very small inhabited islands.
2.1 Research design and analytical approach
This study adopts a descriptive–analytical design based on secondary policy [7, 8] and infrastructure data verification to examine drinking water access challenges on very small inhabited islands and to identify scale-appropriate supply strategies. The analysis applies comparative pattern assessment and cross-case comparison to examine recurring relationships between island physical and demographic conditions, freshwater limitations, household water access conditions, and drinking water supply strategies. The study emphasizes comparative interpretation and policy relevance rather than formal statistical inference [9-11].
2.2 Analytical perspective
This study examines drinking water provision on very small islands through a comparative policy–infrastructure perspective focused on structural constraints, limited demand scale, and household water access conditions. Rather than testing a formal causal theory, the analysis identifies recurring cross-case patterns linking hydrogeological limitations, small population size, high water costs, and dependence on non-networked water supply systems. The framework is used to assess the suitability of different drinking water strategies under small-island conditions.
2.2.1 Structural hydrogeological and spatial constraints
Very small islands are characterized by limited land area, narrow catchments, shallow aquifers, high evapotranspiration rates, and strong vulnerability to seawater intrusion [12-14]. As a result, freshwater scarcity in these environments is primarily hydrogeological rather than meteorological.
These conditions restrict the availability of reliable freshwater sources and increase dependence on rainwater harvesting, transported water, or small-scale desalination systems. Geographic fragmentation and low population density also limit the feasibility of centralized network-based water supply systems [3].
Small islands are widely recognized as highly vulnerable freshwater environments because groundwater resources are typically limited to thin freshwater lenses that are sensitive to seawater intrusion, prolonged drought, and climate variability. These hydrogeological limitations have been documented across Pacific and Caribbean island systems and are considered one of the principal constraints to long-term water security in island environments [13, 15-17].
2.2.2 Scale mismatch in centralized utility design
Centralized water supply systems are generally designed for larger and more concentrated populations that can support network infrastructure and economies of scale. Centralized water systems generally operate more efficiently when serving larger and spatially concentrated populations.
However, in very small island settlements, demand volumes are often too low to sustain conventional utility-based systems efficiently. Under these conditions, infrastructure costs remain high relative to output, while geographic isolation further constrains operational feasibility.
2.2.3 Water costs and demand scale
Water supply systems on small islands frequently operate at very low production volumes. While decentralized systems may function under small-demand conditions, centralized utility systems generally require larger demand aggregation to distribute infrastructure and operational costs efficiently [18, 19]. In many very small island settlements, limited freshwater availability and low demand magnitude are associated with relatively high unit water costs compared with mainland systems.
Previous studies have shown that water supply systems serving small and geographically isolated populations frequently experience diseconomies of scale, resulting in higher unit production costs and reduced operational efficiency compared with mainland systems. Under such conditions, decentralized and modular systems often provide more economically viable service arrangements [20-22].
2.2.4 Household water costs and socio-economic implications
Households on very small islands commonly rely on purchased water, transported water, or rainwater harvesting systems.
These conditions are associated with relatively high household water expenditures compared with mainland utility users [23, 24].
High water prices may reduce affordability and increase household vulnerability, particularly in low-income island communities. Household water access patterns, therefore, provide an important indicator for evaluating the economic implications of small-island drinking water systems.
2.2.5 Appropriate drinking water strategies
The recurring conditions observed across very small islands indicate that drinking water provision requires approaches compatible with local physical and demographic conditions. Uniform infrastructure models designed for mainland or urban systems may not align with the operational realities of very small islands.
Under these conditions, decentralized and scale-appropriate systems-including rainwater harvesting, micro-scale desalination, and hybrid off-grid configurations—represent important alternatives for improving drinking water access in archipelagic settings.
A growing body of evidence suggests that rainwater harvesting, small-scale desalination, and hybrid water supply systems offer practical and resilient solutions for remote island communities where conventional utility expansion is technically difficult and economically inefficient. Such approaches have been increasingly promoted as key components of water security strategies for Small Island Developing States (SIDS) and other archipelagic regions.
2.3 Case selection and study area
This study adopts a comparative case study approach to examine freshwater supply challenges on very small inhabited islands in Indonesia. The selected cases are intended to represent typical hydrogeological and socio-technical conditions faced by islands with limited population size, constrained natural freshwater resources, and the absence of formal water utility services. Emphasis is placed on ensuring both geographic representativeness and data availability to support cross-regional analysis.
2.3.1 Selection criteria
Fifteen very small inhabited islands were selected as illustrative case studies based on the following criteria: (1) small permanent population of fifteen inhabited islands was selected, consisting of thirteen micro islands (≤ 200 inhabitants) and two upper-bound comparator islands representing larger small-island conditions. Comparator islands were included to assess how water demand and system feasibility evolve beyond the micro-island threshold; (2) absence of centralized water utilities–based supply; (3) geographic representation of Indonesia’s five major island groups; (4) availability of secondary data from government reports or previous studies. These criteria ensure that the selected islands share comparable scale-related constraints while capturing regional variability in physical and institutional settings.
Population ranges are based on regional statistical records and secondary sources. Permanent freshwater availability refers to the presence of reliable, non-saline surface or groundwater sources. PDAM responsibility is administrative; none of the islands are served by operational on-grid utility systems. For each selected island, data were compiled on geographic location, population range, freshwater availability, dominant household water source, and administrative responsibility for drinking water provision. These variables were used to operationalize structural constraints and household water access patterns at the island level.
2.3.2 Case distribution
Based on the above criteria, the selected islands are evenly distributed across Indonesia’s five major island groups to avoid regional bias and to reflect diverse environmental and demographic conditions. The spatial distribution of the fifteen case study islands across Indonesia is presented in Figure 1, while their demographic and water access characteristics are summarized in Table 2.
