Environmental Impacts of Rainwater Harvesting Systems in Urban Areas Applying Life Cycle Assessment—LCA

Abstract

Climate change poses a series of challenges to water management to satisfy society’s current and future needs. Considering water an essential resource for life, this research is dedicated to comparing the environmental impacts caused by the conventional water supply system and Conventional Water Supply system complemented with rainwater harvesting, considering in the first one not only the components of the municipal public supply, but also the hydraulic components of the residences until the point of use. A life cycle assessment—LCA—was conducted for the two systems, considering them from the catchment to the point of use. This methodology quantifies environmental impacts throughout the product or process life cycle to obtain sustainable options, from raw material extraction to ultimate disposal. The results expose that the hybrid system represents an increase in environmental impacts. However, at a building scale, this increase was very small, and this can be seen as favorable to the hybrid system due to the benefit it brings. The seven percent reduction in water demand over the conventional system can represent significant relief in regions that already have water stress as a reality. This study has the potential to guide managers and designers of public water policies, providing data for a better decision-making process.

1. Introduction

Water is an essential resource for life. Factors such as population growth, urban concentration, illegal connections, pollution, climate change, and the poor preservation of springs and ecosystems threatens its existence and supply [1,2,3]; therefore, there is no uniformity in its spatial distribution. An increase of 55% in global water demand is expected by 2050 [4], mainly due to growing demand from the industrial sector, thermoelectric power generation systems, and domestic users. For this reason, several countries have devised some strategies to ensure the sufficiency of this resource, especially in urban areas [5,6,7].

Currently, the processes of planning, executing, and operating water systems in densely populated areas require more complex actions and greater investments to deal with the necessary supply [8]. SANEPAR, a sanitation company in Brazil, claims that this situation occurs because a hydrographic basin with urban and industrial use generates pollutants that accumulate along the downstream watercourse, compromising the water quality [5]. The big challenge for cities is finding viable alternative water sources to avoid economic, environmental, political, and societal stress. For example, in the Brazilian city of São Paulo, the high demand for water for urban-industrial activities led to an infrastructure project. It permitted the transference of water from the basins of the Juquiá, Itapanhaú, São Lourenço, and Itatinga rivers, promoting a contribution of around 6.2 m³/s to the main system that supplies the metropolitan region of São Paulo [9]. Another example is in Asia, where China’s government has been building many dams to improve the availability of water resources [10].

However, the concern with water supply is not just local but global. Mhlongo et al. [11] highlighted that in South Africa, fresh water is limited and vulnerable in terms of quality and quantity, requiring improvements in social and environmental conditions using available technology to boost the quality of the water. Indeed, Hanumante et al. [12] concluded that the overall global water availability is adequate to support the increasing demand for this resource in the next century. Still, due to regional characteristics, Africa suffers from greater water stress. In America, Garcia et al. [13] exposed that Los Angeles, Las Vegas, and Miami have experienced several situations of water stress, shown as droughts, reservoirs at the lowest levels, and threats from saltwater intrusion. These situations have motivated the development of environmental policies based on sustainable strategies to deal with the accelerated changes in water management.

Bashar et al. [14] worked on implementing rainwater harvesting systems in Bangladesh to help the supply in rural areas. They highlighted that reliabilities between 30% and 40% could be found in some cities, meaning savings of between 500 and 800 m³ of water per year that could complement the cities’ conventional water supply. In Poland, Krauze and Wagner [15] studied diverse nature-based solutions to manage water in Europe, following the United Nations Sustainability Goals. They presented a series of best practices for urban water management to comply with the water demands, applied in the early stages of a water supply plant’s planning and design process.

Some countries support the implementation of rainwater harvesting systems, such as Brazil, Australia, Germany, Italy, the United Kingdom, and the United States of America [16,17]. According to Espíndola et al. [18], the WHO/UNICEF Water Supply and Sanitation Monitoring Program report pointed out that 86 million people (around 17 million households) were using rainwater for human consumption, or 1.6% of the 5.3 billion people included in the sample. Considering the rural population also, the usage goes up to 2.4%.

