Evaluating Freshwater, Desalinated Water, and Treated Brine as Water Feed for Hydrogen Production in Arid Regions

Abstract

Hydrogen production is increasingly vital for global decarbonization but remains a water- and energy-intensive process, especially in arid regions. Despite growing attention to its climate benefits, limited research has addressed the environmental impacts of water sourcing. This study employs a life cycle assessment (LCA) approach to evaluate three water supply strategies for hydrogen production: (1) seawater desalination without brine treatment (BT), (2) desalination with partial BT, and (3) freshwater purification. Scenarios are modeled for the United Arab Emirates (UAE), Australia, and Spain, representing diverse electricity mixes and water stress conditions. Both electrolysis and steam methane reforming (SMR) are evaluated as hydrogen production methods. Results show that desalination scenarios contribute substantially to human health and ecosystem impacts due to high energy use and brine discharge. Although partial BT aims to reduce direct marine discharge impacts, its substantial energy demand can offset these benefits by increasing other environmental burdens, such as marine eutrophication, especially in regions reliant on carbon-intensive electricity grids. Freshwater scenarios offer lower environmental impact overall but raise water availability concerns. Across all regions, feedwater for SMR shows nearly 50% lower impacts than for electrolysis. This study focuses solely on the environmental impacts associated with water sourcing and treatment for hydrogen production, excluding the downstream impacts of the hydrogen generation process itself. This study highlights the trade-offs between water sourcing, brine treatment, and freshwater purification for hydrogen production, offering insights for optimizing sustainable hydrogen systems in water-stressed regions.

1. Introduction

The global transition toward low-carbon energy systems has positioned hydrogen as a key energy carrier for decarbonizing industries, transportation, and energy storage. Its versatility, high energy density, and compatibility with renewable energy sources make hydrogen an attractive solution for meeting the increasing demand for clean energy and reaching international Net-zero targets. Renewable energy in this paper refers to typical renewable energy resources like photovoltaic, wind, or hydropower and does not include nuclear energy. Global hydrogen demand is projected to increase exponentially by 2050, driven by its critical role in achieving energy transition goals and decarbonization strategies [1]. However, hydrogen production is extremely resource-intensive: it requires substantial quantities of water and energy [2], especially in technologies such as electrolysis or steam methane reforming (SMR). According to an International Renewable Energy Agency (IRENA) and Bluerisk report [3], global freshwater withdrawal for hydrogen production is expected to surge from the current 2.23 billion m³ to 12.09 billion m³ by 2050, with consumption rising from 1.72 billion m³ to 8.32 billion m³. This highlights the increasing water intensity of hydrogen systems and the urgent need for sustainable water strategies in hydrogen production. This raises important concerns about the sustainability of hydrogen production in arid regions, given its significant demand for clean water and energy. In many arid and semi-arid regions around the world, seawater desalination could be the main source for providing freshwater. Desalination is a process that involves extracting dissolved salts and other impurities from water sources like seawater and brackish water [4].

Common hydrogen production methods include green hydrogen, generated through electrolysis powered by renewable energy; blue hydrogen, produced via SMR with carbon capture; and grey hydrogen, derived from SMR without carbon capture. Electrolysis is particularly considered a sustainable solution because it utilizes renewable energy. However, it requires substantial inputs of high-purity water, with an average theoretical consumption of 9 L per kg of hydrogen produced [5,6]. Nnabuife et al. [6] noted that additional water may be required for system cooling purposes. As highlighted in IRENA and Bluerisk [3], approximately 56% of total water withdrawals for electrolysis are used for cooling, equating to around 14.4 L/kg H₂ for proton exchange membrane electrolysis and 18.0 L/kg H₂ for alkaline electrolysis. On the other hand, SMR remains the most adopted method in the world today because it is cost-effective, but it also relies on water [6]. The stoichiometric water demand for SMR is approximately 4.5 L per kg of hydrogen produced in theory, reaching up to 5.85 L per kg, while water use for cooling can add approximately 38 L per kg of hydrogen produced [6]. The assumed values for the pure water requirement are based on the stoichiometric values for the sake of simplification, as consistent specific pure water requirement values for the hydrogen production process were not available. The real values are generally higher, but they vary significantly among different systems. Therefore, while most studies on hydrogen production focus on comparing the environmental impacts of different hydrogen production methods, there is a significant gap in understanding the role of water sourcing and its associated energy in shaping these impacts. This study addresses this gap by examining how the choice of water source, whether freshwater or seawater, affects the environmental footprint of hydrogen production, stressing the need for comprehensive analyses of water systems within hydrogen production frameworks. It also compares the impacts of feedwater between two different hydrogen production methods: hydrogen produced via electrolysis or by SMR, as they are the most common methods of producing hydrogen today. The environmental footprints this study focuses on are the impacts on human health, ecosystem quality, climate change, water scarcity, and marine eutrophication.