Figure 1. Spatial distribution of fifteen small inhabited islands across five major island regions of Indonesia
Table 2. Distribution of case study islands
|
Five Major Island Regions |
Island |
Approx. Latitude |
Approx. Longitude |
Province |
Population Estimate |
Dominant Water Source |
Responsible Water Utility (PDAM) |
|
Sumatra |
Tailana |
2.229148° N |
97.222697° E |
Aceh |
200–250 |
Rainwater harvesting |
PDAM Aceh Singkil Regency |
|
Salah Namo |
3.342222° N |
99.722778° E |
North Sumatra |
~ 100–300 (est.) |
Transported water |
PDAM Langkat Regency |
|
|
Berhala Kecil |
−0.857462° S |
104.405733° E |
Jambi |
~ 80–250 (est.) |
Rainwater harvesting |
PDAM East Tanjung Jabung Regency |
|
|
Java |
Pamujan Kecil |
−5.964167° S |
106.193056° E |
Banten |
~ 90–180 (est.) |
Purchased water |
PDAM Serang Regency |
|
Bidadari |
−7.250315° S |
113.726014° E |
East Java |
~ 80–200 (est.) |
Brackish shallow wells |
PDAM Sumenep Regency |
|
|
Gili Raja Kecil |
−7.218441° S |
113.784107° E |
East Java |
~ 90–190 (est.) |
Rainwater harvesting |
PDAM Sumenep Regency |
|
|
Kalimantan |
Nunukan Island |
4.053292° N |
117.666724° E |
North Kalimantan |
~ 80–170 (est.) |
Purchased water |
PDAM Nunukan Regency |
|
Maratua Kecil |
2.211147° N |
118.622775° E |
East Kalimantan |
~ 70–150 (est.) |
Rainwater harvesting |
PDAM Berau Regency |
|
|
Pulau Manti |
−3.519212° N |
116.376819° E |
South Kalimantan |
~ 60–140 (est.) |
Transported water |
PDAM Kotabaru Regency |
|
|
Sulawesi |
Langkai |
−5.031662° S |
119.094481° E |
South Sulawesi |
~ 90–150 (est.) |
Rainwater harvesting |
PDAM Makassar City |
|
Kulambing |
−4.786100° S |
119.431138° E |
South Sulawesi |
~ 80–160 (est.) |
Brackish shallow wells |
PDAM Pangkep Regency |
|
|
Gangga |
−1.767410° S |
125.054326° E |
North Sulawesi |
~80–160 (est.) |
Purchased water |
PDAM North Minahasa Regency |
|
|
Papua & Maluku |
Owi |
−1.241729° N |
136.207663° E |
Papua |
300–450 |
Rainwater harvesting |
PDAM Biak Numfor |
|
Ur |
-0.847911° S |
132.536052° E |
Maluku |
745–4783 |
Transported water |
PDAM Kota Tual |
|
|
Numfor Kecil |
−1.005270° S |
134.848952° E |
Papua |
~ 100–180 (est.) |
Shallow wells |
PDAM Biak Numfor Regency |
2.4 Data sources and materials
This study employs a combination of secondary and limited primary data to ensure both breadth and contextual depth in analysing freshwater supply conditions on very small islands. Secondary data provide national-scale consistency, while primary data from selected locations offer field-based validation of documented conditions.
2.4.1 Primary field verification, sampling design, and measurement period
This study is primarily based on secondary national datasets (BIG, BPS, BRIN, PUPR, 2019–2024); limited field verification was conducted between (1) July 2019–September 2019 (Sumatra cases); (2) August 2021–October 2021 (Sulawesi cases); (3) May 2023–July 2023 (follow-up validation and price confirmation).
These visits were conducted to validate household water consumption levels, price structures, and seasonal supply patterns documented in secondary sources.
Sampling Design. Across the 15 islands, household interviews were conducted using structured questionnaires. Sampling was purposive but proportionally distributed to reflect settlement size. Total estimated households across 15 islands: ~ 420–480 households. Sampled households: (1) 8–15 households per island; (2) Total verified households: 132 households; (3) Estimated coverage: 25–32% of households on micro islands (≤ 200 inhabitants). Household size assumptions (cross-checked with BPS 2019–2023 district data): Average household size: 4–5 persons per household.
Consumption Measurement Method. Water consumption data were collected using three complementary approaches: (1) Self-reported daily consumption (L/household/day); (2) Volumetric estimation based on container counts (20 L jerrycans; 200 L drums); and (3) Refill frequency tracking (per week). Since no island had functioning water meters, standardized per capita consumption (L/cap/day) was calculated as (see Eq. (1)):
$L_{\text {cap} / \text {day}}=\frac{Q_h}{N_h}$ (1)
where, (1) $L_{\text {cap} / \text {day}}$ = per capita water consumption (L/capita/day); (2) $Q_h$ = total household water consumption per day (L / day); (3) $N_h$ = number of household members (persons).
Reported values were cross-checked against physical storage capacity observed during visits.
Per Capita and Total Demand Calculation. Standardized per capita consumption (Lcap/day) was derived from household-level volumetric estimation. Total island-level water demand was then calculated using:
$Q_d=\frac{P \times C_p}{1000}$ (2)
where, (1) $Q_d=$ total daily water demand $\left(m^3 / d a y\right)$; (2) $P=$ island population (persons); (3) $C_p=$ per capita water consumption $\left(L_{\text {cap } / \text { day }}\right)$; (4) $1000=$ conversion factor from liters to cubic meters $\left(1 \mathrm{~m}^3=1000 \mathrm{~L}\right)$. Population refers to the most recent district-level demographic estimate (2019-2023).
Seasonal Framing. Respondents were asked to distinguish between: (1) Wet season (November–April); (2) Dry season (May–October). All water demand values as presented later in the results section reflect: Dry-season effective consumption, representing the most constrained operational condition.