A solution to the water stress situation is using rainwater harvesting systems. These systems collect and store water from rain in areas such as terraces, rooftops, and courtyards, to storage devices such as cisterns or tanks, for further use in commercial, domestic, and industrial activities [19,20,21]. The process presents economic benefits, especially for rural areas, and has presented some advantages in urban areas to supply the demand for nonpotable purposes [22].

Water scarcity is a situation in which consumption approaches or exceeds the natural regeneration of the water body [23]. In Brazil, water imports have been a growing inconvenience in semiarid regions and large urban centers that often have abundant water resources. They are insufficient to satisfy the demands or are contaminated, exposing a supply–demand imbalance and presenting legal and political problems [24,25]. Several researchers have utilized the LCA methodology to assess water supply systems worldwide [26,27,28,29,30,31,32]. Some of the rainwater harvesting system studies are summarized next.

Rodriguez et al. [33] evaluated four treatment plants. In two of them, flocculation was the process with the highest loading, while in the other two, energy consumption was the most significant impact source. Stokes and Horvath [34] analyzed a hybrid system (of conventional and rainwater harvesting) in California (USA), formed by desalination, reuse, and imports. The impact analysis considered atmospheric and energy emissions. The system showed favorable results for reuse and imports, with insignificant differences.

Godskesen et al. [35] applied the LCA methodology to compare four types of water supply. The types of supply evaluated were rainwater collection in several city blocks, groundwater collection, accompanied by compensatory measures, the establishment of well areas in the most remote regions of the city, and seawater desalination. Rainwater harvesting, with a system comprising 68.50 m² of a catchment area and including both roof and road areas, was the system that presented the most favorable results for the environment.

Vialle et al. [36] compared the production of the whole system of municipal drinking water with the nonpotable rainwater harvesting system. The results showed that, for all impact categories, rainwater harvesting was slightly worse for the environment. Concerning the evaluated subsystems referring to rainwater collection, more than 60% of the impacts caused were due to the structure’s construction. Polyethylene tanks were identified as the biggest pollutant.

Ghimire et al. [37] also developed a work comparing rainwater harvesting with the municipal water supply service in Washington DC (USA). The study found 0.40 kg CO₂—eq./m³ for the municipal water supply and 0.33 kg CO₂—eq./m³ for rainwater harvesting. It concluded that the rainwater harvesting system performed better than the municipal supply system in all categories except ozone depletion. The difference between the two systems exceeded 40%.

In Brazil, Tarpani et al. [38] carried out a study in Florianópolis that compared the impacts caused by three types of supply: desalination by reverse osmosis, reuse of wastewater, and rainwater collection to replace the potable supply of the local distribution network. As a result, for the three of them, rainwater harvesting appeared as the preferable option, with emphasis on the categories climate change (591 kg CO₂—eq./1000 m³) and human toxicity. However, the study pointed out that rainwater harvesting presents the worst conditions regarding water depletion and marine toxicity.

There are contradictions in the environmental impacts between conventional supply and rainwater harvesting systems, demanding a re-definition of the variables and scopes of the study to validate the most feasible option. Additionally, some research related to hypothetical rainwater harvesting projects guided the development of this study’s methodology [13,14,15,22,26].

Mo et al. [39] conducted a study evaluating the energy used in the water supply system. The authors divided their analysis into two parts: the construction phase of the facilities and the operation and maintenance stage. In the construction phase, the diesel used in the concrete mixture was considered direct energy spent. In the operation and maintenance phase, the electricity connected to pumping water was considered direct energy. The energy linked to producing and delivering materials related to this stage, such as chemical products, was considered indirect energy. Finally, the diesel used to manufacture cement and its delivery to the site was considered indirect energy. The results showed that the energy consumed in the construction phase was insignificant compared to that in the operation and maintenance phase.

