Sacred Water Architecture and Blue-Green Infrastructure for Sustainable Urban Planning in Bali

The intersection of cultural heritage and sustainable urban development is one of the most pressing challenges facing rapidly urbanizing regions worldwide [1]. In Bali, Indonesia, this challenge is visible in the condition of sacred water architecture, a system of water temples, springs, canals, and ceremonial bathing places that has sustained both spiritual practice and ecological balance for over a millennium [2]. The Historic Urban Landscape (HUL) approach, developed by UNESCO and subsequent scholarship, calls for integrating cultural values, traditional knowledge systems, and natural heritage into urban planning frameworks [1].

Balinese water temples exemplify what Lansing and Fox [2] termed niche construction, a process in which cultural practices and ecological systems co-evolve to produce sustainable human-environment relationships. The subak irrigation system, inscribed on the UNESCO World Heritage List in 2012, demonstrates how the Tri Hita Karana philosophy, which promotes harmony among the spiritual realm, human community, and natural environment, has produced resilient socio-ecological systems [3]. Contemporary urbanization in Denpasar and surrounding areas, however, is fragmenting these sacred landscapes and disrupting water management practices that have proven sustainable over centuries.

Recent scholarship increasingly treats sacred water bodies as multifunctional blue-green infrastructure (BGI), capable of providing ecosystem services while retaining cultural and spiritual significance [4, 5]. These spaces function simultaneously as centers of community life, ecological assets, and providers of regulating, provisioning, and cultural services [6, 7]. BGI refers to networks of natural and semi-natural features that provide water-related ecosystem services; this concept offers a framework for understanding how traditional sacred water architecture can inform contemporary sustainable urban planning [8, 9].

Despite growing recognition of cultural heritage's value in sustainable development, as reflected in Sustainable Development Goal Target 11.4 on protecting cultural and natural heritage [8, 10], practical frameworks for integrating sacred water architecture into urban planning remain underdeveloped. The gap between policy aspiration and implementation is especially acute in Global South contexts, where rapid urbanization intersects with dense cultural heritage [11]. Indigenous knowledge systems, including traditional ecological knowledge (TEK) embedded in sacred site management, offer insights for climate adaptation and sustainable development but are frequently marginalized in technocratic planning processes [10, 12].

This gap is not unique to Bali. Cities across South and Southeast Asia, as well as parts of Latin America and sub-Saharan Africa, contend with comparable tensions between historically-evolved sacred water systems and contemporary infrastructure planning. A transferable model for reading, valuing, and integrating these systems into formal planning would therefore have relevance well beyond the Indonesian context. This research addresses that gap by developing the Sacred Water Urban Integration Model (SWUIM), a preservation framework that synthesizes Balinese water temple typology with contemporary BGI planning. The work contributes in three ways: (1) systematic documentation and analysis of sacred water architecture typologies and their fragmentation patterns in an urban context; (2) empirical assessment of ecosystem services provided by sacred water sites functioning as urban BGI; and (3) an actionable, transferable framework for heritage-led urban regeneration that other rapidly urbanizing regions with comparable sacred-water traditions can adapt and test.

3.1 Research design, study area, and case selection

This study used a sequential mixed-methods design combining spatial, qualitative, ecological, and survey-based evidence. The study area was the Greater Denpasar metropolitan region in Bali, Indonesia, selected because it contains a concentration of sacred water sites exposed to rapid urbanization pressure. As Bali’s capital and largest urban region, Greater Denpasar provides an appropriate setting for examining the interaction between traditional sacred-water practices and contemporary urban development.

The research proceeded in four phases: (1) identification and documentation of 27 sacred water sites; (2) intensive comparative analysis of nine case-study sites representing varied typologies and fragmentation conditions; (3) socio-ecological assessment of the Tukad Badung river corridor; and (4) participatory appraisal and refinement of the literature-derived SWUIM with traditional leaders, planners, civil-society representatives, and community members.

The 27-site inventory was assembled through purposive, criterion-based selection rather than probability sampling. Sites were included when they were located within Greater Denpasar, contained a physically identifiable water feature associated with a temple, ritual, or customary practice, retained continuing or historically documented sacred-water significance, and could be accessed with the permission of the relevant custodians. The nine intensive case studies were selected using maximum-variation logic to represent different sacred-water typologies, contrasting fragmentation patterns, and varied positions within the Tukad Badung corridor and surrounding urban context. Consequently, the reported prevalence of fragmentation describes the analyzed inventory and should not be interpreted as a statistically representative estimate of all sacred water sites in Greater Denpasar. Figure 2 shows the spatial distribution of the 27 inventoried sites and the nine intensive case-study locations.

tu_pian_2.png

3.2 Data collection methods

Spatial data on all 27 sites were collected using GPS coordinates, satellite imagery analysis, and GIS mapping. Historical maps and colonial-era documentation were digitized to assess landscape transformation over time. Spatial analysis included buffer zone delineation, land use change assessment, and connectivity analysis using ArcGIS 10.8.