Water Price Data Collection. Water prices were documented during: (1) Field interviews (2019, 2021, 2023); (2) Vendor transaction observation; (3) District government records (2019–2024). Prices were standardized into: (1) Rp per liter; (2) Rp per m³ equivalent. Mainland PDAM tariffs used for comparison were taken from District tariff schedules (2022–2024).
Scope of Water Demand Calculation. Water demand calculations include: Household domestic use (drinking, cooking, washing, sanitation), and exclude: (1) Institutional demand (schools, mosques, public buildings); (2) Commercial demand (fish landing, ice production); (3) System losses (non-revenue water). Institutional demand was separately estimated (see Results section) but not included in per capita demand calculations.
2.4.2 Secondary data sources
The study exclusively uses secondary data obtained from: (1) National island inventories (Badan Informasi Geospatial-BIG); (2) Population statistics (BPS); (3) Technical reports from BRIN and the Ministry of Public Works and Housing (PUPR); (4) Peer-reviewed journal articles indexed in Scopus and Web of Science; (5) WHO guidelines on small drinking water systems. In addition, primary data were collected through field visits to several very small inhabited islands in Sumatra and Sulawesi. These visits provided qualitative and observational information on existing water sources, storage practices, and operational conditions, which were used to corroborate and contextualize the secondary data.
2.4.3 Materials used
The materials and data inputs employed in this study are summarized in Table 3, including spatial datasets, technical reports, economic information, and policy documents relevant to small-island water supply systems.
Table 3. Materials and data inputs
|
Material |
Description |
Purpose |
|
Island inventory datasets |
Island size, location, population |
Structural classification |
|
Water supply reports |
Existing water sources & systems |
Access pattern analysis |
|
Cost data |
Water prices, capital expenditure, and operational expenditures |
Comparative cost analysis |
|
Policy documents |
National & regional regulations |
Policy synthesis |
2.5 Analytical dimensions and measurement
This study applies a parsimonious analytical structure appropriate for comparative policy and infrastructure assessment. Analytical dimensions are defined to capture recurring physical, economic, and operational conditions affecting drinking water provision on very small inhabited islands. The framework focuses on structural conditions, household water access conditions, and drinking water access outcomes relevant to comparative evaluation.
2.5.1 Structural conditions
Structural constraints of small islands are defined as natural and demographic conditions that inherently limit both the availability of raw freshwater resources and the technical–economic feasibility of centralized water supply systems. These constraints are considered exogenous and relatively fixed, particularly for very small inhabited islands (see Table 4).
Table 4. Operational indicators of structural constraints of small islands
|
No. |
Indicator |
Operational Description |
Data Source |
|
1 |
Island size and catchment limitation |
Classification of islands based on land area and effective catchment capacity (small vs. very small islands) |
National Island Inventories (BIG) |
|
2 |
Groundwater availability |
Presence, absence, or salinity condition of groundwater resources |
BRIN–PUPR technical reports; peer-reviewed literature |
|
3 |
Rainfall dependency |
Degree of reliance on rainfall as the primary or sole freshwater source |
BRIN–PUPR technical reports; peer-reviewed literature |
|
4 |
Population size category |
Population classification (≤ 200; 200–1,000; 1,000–10,000; > 10,000 inhabitants) |
BPS population data |
2.5.2 Household water access conditions
Household water access pattern is defined as the predominant strategies adopted by households to obtain freshwater in response to the structural constraints of small islands, including limitations in natural water availability and the absence of centralized water supply systems (see Table 5).
Table 5. Operational indicators of household water access patterns
|
No. |
Indicator |
Operational Description |
Data Sources |
|
1 |
Dominant water source |
Main source of water used by households (rainwater harvesting, purchased water, or transported water) |
Published case studies; regional government reports |
|
2 |
Relative water price |
Household water cost relative to mainland utility tariffs (lower, comparable, or higher) |
Published case studies; empirical findings from previous island water studies |
2.5.3 Drinking water access conditions
The level of drinking water access is defined as the extent to which households on small islands achieve safe, affordable, and sustainable access to drinking water, reflecting service coverage, economic affordability, and system reliability (see Table 6).
Table 6. Operational indicators of drinking water access level
|
No. |
Indicator |
Operational Description |
Data Sources |
|
1 |
Access to safe drinking water |
Percentage of households with access to drinking water meeting safety standards |
Published case studies; regional government reports |
|
2 |
Relative affordability |
Household water expenditure as a proportion of total household income |
Published case studies; empirical findings from island water studies |
|
3 |
System reliability |
Continuity of water availability across seasons (non-seasonal vs. seasonal access) |
Published case studies; regional government reports |
2.5.4 Outcome (Output indicators)
The outcome indicators aligned with the research objectives, translating key analytical dimensions into measurable variables. Table 7 clarifies how physical, economic, institutional, and access-related factors interact to explain drinking water outcomes on very small inhabited islands.
Table 7. Outcome indicators aligned with research objectives
|
Research Objective |
Outcome Dimension |
Output Indicator (Measurable) |
Unit / Proxy Data |
Relevance |
|
Objective 1 |
Physical feasibility |
Absolute daily water demand |
m³/day (≤ 20–30 for micro islands) |
Indicates demand scale condition |
|
Characterize structural constraints |
Resource availability |
Presence/absence of permanent freshwater |
Binary (yes/no) |
Indicates freshwater limitation |
|
Objective 2 |
Economic burden |
Household water price ratio |
Ratio vs mainland utility (50–200×) |
Indicates relative household burden |
|
Examine utility limitations |
Affordability |
Share of water expenditure |
% of household income (10–20%) |
Indicates affordability condition |
|
Service reach |
Coverage by utilities-based systems |
% islands served (≈ 0% for ≤ 200) |
Indicates service limitation |
|
|
Objective 3 |
Access outcome |
Share of households with safe water |
% households |
Main drinking water outcome |
|
Develop scale-appropriate framework |
System reliability |
Non-seasonal availability |
Months/year |
Indicates operational continuity |
|
Sustainability |
OPEX vs household ability to pay |
Rp/m³ vs income |
Indicates sustainability consideration |
2.5.5 Evaluative criteria
To systematically assess the suitability of drinking water supply strategies for small islands, this study applies a set of evaluative criteria encompassing technical, economic, institutional, and social dimensions, as summarized in Table 8.