Angrill et al. [40], when trying to identify the optimal configuration for buildings with mixed supply, found values between 0.27 and 1.9 kg of CO₂—eq. For the most similar scenario to the building represented here (i.e., a building of six floors, with water distributed to the entire building—although it only met part of the demand—and an underground tank), the value was 0.79 kg CO₂—eq. The study concluded that the tank’s position was the most important factor. Additionally, placing it above the building represented an impact reduction of at least 25%.

Faragò et al. [41] compared four rainwater systems. Decentralized RWH required up to 14 times more plastic than any of the other alternatives due to PVC underground storage tanks. The use of electricity in the operational phase has contributed markedly to the impacts of climate change and the depletion of fossil resources in all alternative scenarios. It is worth remembering, however, that all the scenarios include some treatment process.

Reducing the dependence on watercourses helps to diminish the environmental impacts related to the water production. However, the need to evaluate the environmental impacts of rainwater harvesting systems emerges to verify how harmful they could be and further comparison with conventional supply systems is required to judge which one is more convenient [42]. To identify which systems or processes are more harmful in terms of environmental impacts, the most commonly used tool worldwide is life cycle assessment (LCA).

This methodology quantifies environmental impacts throughout the product or process life cycle to obtain sustainable options, from raw material extraction to ultimate disposal [43,44,45]. It consists of four stages: goal and scope definition; life cycle inventory (LCI); life cycle impact assessment (LCIA); and interpretation [46,47].

Since water is a resource that plays an important role in health, social and economic aspects [48], this study focuses on the application of an LCA analysis on the consumption of materials and energy associated with the conventional water supply (CWS) system and the supply system complemented with rainwater harvesting (RWH)—under typical conditions. The objective is to determine the environmental impacts of these two settings of water supply at the scale of a house and a building. Despite the significant amount of water in this city that exists, the location selected for the analysis is the city of Rio de Janeiro, Brazil, due to the deficiencies in the service of conventional supply systems in this city and the significant amount of people affected [49].

The novelty of this article remains in the application of the LCA methodology to compare the environmental impacts caused by the conventional supply system and conventional supply system complemented with the rainwater harvesting. Moreover, the first system considers the municipal public supply components and the residences’ hydraulic components until the point of use. In this way, the impacts produced by the conventional supply system is better observed. With this, a comparison between the conventional system and the system complemented by the capture of rainwater is offered more effectively. This comparison aims to guide managers and designers of public water policies, providing data for a better decision-making process.

2. Materials and Methods

A literature review was initially conducted through the main scientific literature databases, such as Science Direct, Scopus, Wiley, and Springer Link, to identify the benefits of rainwater harvesting systems compared to other available solutions. It also sought to identify the critical elements pointed out by similar studies in other locations to verify its application. Regarding the conventional supply system, an existing installation was considered, and access to data was given through a questionnaire sent to one of the engineers who worked in the water supply company. For the rainwater harvesting system, a hypothetical case study was prepared. Finally, it was possible to build the life cycle analysis per cubic meter of water produced by both proposals: the conventional water supply (CWS) system and conventional system complemented by rainwater harvesting (RWH). Figure 1 presents the different steps of the method mentioned above.

2.1. Hydraulic Project

First, conventional water supply (CWS), which is divided into Conventional Water Supply 1—CWS1 (considering elements commonly found in the literature); and Conventional Water Supply 2—CWS2 (considering, in addition to the elements normally found in the literature, the concrete used in the WTP and the hydraulic system of the residence to the point of use of drinking water). Then, the Rainwater Harvesting (RWH) as a complementary system is divided into: rainwater harvesting as a complementary system—RWH1 (considering elements normally found in the literature); and rainwater harvesting as a complementary system—RWH2 (considering, in addition to the elements normally found in the literature, the concrete used in the construction of the WTP and the hydraulic system of the residence to the point of use of both drinking and abstracted water).

Each of these assessments is made at the scale of (a) a single-story house where the projection area is equivalent to the catchment area of 150 m² and in which the potable water supplied by the network has enough pressure to reach the reservoir, and of (b) a building block with three floors and four apartments per floor, where the projection area is equivalent to the catchment area of 600 m².