Ethnographic data were gathered through semi-structured interviews with 45 key informants, including temple priests (pemangku), subak leaders (pekaseh), community elders, and regular worshippers. Interview protocols addressed historical site significance, ritual practice, perceived threats, and community stewardship mechanisms. Participant observation was conducted during major water ceremonies, including Melasti and Tirtayatra.

Ecological data were collected at 15 sacred water sites during the wet and dry seasons. The monitored water-quality parameters were pH, dissolved oxygen, turbidity, nitrate, and phosphate. Hydrological observations addressed flow variation and short-term stormwater retention, while microclimate observations recorded temperature and humidity at sacred sites and nearby built-up comparison locations. Biodiversity observations documented the presence of aquatic and riparian flora and fauna. The ecological component was designed as an exploratory site assessment rather than as a controlled experiment; consequently, the data were summarized descriptively by site and season, and no inferential significance tests or causal estimates are reported.

Stakeholder surveys used structured questionnaires (n = 312) to assess community perceptions of ecosystem services, cultural attachment, and management preferences. Sampling employed stratified random selection across urban, peri-urban, and rural zones, with survey instruments adapted from validated scales in the ecosystem-service perception literature [5]. Of 360 eligible individuals invited to participate, 312 complete questionnaires were retained for analysis, corresponding to a response rate of 86.7%.

The demographic profile of the respondents is presented in Table 1. The sample included 148 male respondents (47.4%) and 164 female respondents (52.6%). Respondents were distributed across four age groups: 18–30 years (72; 23.1%), 31–45 years (96; 30.8%), 46–60 years (88; 28.2%), and over 60 years (56; 17.9%). In terms of residential location, 156 respondents (50.0%) lived in urban areas, 96 (30.8%) in peri-urban areas, and 60 (19.2%) in rural areas. Educational attainment was distributed as secondary education (74; 23.7%), diploma or bachelor's degree (178; 57.1%), and postgraduate degree (60; 19.2%). Regarding their relationship with sacred water sites, 112 respondents (35.9%) were local residents or community members, 88 (28.2%) were regular worshippers or site users, 56 (17.9%) were temple custodians or traditional leaders, and 56 (17.9%) were other community stakeholders.

The questionnaire contained ten ecosystem-service recognition items: spiritual practice, cultural identity, aesthetic appreciation, education and awareness, water-quality filtration, urban heat mitigation, ceremonial water supply, urban agriculture, subsistence fishing, and traditional medicine. Responses were recorded using a five-point Likert scale: 1 = strongly disagree to 5 = strongly agree. An item was classified as endorsed when a respondent selected either “agree” or “strongly agree” (scores 4 or 5). Percentages therefore represent item-level endorsement among the 312 valid respondents and are not composite scores for the cultural, regulating, or provisioning domains. Because the items represent distinct ecosystem-service functions rather than a single reflective latent construct, an overall Cronbach's alpha was not interpreted as evidence of scale reliability. Survey data were analyzed descriptively using valid frequencies and percentages. The ten survey items and their corresponding endorsement results were therefore used to describe perceptions of ecosystem-service functions provided by sacred water sites.

Table 1. Profile of survey respondents (n = 312)

3.3 Participatory appraisal of Sacred Water Urban Integration Model

Participatory appraisal of SWUIM was conducted with 32 stakeholders purposively recruited through customary institutions, municipal planning offices, civil-society networks, and communities associated with the selected sites. The participants comprised 8 traditional leaders and temple custodians, 8 government planners, 6 NGO or civil-society representatives, and 10 local community members, for a total of 32 participants. Eligibility required direct knowledge of, responsibility for, or regular engagement with at least one case-study site. Workshop sessions applied the four literature-derived dimensions to three sites representing different typologies and fragmentation conditions. Participants worked with site maps, fragmentation summaries, and a four-dimension strategy matrix.

Workshop notes, annotated maps, and strategy matrices were reviewed using four explicit appraisal criteria: empirical fit with site conditions, cultural legitimacy, ecological relevance, and planning feasibility. A dimension was treated as retained when it remained applicable across the three case applications; it was treated as refined when participants proposed additional safeguards, actors, or implementation mechanisms. The exercise was designed as participatory appraisal and operational refinement, not as statistical, predictive, or psychometric validation.