Table 8. Evaluative criteria for assessing drinking water strategies on small islands
|
Criterion |
Definition |
Evaluation Question |
Threshold / Reference |
|
Technical viability |
Compatibility with island physical conditions |
Can the system operate without permanent freshwater? |
Must function with seawater/rainwater only |
|
Scale compatibility |
Match between system capacity and demand |
Is demand within minimum viable scale? |
≤ 30 m³/day → off-grid only |
|
Economic sustainability |
Balance between CAPEX–OPEX and affordability |
Is water cheaper than purchased water? |
OPEX < Rp20,000/m³ |
|
Equity of access |
Fairness compared to mainland households |
Does it reduce price ratio? |
< 10× mainland tariff |
|
Institutional fit |
Alignment with operator capacity |
Can it be run outside utilities? |
Community/BUMDes feasible |
|
Resilience |
Climate and seasonal robustness |
Can it operate during dry months? |
≥ 10–12 months/year |
2.6 Evaluative framework
This study applies an evaluative framework to compare drinking water conditions and supply strategies across very small inhabited islands. The framework focuses on recurring structural characteristics, including freshwater availability, population scale, household water access conditions, and operational suitability of different supply systems. Rather than applying formal explanatory modelling, the framework is used to organize comparative assessment across cases and to evaluate the compatibility of centralized and decentralized drinking water approaches under small-island conditions. Table 9 maps key analytical dimensions to measurable indicators and evaluative criteria used throughout the study.
An evaluative structure is required to assess where and how policy intervention can effectively modify this pathway. Accordingly, the causal logic of the conceptual model is translated into an evaluative flow chart (see Table 9) and evaluative sequences, which serve as the analytical backbone for outcome assessment and policy comparison.
Table 9. Maps key analytical dimensions to measurable indicators and evaluative criteria used throughout the study
|
Analytical Dimension |
Indicator |
Evaluative Criterion |
|
Structural constraints |
Absence of freshwater |
Technical viability |
|
Household water conditions |
Water price ratio |
Economic sustainability |
|
Socio-economic conditions |
Water expenditure share |
Affordability |
|
Supply strategy |
Technology selected |
Scale compatibility |
|
Drinking water outcome |
Safe water access |
Reliability & resilience |
2.7 Analytical methods
This study applies a qualitative–comparative analytical approach consistent with its mixed-evidence design. The analysis focuses on identifying recurring structural patterns and policy-relevant implications across the selected islands rather than formal statistical inference. The analytical process consists of: (1) descriptive profiling of island physical and demographic conditions; (2) comparison of household water access conditions and freshwater limitations; (3) comparative assessment of household water prices and indicative CAPEX–OPEX conditions; and (4) synthesis of scale-appropriate drinking water supply strategies. Cross-case consistency and data triangulation are used to strengthen analytical interpretation.
2.8 Comparative analytical strategy
The analysis applies descriptive comparison and cross-case assessment to identify recurring structural patterns across the selected islands. Emphasis is placed on consistency between population scale, freshwater availability, household water access conditions, and the suitability of different drinking water supply strategies. The study focuses on comparative interpretation and policy relevance rather than formal statistical inference.
3.1 Results
3.1.1 Characteristics of the 15 case study islands
A total of 15 inhabited islands were analyzed across Indonesia’s five major island regions. A detailed summary of island characteristics, estimated water demand, dominant water sources, water prices, and freshwater availability is presented in Table 10.
Table 10. Characteristics of case study islands and water access conditions
|
Island |
Province |
Population |
Dominant Source |
Estimated Demand (m³/day)* |
Water Price (Rp/L) |
Price Ratio vs. Mainland |
Responsible Water Utility (PDAM) |
Permanent Freshwater |
|
Tailana |
Aceh |
200–250 |
Rainwater |
10–14 |
300–500 |
50–120× |
PDAM Aceh Singkil Regency |
No |
|
Salah Namo |
North Sumatra |
100–300 |
Transported |
5–12 |
400–600 |
80–150× |
PDAM Langkat Regency |
No |
|
Berhala Kecil |
Jambi |
80–250 |
Rainwater |
4–10 |
300–500 |
50–120× |
PDAM East Tanjung Jabung Regency |
No |
|
Pamujan Kecil |
Banten |
90–180 |
Bottled |
4–8 |
400–600 |
80–150× |
PDAM Serang Regency |
No |
|
Bidadari |
East Java |
80–200 |
Brackish wells |
4–12 |
300–500 |
50–120× |
PDAM Sumenep Regency |
No |
|
Gili Raja Kecil |
East Java |
90–190 |
Rainwater |
5–13 |
300–500 |
50–120× |
PDAM Sumenep Regency |
No |
|
Nunukan Island |
North Kalimantan |
80–170 |
Purchased |
4–10 |
400–600 |
80–150× |
PDAM Nunukan Regency |
No |
|
Maratua Kecil |
East Kalimantan |
70–150 |
Rainwater |
3–8 |
300–500 |
50–120× |
PDAM Berau Regency |
No |
|
Pulau Manti |
South Kalimantan |
60–140 |
Transported |
3–7 |
400–600 |
80–150× |
PDAM Kotabaru Regency |
No |
|
Langkai |
South Sulawesi |
90–150 |
Rainwater |
5–8 |
300–500 |
50–120× |
PDAM Makassar City |
No |
|
Kulambing |
South Sulawesi |
80–160 |
Brackish wells |
4–10 |
300–500 |
50–120× |
PDAM Pangkep Regency |
No |
|
Gangga |
North Sulawesi |
80–160 |
Purchased |
4–9 |
400–600 |
80–150× |
PDAM North Minahasa Regency |
No |
|
Owi |
Papua |
300–450 |
Rainwater |
24–40 |
300–500 |
50–120× |
PDAM Biak Numfor |
No |
|
Ur |
Maluku |
745–783 |
Transported |
45–62 |
400–600 |
80–150× |
PDAM Kota Tual |
No |
|
Numfor Kecil |
Papua |
100–180 |
Shallow wells |
5–13 |
300–500 |
50–120× |
PDAM Biak Numfor Regency |
No |
While most cases fall within the micro-island category (≤ 200 inhabitants), two islands (Owi and Ur) represent upper-bound small island conditions to enable scale comparison.