The present research did not use a pre-existing commercial capture system. However, it proposed a simple model with usual materials in local constructions, respecting the legislation. For the RWH systems, the simplest possible treatment was adopted among the ones found on the market (with coarse solid filtration). Once the results presented by [39] pointed out that the energy consumed in the construction phase was insignificant compared to the operation and maintenance phase, only the energy used in operation and maintenance was considered. Although the authors of [40] have pointed out that upper reservoirs present much more efficient results, the greater volumes made this load increase unfeasible in a pre-existing structure that was not prepared for it. Then, lower rainwater catchment reservoirs were considered.

2.2. Conventional Supply System

The conventional supply system station comprises a catchment dam, a pumping station, a raw water pipeline, a treatment station, a reservoir, a treated water pipeline, and a distribution network. The treatment performed is a complete treatment (grading, desanding, coagulation, flocculation, decantation, filtration, chlorination, and filtration).

2.3. Consumption per Residence

The demand calculation was based on Rio de Janeiro’s building code [50], which is 300 L/day per habitable room and 50 L/day per parking space day. Considering the design of the present work, each residential unit contained one living room, four bedrooms, and one laundry room; in other words, each unit contained six habitable rooms, in addition to a garage, so that each housing unit had a project consumption value equal to (6 × 300) + (50 × 1) = 1850 L/day.

Data from IBGE (Instituto Brasileiro de Geografia e Estatística) indicates that the average number of inhabitants per household in the city of Rio de Janeiro is 2.9. This number being associated with the consumption estimated in the previous paragraph means that the estimated per capita consumption is 1850/2.9 = 637.9 L/inhab per day. This is a very conservative calculation since the World Health Organization states that 100 L/inhabitant/day would meet all needs. Additionally, values between 110 and 140 L/inhab.day are commonly found in the literature [4].

2.4. Rainwater Harvesting Volume

The Brazilian standard [51] presents the following methods to determine reservoir volumes: the Rippl daily method, Rippl monthly method, simulation method, Azevedo Neto method, German practical method, English practical method, and Australian practical method. Brandão and Marcon [52] assessed the performance of these methods based on the levels of guarantee of supply achieved; they found out that for all the considered methods used by the Associação Brasileira de Normas Técnicas, the Rippl daily and Rippl monthly methods were distinguished for generating reservoirs associated with high supply guarantees (between 98 and 100%). The reservoirs present a large volume due to the high guarantee.

For the present study, the Rippl method pointed out that the average precipitation can supply 22% of the total demand for a house and 7% of the demand for a 3-story building. Sanitary basins represent a significant percentage of total consumption, so the rainwater catchment can supply three out of five toilets in a house and one out of five toilets in each apartment.

2.5. Life Cycle Analysis

The life cycle analysis aimed to assess the environmental impacts of drinking water produced using the conventional urban supply system and the decentralized rainwater harvesting system associated with the centralized supply. These systems were selected considering the variations in materials of the reservoirs and the source of energy established. Next, the steps were divided according to the application of the LCA methodology [53,54].

2.5.1. Goal and Scope of the LCA Study

In this step, the aim of the investigation, the interactions, and the depth of the study were established. The system boundaries, indicating which processes were considered, are represented in Figure 2. It followed the steps mentioned by [55], as expected in a conventional supply system. A functional unit is a quantified description of the performance of a product or system used as a reference unit. It referred to the subject in which the impacts are calculated [55]. In this case, the functional unit was 1 m³ of water delivered to the end-user. Since the systems have different functions, the comparison was made as a complementary system to the other.

Assessment 1 was the analysis that was carried out just as it appears in the bibliographic reviews; it was carried out on for the conventional water supply, considering collection to distribution. However, the concrete used in the construction of the treatment plant was not considered, nor was the internal hydraulic structure of the building (building branches, pumps, reservoirs, and energy).

Assessment 2 was the analysis that was carried out from water distribution point (public domain) to the point of use (private domain). This included the subsystems that are normally left out of analyses (building branches, pumps, reservoirs, and energy).