3.4 Analytical framework

Sacred water sites were classified according to physical characteristics, ritual function, and governance structure. Fragmentation was analyzed through three dimensions: spatial discontinuity, functional degradation, and cultural disconnection. Ecosystem services were organized using the Common International Classification of Ecosystem Services (CICES), adapted to the sacred-water context, and grouped into regulating, provisioning, and cultural services. Evidence was synthesized from ecological observations, structured survey responses, interviews, participatory mapping, and photo-elicitation [5]. No monetary valuation or benefit-transfer analysis was undertaken; accordingly, the study reports biophysical indicators and socio-cultural perceptions rather than economic values.

Survey data were analyzed using valid frequencies and item-level percentages. Interview, observation, mapping, and workshop materials were synthesized thematically, using the four literature-derived SWUIM dimensions as deductive categories while allowing site-specific themes to emerge inductively. SDG alignment was assessed against established heritage-SDG methodologies [8, 9], with primary attention to SDG 6.6, SDG 11.4, and SDG 15.1.

5.1 Integrating indigenous knowledge with contemporary planning

The SWUIM model demonstrates a practical application of two-eyed seeing principles [10], bridging Balinese TEK embedded in the Tri Hita Karana philosophy with contemporary BGI science. This integration goes beyond tokenistic consultation toward substantive co-design, in which indigenous cosmological concepts inform technical planning decisions. Traditional sacred buffer zone concepts, such as telajakan and lebuh, align with contemporary riparian corridor planning while carrying spiritual significance that strengthens community stewardship.

Power asymmetries nonetheless persist. Urban planning institutions continue to privilege technical-scientific knowledge over traditional wisdom, consistent with Baresi and Russo’s [23] observations of SEA processes. Workshop discussions indicated that traditional leaders perceived themselves as marginalized despite nominal inclusion, suggesting the need for a more fundamental restructuring of decision-making authority rather than merely expanding stakeholder consultation. Evidence from community-led Participatory Action Research in sacred-site management [25] further suggests that genuine power-sharing is more consistent with indigenous knowledge sovereignty than expert-driven consultation alone. Figure 5 illustrates the knowledge integration framework linking TEK and contemporary urban planning through participatory co-design.

5.2 Sacred water sites as blue-green infrastructure

These findings are consistent with international research showing that historically-evolved water management systems can inform contemporary BGI development [7, 19]. As in Singapore's ABC Waters Program in high-density contexts [19], Balinese sacred water architecture shows that even small water bodies can provide ecosystem service benefits disproportionate to their spatial footprint. Cultural ecosystem services, particularly spiritual significance and place attachment, generate community stewardship that supports long-term sustainability, a dimension that technocratic BGI planning tends to undervalue [16, 20].

The three-dimensional fragmentation framework used here, covering spatial, functional, and cultural dimensions, offers more analytical precision than typical habitat fragmentation models. Cultural disconnection is a critical but under-recognized threat: even physically intact sites lose sustainability when traditional governance erodes and intergenerational knowledge transmission fails, a finding consistent with research on indigenous knowledge systems showing that cultural continuity is essential for environmental stewardship [21, 22].

The Tukad Badung corridor analysis reveals tension between ecosystem service provision and sacred landscape integrity. Ecological restoration could enhance regulating services, but traditional communities prioritize ritual purity and spiritual access over ecological metrics. This value plurality calls for negotiated solutions rather than technical optimization, a principle central to the HUL approach [1] but difficult to implement within planning frameworks that require quantifiable outcomes.

tu_pian_5.png

Figure 5. Sacred Water Urban Integration Model (SWUIM) knowledge integration framework

tu_pian_6.png

Figure 6. Policy and practice implications of the Sacred Water Urban Integration Model (SWUIM) model

5.3 Policy and practice implications

The SWUIM model offers actionable pathways for heritage-led urban regeneration [1, 13, 27]. Key policy recommendations include: (1) formal recognition of sacred water sites as critical urban infrastructure within spatial plans; (2) hybrid governance mechanisms that integrate customary law (adat) with municipal authority; (3) financial instruments, including heritage-led blended finance [28] and payment-for-ecosystem-services schemes that value cultural as well as Ecological Functions; and (4) educational programs that address both traditional knowledge transmission and contemporary sustainability literacy.