These islands generally lack permanent freshwater sources and rely primarily on rainwater harvesting or purchased water. Water prices on small islands range from Rp 300 to Rp 600 per liter, compared to mainland PDAM tariffs of approximately Rp 3–6 per liter, resulting in households on small islands paying 50–200 times higher prices. For low-income households, water expenditures can account for 10–20% of total monthly household spending.
3.1.2 Observed water consumption (2019–2023)
The observed consumption ranges and scale characteristics across island categories are summarized in Table 11.
Table 11. Observed consumption and scale characteristics
|
Category |
Population Range |
Observed L/Cap/Day |
Absolute Demand Range |
Institutional Share |
On-Grid System |
|
Micro |
≤ 200 |
45–70 |
2.5–14 m³/day |
< 8% |
None |
|
Small |
200–1,000 |
60–80 |
12–80 m³/day |
< 10% |
None |
|
Medium |
1,000–10,000 |
80–100 |
80–1,000 m³/day |
Moderate |
Limited |
|
Large |
> 10,000 |
100–120 |
> 1,000 m³/day |
Significant |
Present |
Across 132 validated households (2019–2023): (1) Micro islands (≤ 200 inhabitants): 45–65 L/cap/day; (2) Upper-bound micro (~ 180–200 inhabitants): 60–70 L/cap/day; (3) No island exceeded 80 L/cap/day during the dry season.
All micro-island cases remain below 15 m³/day.
Institutional water demand across micro islands was minimal. Observed elementary schools (20–60 students) and small prayer facilities contributed an estimated additional 1–2 m³/day, representing less than 8–10% of total island demand. This increment does not materially alter the scale classification of micro islands (< 15 m³/day).
3.1.3 Seasonal variation (Observed pattern)
As summarized in Table 12, seasonal variations affect water consumption patterns, purchased water dependency, refill frequency, and water prices across all study islands. No island achieved year-round freshwater sufficiency without supplementation.
Table 12. Seasonal dynamics of water access
|
Indicator |
Wet Season |
Dry Season |
|
Per capita consumption |
+ 10–15% |
Baseline |
|
Purchased water share |
20–40% |
60–70% |
|
Refill frequency |
Moderate |
+ 30–50% |
|
Price variation |
Baseline |
+ 10–15% |
3.1.4 Verified water prices (2019–2024)
As shown in Table 13, water prices on micro islands remain substantially higher than mainland PDAM tariffs. The 50–200× ratio is consistently observed across all micro-island cases.
Table 13. Water price comparison
|
Indicator |
Micro Islands |
Mainland PDAM |
|
Retail price (Rp/L) |
300–600 |
3–6 |
|
Equivalent (Rp/m³) |
300,000–600,000 |
3,000–6,000 |
|
Expenditure share |
10–20% income |
< 3% income |
3.1.5 Comparative CAPEX–OPEX across drinking water supply strategies
Table 14 summarizes the indicative capital and operating cost ranges associated with different drinking water supply strategies across island population scales.
Table 14. Indicative capital and operating cost ranges
|
Strategy |
Scale |
CAPEX (Rp/Person) |
OPEX (Rp/m³) |
|
Micro RO |
≤ 200 |
8–15 million |
10,000–20,000 |
|
RWH + UV |
≤ 200 |
3–7 million |
< 5,000 |
|
Hybrid RWH–RO |
200–1,000 |
6–10 million |
8,000–15,000 |
|
Off-grid SPAM |
1,000–10,000 |
4–8 million |
5,000–10,000 |
|
On-grid |
> 10,000 |
2–5 million |
< 5,000 |
3.2 Discussion
The analysis results indicate that water scarcity on small islands is structural in nature and cannot be resolved through on-grid approaches or population relocation. The most rational solution is island-scale off-grid systems, such as micro-scale reverse osmosis (RO) and integrated rainwater harvesting, managed at the community level with support from state policies.
3.2.1 Structural constraint and demand magnitude
Empirical findings show that all micro-island cases (≤ 200 inhabitants) exhibit absolute domestic water demand below 15 m³/day, absence of permanent freshwater sources, and dependence on rainwater, transported water, or saline groundwater. These three variables—demand magnitude, raw water availability, and supply dependence—form a consistent structural pattern across geographically dispersed islands.
The critical feature is not merely water scarcity, but the interaction between hydrogeological limitation and extremely small demand volume. Even where rainfall is present, limited catchment area and shallow aquifers prevent stable freshwater accumulation. Thus, water insecurity is not seasonal variability alone, but structural hydrological constraint combined with minimal demand aggregation.
3.2.2 Scale threshold and system feasibility
Observed demand volumes (< 15 m³/day for micro islands) are substantially below the production scales typically associated with centralized utility systems. Centralized systems rely on: (1) Continuous raw water abstraction; (2) Aggregated demand; (3) Network-based distribution; (4) Capital amortization over large customer bases.