2.5.2. Life Cycle Inventory

Inventory data considered the two main components in the material volume of each subsystem. The inventory considered was based on the information specified in the data survey field associated with the Ecoinvent 3.8 database. The impacts were calculated with the help of an Excel spreadsheet. The elements accounted for corresponded to the two largest mass contributors of each subsystem. Since this hierarchy was not evident, more elements were considered. In the case of the hydraulic structure of the house, in addition to the tank material (e.g., concrete or polyethylene), the piping (PVC) and the booster pump (steel or cast iron) were considered. In the treatment station, each chemical linked to the treatment process was accounted for in addition to the concrete.

Table 1. Elements considered in each proposed assessment.

What Was Considered in Each Analysis?
	Subsystems	Element	Lifetime	CWS1	CWS2	RWH1	RWH2
Municipal Public Water Supply	Dam	Rock	50 years	Ok	Ok	Ok	Ok
		Sand	50 years	Ok	Ok	Ok	Ok
	R.W. Adduction	FoFo	50 years	Ok	Ok	Ok	Ok
	Treatment	Steel	50 years	Ok	Ok	Ok	Ok
		PAC	X	Ok	Ok	Ok	Ok
		Polymer	X	Ok	Ok	Ok	Ok
		Hypochlorite	X	Ok	Ok	Ok	Ok
		Ac. Fluosilicic	X	Ok	Ok	Ok	Ok
		Concrete	50 years	X	Ok	X	Ok
	Public Reservoir	Steel	50 years	Ok	Ok	Ok	Ok
	T.W. Adduction	Casting Iron	50 years	Ok	Ok	Ok	Ok
	Network	PVC	50 years	Ok	Ok	Ok	Ok
		Casting Iron	50 years	Ok	Ok	Ok	Ok
	Energy—electricity	Energy (KWh)	24 h	Ok	Ok	Ok	Ok
	Energy—diesel	Energy (Kg)	X	Ok	Ok	Ok	Ok
Private Hydraulic Structure	Building Hydraulic Structure for Drink Water (D.W.)	PVC	50 years	X	Ok	X	Ok
		Concrete (reservoir)	50 years	X	Ok	X	Ok
		Steel (reservoir)	50 years	X	Ok	X	Ok
		Steel pump	15 years	X	Ok	X	Ok
	Energy 2 (D.W.)	Energy	X	X	Ok *	X	Ok
	Building Hydraulic Structure for Rain Water Harvesting (R.W.H.)	PVC	50 years	X	X	Ok	Ok
		Concrete	50 years	X	X	Ok	Ok
		Steel	15 years	X	X	Ok	Ok
		Polyethylene	30 years	X	X	Ok	Ok
	Energy 2 (R.W.H)	Energy	X	X	X	Ok	Ok

* for elements of the private hydraulic structure, the energy to pump potable water was only counted for the building, not the house.

shows the elements considered in each analysis. Figure 3 summarizes how they were quantified for the calculation of impacts.

2.5.3. Life Cycle Impact Assessment

In this step, the type and number of environmental impacts that the systems generate were quantitatively assessed based on the data obtained from the inventories [42]. The analysis of water-related studies by [56] revealed current preferences for the impact assessment methodology. CML appeared as the most used method in evaluating integrated urban water (of two or more infrastructure types) [56]. As the present study intends to evaluate one supply as a complement to the other CML was decided to be used (CML v 4.8 2016). The results will be presented based on the climate change category. Furthermore, the allocation method adopted was the cut-off method, in which only loads directly linked to the product are associated with it.

The Paris Agreement aims to limit the change in global average temperature from preindustrial levels to 1.5 °C [57]. Achieving the ambition requires a global transformation to net zero greenhouse gases (GHG) by midcentury in all sectors of society. Due to this situation, the results of the present research will be measured in kg of CO₂-eq, as they serve to quantify the effects on climate change (greenhouse gases are also a measure, depending on the LCA method adopted). The formula to calculate the climate change impact category is shown below (Equation (1)).

$$Impact Category=\frac{Amount of Material}{m^3 of water produced}\times \left(\frac{{Impact}^{*}}{Amount of Material}\right)$$

(1)

*: Data taken from EcoInvent database.