Digital technologies present both opportunities and risks. Building Information Modeling (BIM) and digital heritage tools can enhance documentation and tourism management [29], but their application must avoid extractive uses of indigenous knowledge that occur without community consent or benefit [23]. Culturally-grounded digital heritage education, as demonstrated in Indonesia [26], offers a more equitable model that strengthens rather than supplants traditional knowledge systems.

These findings also bear on SDG implementation through culture-based pathways [8, 30]. Global frameworks, including the UN 2030 Agenda and the Paris Agreement, would benefit from fuller acknowledgment of indigenous knowledge's role in climate adaptation and sustainable development [10]. The SWUIM model shows that cultural heritage is not merely protected under SDG 11.4 but can actively enable multiple interconnected goals, a systems perspective that is relevant to integrated sustainable development well beyond Bali. Figure 6 extends these linkages by illustrating how sacred-water heritage can contribute to interconnected pathways across heritage protection, climate action, water management, terrestrial ecosystems, and cross-sectoral partnerships.

5.4 Transferability and global relevance

SWUIM was developed and participatorily appraised in a single metropolitan region; consequently, its transferability should not be assumed. That said, the underlying logic, treating sacred or historically significant water systems as functioning BGI rather than as static heritage objects, is not specific to Bali. Comparable configurations exist in India's temple tanks and sacred groves, in the qanat and sabil water systems of parts of the Middle East and North Africa, in cenotes and sacred springs across Mesoamerica, and in holy wells and river shrines throughout South and East Asia. The transferable contribution of the study lies primarily in the three-dimensional fragmentation framework and the four-dimension SWUIM structure, both of which require adaptation to local typologies, governance systems, and belief structures. Accordingly, transferability is framed as a testable proposition for future comparative research rather than as a settled claim.

5.5 Limitations and future research directions

First, the 27-site inventory was purposive and criterion-based rather than probabilistic. This design enabled comparative analysis across accessible sites with documented sacred-water significance, but it may underrepresent abandoned, inaccessible, or fully absorbed sites. The reported prevalence of spatial discontinuity, functional degradation, and cultural disconnection is therefore specific to the analyzed inventory and should not be generalized as a population estimate for all sacred water sites in Greater Denpasar.

Second, three of the four authors are based in Balinese architecture and engineering programs, and the research team's positioning as cultural insiders shaped both access to sacred sites and the interpretation of ritual significance. This proximity was valuable for building trust with pemangku and community elders, but it also raises the possibility that findings emphasizing the resilience and adaptability of Tri Hita Karana practice reflect, in part, the research team's own investment in that framework. Independent replication by researchers without this positioning would strengthen confidence in the results.

Third, ecosystem service quantification relied primarily on qualitative and perception-based methods rather than continuous ecological monitoring, and the study lacks a genuine pre-urbanization baseline; the 1960s riparian width figures for the Tukad Badung corridor come from historical maps rather than instrumented data, which limits the precision of any claim about the rate of ecological change. Future research should integrate longer-term ecological monitoring, economic valuation using both revealed- and stated-preference methods, and social network analysis of governance dynamics. Climate change impacts on sacred water site hydrology, and their implications for ritual practice, remain a significant knowledge gap.

Finally, this research focused on Greater Denpasar, and transferability to other contexts, in Bali and elsewhere, requires careful attention to local cultural specificities, governance structures, and urbanization patterns; the SWUIM framework's general principles may apply broadly, but implementation details must emerge from place-based co-design rather than direct transplantation. Comparative research across Southeast Asian and other sacred-water contexts could clarify which elements of the framework generalize and which are specific to the Balinese case, and should give particular attention to gendered dimensions of sacred site stewardship, to ensure that integration efforts do not reinforce existing power inequalities within communities.

The participatory appraisal involved 32 stakeholders and three case applications. It provides evidence of contextual relevance, cultural legitimacy, and implementation feasibility, but it does not establish psychometric, predictive, or cross-cultural validity. Further application across additional sites and independent stakeholder groups is therefore required before broader validation can be claimed.

The authors are grateful to the temple priests (pemangku), subak leaders (pekaseh), and community members who shared their knowledge and facilitated access to sacred spaces. Institutional support from the Denpasar City Planning Agency and the Bali Provincial Heritage Office is gratefully acknowledged, as is the assistance of the research assistants and students who contributed to data collection. The research was conducted with attention to Balinese cultural protocols and in the spirit of collaborative knowledge creation.

The first author gratefully acknowledges the Doctoral Program in Engineering Science, Faculty of Engineering, Udayana University, and the supervisory team for their academic guidance, critical feedback, and support throughout the research.