Where production volume remains permanently small, fixed infrastructure costs are distributed over extremely low output. Under such conditions, centralized expansion does not generate efficiency gains.
The empirical evidence therefore indicates that micro islands operate below the functional scale at which conventional utility models become viable. This mismatch is structural and persistent. This finding is consistent with previous studies on small-island water systems, which demonstrate that low population density, geographic isolation, limited freshwater availability, and high infrastructure costs constrain the economic viability of centralized water supply networks. Consequently, decentralized approaches such as rainwater harvesting, small-scale desalination, and community-managed systems are frequently identified as the most appropriate solutions for remote island settlements [18].
Similar observations have been reported in other island settings, where low demand volumes, geographic isolation, and limited resource availability constrain the viability of centralized water supply systems. In such contexts, decentralized and community-scale systems are frequently identified as more appropriate service models due to their flexibility and lower infrastructure requirements [25-27].
3.2.3 Cost structure and household burden
Water price data reveal a consistent ratio of 50–200× relative to mainland tariffs, with household expenditures reaching 10–20% of monthly income. This pattern reflects the economic consequence of small-scale supply fragmentation.
Indicative CAPEX–OPEX analysis shows that while per-capita capital costs of micro systems are higher than large-scale systems, total absolute investment remains modest due to small population size. Moreover, operational costs of modular systems are substantially lower than prevailing purchased-water prices.
This implies that replacing informal purchased-water dependence with structured off-grid systems would reduce household economic burden despite higher per-capita capital cost.
3.2.4 Governance alignment with scale conditions
Although all islands fall under administrative responsibility of local utilities, none of the micro-island cases is operated through centralized on-grid systems. Operational arrangements are community-based, semi-formal, or dependent on external transport.
This divergence between administrative mandate and operational reality reflects scale misalignment rather than institutional absence. Expanding centralized mandate without adjusting system architecture would not alter the underlying scale constraint.
3.2.5 Technology emergence as structural response
Across cases, three system types appear consistently: (1) Micro reverse osmosis units in islands without viable freshwater; (2) Rainwater harvesting with basic treatment where rainfall reliability permits; (3) Hybrid configurations in upper-bound population cases.
Technology selection aligns directly with demand magnitude and raw water conditions. No case demonstrates sustainable centralized on-grid operation under micro-island conditions.
The empirical pattern therefore indicates that system choice is constrained by physical and demographic parameters rather than by policy preference.
Recent advances in modular desalination technologies, renewable-energy integration, and distributed rainwater harvesting systems further strengthen the feasibility of decentralized water supply approaches for remote island settlements. These technologies are increasingly recognized as practical alternatives for achieving water security under severe physical and demographic constraints.
Across all surveyed micro islands (≤ 200 inhabitants), three structural characteristics consistently recur: limited land area with narrow hydrological catchment, high vulnerability to seawater intrusion, and extremely small aggregated domestic water demand (< 15 m³/day). Even where rainfall is adequate, storage limitations and shallow or saline aquifers prevent the establishment of permanent freshwater sources. Water scarcity under these conditions is therefore structural and spatial in nature rather than a temporary seasonal imbalance.
From a demand perspective, populations at or below 200 inhabitants generate volumes that are inherently small in absolute terms. This minimal aggregation fundamentally shapes system feasibility. The operational environment of micro islands is defined by the simultaneous presence of hydrogeological fragility and permanently low demand magnitude.
Centralized utility systems, by contrast, depend on continuous raw water abstraction, demand aggregation, network economies, and capital amortization across large user bases. Empirical evidence shows that none of the surveyed micro islands operates under such conditions. Production volumes remain permanently low, and raw water sources are unreliable or saline. Under these circumstances, fixed infrastructure costs cannot be efficiently distributed, and unit production costs increase as output declines. Expanding centralized networks into micro-island environments does not alter these scale fundamentals.
Observed system patterns indicate that only three configurations consistently align with micro- and small-island conditions: micro reverse osmosis units where seawater or brackish sources dominate, rainwater harvesting with basic disinfection where rainfall reliability permits, and hybrid RWH–RO systems for upper-bound micro populations. These systems match low production volumes, avoid extensive distribution networks, and allow modular capital deployment with localized operation. No empirical case supports sustainable centralized on-grid provision under micro-island conditions.
Taken together, the findings demonstrate that drinking water insecurity on very small inhabited islands is the product of interacting structural variables: hydrogeological limitation, extremely small demand volume, and absence of scalable aggregation. Micro islands operate below the effective scale required for centralized utility systems. Under these conditions, modular off-grid systems constitute the only configurations consistently compatible with observed demand levels and raw water realities. Water provision in archipelagic settings must therefore be structured around scale-differentiated system design rather than uniform infrastructure expansion.
Although not explicitly framed as interrogative statements in the introduction, this study is guided by three underlying research questions derived from the structural, scale, and policy gaps identified in earlier sections. The empirical synthesis directly addresses each of these questions, as summarized in Table 15.
Table 15. Research questions, key findings, and structural implications
|
Research Question |
Analytical Focus |
Key Empirical Findings |
Structural Implication |
|
RQ-1: What structural characteristics constrain freshwater availability on very small islands? |
Hydrogeological–demographic |
Micro islands (≤ 200 inhabitants) exhibit limited catchment areas, saline or brackish groundwater, absence of permanent freshwater, and absolute demand < 15 m³/day. |
Water scarcity is inherent to physical and demographic scale, not a temporary supply failure. |
|
RQ-2: Why are centralized water utilities–based systems unviable? |
Institutional–scale |
Observed demand volumes remain far below typical centralized production thresholds; no island in the sample operates an on-grid system; high unit costs persist under low output. |
Service failure reflects structural scale incompatibility rather than institutional weakness. |
|
RQ-3: What scale-appropriate solutions ensure sustainable access? |
System design–policy |
Micro RO, rainwater harvesting with basic treatment, and hybrid systems consistently align with demand magnitude and raw water constraints. |
Island-category-based system differentiation is required for sustainable provision. |
The answers to these research questions converge on a single structural insight: scale is the decisive variable linking physical constraint, economic feasibility, institutional performance, and technology selection. In very small island contexts, infrastructure design cannot be detached from demographic magnitude and hydrogeological limitations. Accordingly, drinking water insecurity in archipelagic states should not be interpreted as a failure of utility expansion, but as a consequence of applying uniform system logic to fundamentally heterogeneous spatial scales. Sustainable provision requires system architectures that are explicitly aligned with island-scale thresholds.