2.5.4. Interpretation of Results

The presentation of results followed the contribution analysis model associated with dominance analysis, to identify the biggest contributors to environmental impacts and determine what improvements are needed. The sensitivity of a model describes the extent to which the variation in an input parameter or a choice leads to a variation in the model result. A model is sensitive to a parameter if a small change in this parameter results in a large change in the model result. This change is analyzed by varying one parameter at a time and called local sensitivity [42]. This document then performs the local sensitivity analysis, with variation in the parameters of energy source and type of residence (single-family residence with one floor or multifamily residence with three floors and a percentage of rainwater harvesting service).

3. Results and Discussion

The results are divided into the types of residential projects mentioned in Section 2.1.

3.1. Hydraulic Project

3.1.1. One-Story House (CWS1 × RWH1)

The results show that the house supplied by RWH1 produced more environmental impacts in all the evaluated categories (see

Table 2. CWS1 × RWH1 (house in a place with a hydroelectricity source).

Environmental Impacts	House CSW1	House RWH1
acidification (kg SO2-eq)	1.00 × 10−3	2.23 × 10−3
climate change (Kg CO2-eq)	1.95 × 10−1	5.50 × 10−1
freshwater ecotoxicity (Kg 1.4-DCB-eq)	1.50 × 10−1	4.13 × 10−1
marine ecotoxicity (Kg 1.4-DCB-eq)	4.59 × 102	1.01 × 103
terrestrial ecotoxicity (Kg 1.4-DCB-eq)	1.12 × 10−3	5.09 × 10−3
energy resources: nonrenewable (MJ)	2.31 × 100	1.09 × 101
eutrophication (Kg PO4-eq)	5.32 × 10−4	9.23 × 10−4
human toxicity (Kg 1.4-DCB-eq)	5.00 × 10−1	1.73 × 100
material resources: metals/minerals (Kg SB-eq)	7.66 × 10−6	8.91 × 10−6
ozone depletion (Kg CFC-11-eq)	1.77 × 10−8	2.77 × 10−8
photochemical oxidation formation (Kg ethylene-eq)	5.40 × 10−5	1.45 × 10−4

). The difference between the two systems ranged from 14% to 78% according to the observed category, and the greatest damage was in the ‘Marine Ecotoxicity’ category.

3.1.2. Three-Story Building (CWS1 × RWH1)

The results showed that the building supplied by RWH1 also produced more impacts on the environment in all the categories evaluated. For the building, the difference between the impacts produced by the two types of supply varied from 14% to 73% according to the observed category, and the greatest damage was again in the ‘Marine Ecotoxicity’ category (

Table 3. CWS1 × RWH1 (building in a place with a hydroelectricity source).

Environmental Impacts	Building CSW1	Building RWH1
acidification (kg SO2-eq)	1.00 × 10−3	1.86 × 10−3
climate change (Kg CO2-eq)	1.95 × 10−1	4.41 × 10−1
freshwater ecotoxicity (Kg 1.4-DCB-eq)	1.50 × 10−1	2.59 × 10−1
marine ecotoxicity (Kg 1.4-DCB-eq)	4.59 × 102	6.95 × 102
terrestrial ecotoxicity (Kg 1.4-DCB-eq)	1.12 × 10−3	2.36 × 10−3
energy resources: nonrenewable (MJ)	2.31 × 100	9.36 × 100
eutrophication (Kg PO4-eq)	5.32 × 10−4	7.66 × 10−4
human toxicity (Kg 1.4-DCB-eq)	5.00 × 10−1	8.97 × 10−1
material resources: metals/minerals (Kg SB-eq)	7.66 × 10−6	8.87 × 10−6
ozone depletion (Kg CFC-11-eq)	1.77 × 10−8	2.40 × 10−8
photochemical oxidation formation (Kg ethylene-eq)	5.40 × 10−5	1.23 × 10−4

). When comparing the supply complemented by rainwater harvesting (RWH1) for a house and a building, the difference varied between 0 and 54% (Table 2 and Table 3).