The implications presented below are derived from RQ-1 (structural constraints), RQ-2 (scale incompatibility), and RQ-3 (scale-appropriate configurations) and directly reflect the structural relationships identified in Section 4.
Empirical findings indicate that drinking water insecurity on micro islands (≤ 200 inhabitants) is structurally determined by the interaction of three key variables: limited hydrogeological capacity, extremely low absolute demand (< 15–20 m³/day), and the absence of scalable aggregation. Under these conditions, centralized utility-based systems consistently operate below their effective production threshold.
Accordingly, infrastructure planning must shift from uniform expansion models toward scale-differentiated system design. First, island categories should be formally defined based on population size and demand magnitude, as micro islands require system typologies fundamentally distinct from mainland utility frameworks. Second, governance structures must be aligned with operational scale. While utilities may retain regulatory and technical oversight roles, operational management in micro-island contexts should be adapted to localized, low-volume configurations. Third, capital allocation should prioritize modular off-grid systems. Although per-capita capital costs tend to be higher at smaller scales, total absolute investment remains limited, and operational costs are significantly lower than prevailing informal water prices. Finally, technology deployment should be guided by demand thresholds rather than uniform policy templates: (1) ≤ 200 inhabitants: modular micro-systems (e.g., RO or RWH-based); (2) 200–1,000 inhabitants: hybrid localized systems; (3) ≥ 10,000 inhabitants: centralized on-grid systems.
In archipelagic contexts, drinking water provision must therefore be structured around demographic and geographic scale as primary design variables. Scale-differentiated planning is not merely a policy preference, but a structural necessity for achieving sustainable water access [28-31].
The authors gratefully acknowledge the support of the National Research and Innovation Agency (BRIN), Indonesia, for facilitating this research through institutional research support and technical collaboration. The authors also thank regional stakeholders, local communities, and field respondents across the surveyed island locations for their assistance during field verification activities conducted between 2019 and 2023. Appreciation is further extended to reviewers and editors for their constructive comments, which substantially improved the quality and clarity of this manuscript.
The study was primarily conceptualized, designed, and led by Nicco Plamonia, who was responsible for developing the research framework, comparative analytical approach, interpretation of findings, and preparation of the original manuscript draft. Nicco Plamonia also coordinated the overall research process, integrated the analytical sections, and served as the corresponding author. Rizki Arizal Purnama and Wahyu Purwanta contributed to methodological development, analytical refinement, and manuscript review. Yeni Novitasari, Elshedevika Rosya, Heru Mulyono, and Rizki Firmansyah contributed to data interpretation, policy analysis, and technical discussion. Dwi Budiyanto Trisnoharjono, Ahmad Nuridha, Indri Mardiyana, Suwarti Wati, Agus Setiawan, and Susilo Raharjo supported technical validation, engineering interpretation, and review of the manuscript. Revina Devitani Putri assisted with literature compilation, data organization, and manuscript preparation. All authors reviewed and approved the final version of the manuscript.
The data supporting the findings of this study were derived primarily from publicly available secondary sources, including datasets and reports from Badan Informasi Geospasial (BIG), Badan Pusat Statistik (BPS), BRIN technical reports, regional government documents, and previously published literature. Limited field verification data collected between 2019 and 2023 were used to corroborate secondary findings. Due to the combination of institutional reports, observational field notes, and partially unpublished verification materials, some supporting datasets are available from the corresponding author upon reasonable request.
[1] Badan Informasi Geospasial. (2024). Pulau Indonesia bertambah jadi 17.380, mengapa angkanya berubah setiap tahun? Badan Informasi Geospasial. https://www.big.go.id/uploads/downloads/Dokumen/Rilis_Jumlah_Pulau.pdf.
[2] Badan Informasi Geospasial. (2021). Gazeter Republik Indonesia. Jakarta, Indonesia, Badan Informasi Geospasial. https://sinar.big.go.id/assets/document/gazeter/gazeter_1643095453.pdf.
[3] Connell, J., Lowitt, K., Ville, A.S., Hickey, G.M. (2020). Food Security in Small Island States. Springer. https://doi.org/10.1007/978-981-13-8256-7
[4] Coulon, C., White, J.T., Pryet, A., Gatel, L., Lemieux, J.-M. (2024). An ensemble-based approach for pumping optimization in an island aquifer considering parameter, observation and climate uncertainty. Hydrology and Earth System Sciences, 28(1): 303-319. https://doi.org/10.5194/hess-28-303-2024
[5] Mohsen, M.S. (2007). Water strategies and potential of desalination in Jordan. Desalination, 203(1-3): 27-46. https://doi.org/10.1016/j.desal.2006.03.524
[6] Still, D.A. (1991). Small scale desalination in South Africa with particular reference to solar distillation. OpenUCY, 1.1-9.1. http://hdl.handle.net/11427/12585.
[7] Patton, C., Sawicki, D., Clark, J. (2015). Basic Methods of Policy Analysis and Planning. Routledge. https://doi.org/10.4324/9781315664736
[8] Dale, A., Arber, S., Procter, M. (2025). Doing Secondary Analysis. London: Routledge.