3.2. Sensitivity Analysis Regarding the Energy Sources

The power generation source used is often classified as one of the critical subsystems in water supply systems. Therefore, we performed a sensitivity analysis in this section, changing only the initially renewable energy source to a nonrenewable one. As shown in Figure 4, this change caused considerably higher CO₂ emissions for this second option, i.e., 29% higher emissions for the building and 7% higher emissions for the house.

3.3. Comparing the Choices Adopted in Assessing Supply Systems (CWS1 × CWS2) and (RWH1 × RWH2)

The rainwater harvesting system was evaluated as a hybrid system, and the environmental impacts generated by the concrete will be present in the conventional and rainwater harvesting systems. As such, the environmental impacts from concrete will be relevant to both systems and are therefore accounted for in these results.

Considering the concrete used in the construction of the water treatment station and the building’s hydraulic structure for drinking water, the environmental impact can increase by up to two orders of magnitude, depending on the category. Figure 5 below illustrates what happens to the climate change impact category. It is notable, however, that when these materials were accounted for, the distance between the impacts produced by the two types of supply, CWS and RWH, became much smaller (RWH2–CWS2) <<< (RWH1–CWS1).

Notably, although they are less favorable to the environment, the impacts produced by the two types of supply systems can reach very similar values. The reported impacts mentioned by several previous research studies highlight the difference in magnitude between a system at a municipal scale and another at a residential scale [26,27,28,29,31].

However, it is possible to reach very close values to those produced by the municipal supply system depending on the adopted design parameters, as shown in the works of [40]. The hypothetical home study pointed to impacts of the same magnitude for a building using polyethylene reservoirs when all subsystems involved in supply were considered (CWS2 × RWH2).

Figure 6 and Figure 7 present the normalized impact data for climate change, the most urgent focus of attention today, as described in the previous sections.

If the concrete used in the construction of the hydraulic block was not accounted for, in CWS1 and RWH1, the main contributors to the environmental impacts were the treatment and the hydraulic structure of the house, respectively. Comparing the conventional supply system (CWS1) and the one complemented with the rainwater capture (RWH1), it is concluded that the increase in the climate change impact category was caused by the hydraulic structure of the house, which became the main contributing subsystem. In Analysis 1, the treatment subsystem had polyaluminum chloride as its main element, which had already been pointed out by [33]. The hydraulic structure of the house has as its main element the polyethylene if the rainwater reservoir.

Alternatively, when the concrete used in the treatment station construction was accounted for (in CWS2 and RWH2), the treatment subsystem appeared as the most harmful in the climate change category in both types of supply. However, it was observed that the greatest impact caused by rainwater collection had a significant contribution from the hydraulic structure of the house. In Analysis 2, the subsystem had concrete as its main element. Additionally, the hydraulic structure of the house had as its main element the polyethylene of the rainwater reservoir.

The building differed from the house mainly in Analysis 2 (when the concrete used in the treatment station and the hydraulic structure of drinking water and rainwater were included in the analysis). The building’s hydraulic structure was evidenced as the main subsystem contributing to climate change; in the conventional supply, the concrete in drinking water reservoirs was the element that contributed most to environmental damage. Additionally, in the system supplemented with rainwater, the polyethylene from the catchment reservoirs was added to the concrete.

It was expected that energy would be a more relevant subsystem for the building, as it participated both in the conventional supply and in the system complemented with rainwater. It is important to remember that for the house, the energy accounted for by the conventional supply was already considered sufficient for the water to reach the upper reservoir of the residence. In contrast, the building needed to pressurize both the water from the conventional system and the water from rainwater harvesting.

Previous authors [33,34,35,37,42] only accounted for the impact of concrete on the conventional drinking water system for the public domain of the water system, as is already known; however, and this work also accounted for the hydraulic structure, in the private domain, up to the point of use for both the conventional system and the hybrid system, enabling a more realistic result of the system’s impact. It was expected that the concrete structures would produce great impacts, and the present work quantified this; therefore, it is relevant as it highlights a value that will only appear if quantified up to the private domain of the water system.