[9] Levy, J.S. (2008). Case studies: Types, designs, and logics of inference. Conflict Management and Peace Science, 25(1): 1-18. https://doi.org/10.1080/07388940701860318
[10] Beach, D., Pedersen, R.B. (2016). Causal Case Study Methods: Foundations and Guidelines for Comparing, Matching, and Tracing. Ann Arbor, University of Michigan Press. https://doi.org/10.3998/mpub.6576809
[11] Borman, K.M., Clarke, C., Cotner, B., Lee, R. (2012). Cross-case analysis. In Handbook of Complementary Methods in Education Research, pp. 123-139. https://www.routledge.com/9780203874769.
[12] Hehanussa, P.E. (2002). Ecohydrology and Tecto-Genesis of small islands in Indonesia. In Hydrology and Water Management in the Humid Tropics.
[13] Holding, S., Allen, D.M., Foster, S., Hsieh, A., Larocque, I., Klassen, J., Van Pelt, S.C. (2016). Groundwater vulnerability on small islands. Nature Climate Change, 6(12): 1100-1103. https://doi.org/10.1038/nclimate3128
[14] Kumar, P., Avtar, R., Dasgupta, R., Johnson, B.A. (2020). Socio-hydrology: A key approach for adaptation to water scarcity and achieving human well-being in large riverine islands. Progress in Disaster Science, 8: 100134. https://doi.org/10.1016/j.pdisas.2020.100134
[15] Terry, J.P., Falkland, A.C. (2009). Responses of atoll freshwater lenses to storm-surge overwash in the Northern Cook Islands. Hydrogeology Journal, 18(3): 749-759. https://doi.org/10.1007/s10040-009-0544-x
[16] Bailey, R.T., Jenson, J.W., Olsen, A.E. (2010). Estimating the ground water resources of atoll islands. Water, 2(1): 1-27. https://doi.org/10.3390/w2010001
[17] White, I., Falkland, T. (2009). Management of freshwater lenses on small Pacific islands. Hydrogeology Journal, 18(1): 227-246. https://doi.org/10.1007/s10040-009-0525-0
[18] Griffin, R.C. (2016). Water Resource Economics: The Analysis of Scarcity, Policies, and Projects. MIT Press. https://mitpress.mit.edu/9780262034043/water-resource-economics/.
[19] O’Connell, E. (2017). Towards adaptation of water resource systems to climatic and socio-economic change. Water Resources Management, 31(10): 2965-2984. https://doi.org/10.1007/s11269-017-1734-2
[20] Birawida, A.B., Ibrahim, E., Mallongi, A., Rasyidi, A.A.A., Thamrin, Y., Gunawan, N.A. (2021). Clean water supply vulnerability model for improving the quality of public health (environmental health perspective): A case in Spermonde islands, Makassar Indonesia. Gaceta Sanitaria, 35(S2): S601-S603. https://doi.org/10.1016/j.gaceta.2021.10.095
[21] Ghaffour, N., Missimer, T.M., Amy, G.L. (2013). Technical review and evaluation of the economics of water desalination: Current and future challenges for better water supply sustainability. Desalination, 309: 197-207. https://doi.org/10.1016/j.desal.2012.10.015
[22] Elliott, M., MacDonald, M.C., Chan, T., Kearton, A., Shields, K.F., Bartram, J.K., Hadwen, W.L. (2017). Multiple household water sources and their use in remote communities with evidence from Pacific Island countries. Water Resources Research, 53(11): 9106-9117. https://doi.org/10.1002/2017WR021047
[23] White, I., Falkland, T. (2015). Integrated management of urban water supply and water quality in developing Pacific Island countries. In Global Issues in Water Policy, Springer, pp. 489-526. https://researchportalplus.anu.edu.au/en/publications/integrated-management-of-urban-water-supply-and-water-quality-in-.
[24] IPCC. (2014). Small islands. in Climate Change 2014: Impacts, Adaptation, and Vulnerability, pp. 1613-1654. https://doi.org/10.1017/CBO9781107415386.009
[25] Olufisayo, O.E., Olanrewaju, O. (2024). A review of renewable energy powered seawater desalination treatment process for zero waste. Water, 16(19): 2804. https://doi.org/10.3390/w16192804
[26] Mycoo, M.A. (2017). Beyond 1.5 ℃: Vulnerabilities and adaptation strategies for Caribbean Small Island Developing States. Regional Environmental Change, 18(8): 2341-2353. https://doi.org/10.1007/s10113-017-1248-8
[27] Campisano, A., Butler, D., Ward, S., Burns, M.J., Friedler, E., DeBusk, K., Fisher-Jeffes, L.N., Ghisi, E., Rahman, A., Furumai, H., Han, M. (2017). Urban rainwater harvesting systems: Research, implementation and future perspectives. Water Research, 115: 195-209. https://doi.org/10.1016/j.watres.2017.02.056
[28] Fritzmann, C., Löwenberg, J., Wintgens, T., Melin, T. (2007). State-of-the-art of reverse osmosis desalination. Desalination, 216(1-3): 1-76. https://doi.org/10.1016/j.desal.2006.12.009
[29] Elimelech, M., Phillip, W.A. (2011). The future of seawater desalination: Energy, technology, and the environment. Science, 333(6043): 712-717. https://doi.org/10.1126/science.1200488
[30] Karagiannis, I.C., Soldatos, P.G. (2008). Water desalination cost literature: Review and assessment. Desalination, 223(1-3): 448-456. https://doi.org/10.1016/j.desal.2007.02.071
[31] Shannon, M.A., Bohn, P.W., Elimelech, M., Georgiadis, J.G., Mariñas, B.J., Mayes, A.M. (2008). Science and technology for water purification in the coming decades. Nature, 452(7185): 301-310. https://doi.org/10.1038/nature06599