The quantification of energy impacts, identified as a critical resource both in the conventional supply [33,34,37] and in rainwater collection [35,36,37], was surpassed when analyzing the hydraulic storage structures. Other works accounted for energy as a source needed for the treatment of captured water with UV rays. Some authors calculated [35] the pumping expenses of their systems and balanced them with the distribution savings; for them, the catchment infrastructure appeared as the main contributor, as was the case in this research.

Regarding the CO₂ levels, there is a record of 360 kg of CO₂ for 326 m³ of produced water in [34]. Additionally, [38] considered a system like the one presented in this research, with 1.10 kg of CO₂/m³ of water in the conventional supply system being reported and a record of 0.591 kg of CO₂/m³ of water from the rainwater capture system. With regard to this, the order of magnitude is consistent with the results presented here. Exactly equal values are unlikely, given the unique characteristics of each project (such as the materials and energy source) and each analysis (such as the system boundary).

4. Conclusions

This study focused on applying an LCA analysis to the consumption of materials and energy associated with the conventional water supply system (ACS) and the supply system complemented with rainwater harvesting (RWH)—under typical conditions. The objective was to determine the environmental impacts of these two water supply scenarios at the scale of a house and a building.

At both scales, the hybrid system represented an increase in environmental impacts. However, on a building scale, this increase was very small, and this can be seen as favorable to the hybrid system due to the benefit it brings; the 7% reduction in water demand over the conventional system can represent significant relief in regions that already have water stress as a reality.

Using rainwater as an option for water supply in dwellings helps to decrease the pressure in conventional water supply systems [19,20,37]. However, considering only the absolute values of the environmental impacts produced, a superficial analysis could discourage public policies favoring this ancient technology.

The present work did not present energy as a prominent element, counting it only for pumping. It was then noted that if rainwater is treated, the treatment represents significant energy consumption. The legislation to regulate rainwater harvesting should therefore focus on ensuring that the water reaches only sufficient parameters to ensure its safe use.

The main contributor of impacts in the hybrid system (with the decentralized capture of rainwater) resulted from the consumption of the material for the construction of the reservoirs. Thinking about new materials, reusing and recycling existing ones, and incorporating circular-economy science are options that diminish the impacts. However, a new investigation is required on this matter.

The results found were considered relevant for creating strategies to provide water in large urban agglomerations, characterized by an accentuated densification of buildings, the presence of verticalization, and the near-absence of nonimpermeable soil. The methodology applied in this research presented interesting results. It provided an important contribution to integrating the two supply systems from the point of view of the LCA field. Since the current study presented some limitations, namely the low representativeness of the model, small size, and few components of the system, a consideration of more consolidated integration between rainwater harvesting systems and conventional supply systems is required. Additionally, a validation of the actual model is demanded, considering real-size models for the buildings contemplated. Furthermore, future works should explore broader aspects such as the issue of drainage, which has been little explored here, or the evaluation of a single rainwater harvesting system adapted to the Brazilian reality.

While the stormwater system has been conventionally constructed using traditional infrastructure (e.g., pipe networks, and underground storage tanks), green infrastructure (e.g., rain gardens, green roofs, and permeable pavements) is becoming increasingly popular as it provides greater opportunities for infiltration and a more natural solution for water storage [56]. These technologies need free areas to be implemented but must be considered in water planning.

Water supply is strongly linked to factors such as geography, climate, topography, and the size of the population to be served. Economic factors are usually the most important parameters in the decision-making process when planning policies. However, understanding the needs and the impacts of the most critical subsystems will certainly guide managers in choosing which should be improved when designing projects.

Future studies need to evaluate centralized rainwater harvesting systems, also considering more actual building designs whose lower and upper drinking water reservoirs are no longer made of concrete. Different materials, such as steel or polyethylene, need to be considered for further research. A significant reduction in environmental impacts is expected to be achieved both in the conventional supply system and the hybrid system.