The use of seawater versus freshwater as a feedstock introduces distinct challenges, with seawater desalination demanding higher energy inputs and raising concerns about marine ecosystem protection due to brine discharge, while freshwater utilization risks exacerbating water scarcity in arid or semi-arid regions. Furthermore, the inclusion of brine treatment (BT) adds another layer of complexity, as it mitigates some impacts on the marine ecosystem but simultaneously increases overall energy demand. The study also explores how different hydrogen production methods, such as electrolysis and SMR, interact with these variables, shaping the overall environmental footprint, especially in relation to human health, climate change, and marine ecosystem quality. By examining these interconnected factors, the study underscores the importance of balancing energy efficiency, water footprints, and ecosystem preservation to support the development of hydrogen production systems with a lower overall environmental impact. This study is intentionally limited in scope to the environmental impacts of water sourcing and pre-treatment for hydrogen production. It does not assess the hydrogen production process itself (e.g., electrolysis or SMR), ensuring that the analysis focuses specifically on water-related pathways and their associated trade-offs. Further discussion of these scope limitations and associated implications is provided in Section 6.2.

The scope of this work includes three main scenario sets: (1) hydrogen production with seawater desalination without BT, (2) hydrogen production with desalination incorporating partial BT, and (3) hydrogen production using freshwater as the water source. Each category is analyzed across the United Arab Emirates (UAE), Australia, and Spain as representatives of different climatic, electricity mix, and water resource contexts. These three regions were chosen as they represent regions that have a high interest in hydrogen production projects [7] and are amongst the top 10 countries in terms of desalination capacities [8]. Each region can serve as a model for other regions in the world with similar characteristics of electricity mixes, seawater quality, freshwater availability or non-availability, and/or dominant type of hydrogen production method. The scenarios are evaluated in view of a comprehensive life cycle assessment (LCA) perspective. The LCA was conducted using SimaPro software version 9.6.0.1 and the IMPACT World+ impact assessment method, incorporating both midpoint and endpoint indicators to provide comprehensive insights.

The main advances reported by this study include the following:

(a)

A comprehensive analysis of the trade-offs in environmental impacts between desalination and freshwater purification for hydrogen production in arid regions.

(b)

The integration of energy use for brine and reject water management into the LCA framework, an aspect that is critical but often overlooked in existing studies.

(c)

Comparative insights into the environmental effects of using desalination as feedwater for hydrogen production in various regions highlight the role of local factors such as electricity mix and disposal methods.

(d)

Comparing the results between using electrolysis or SMR as the main hydrogen production method enables understanding of the negative environmental impacts of their respective water demands.

The findings provide insights into understanding hydrogen production systems’ impacts in terms of human health, marine ecosystem quality, and climate change for the studied regions that face acute water scarcity and are undergoing energy transitions. This study could contribute to the growing knowledge base on the energy-water intersection with indications toward optimized hydrogen production processes, coupled with environmental impact minimization.

After this introduction and the literature review (Section 2), the methodology is outlined in Section 3. The results and discussion begin with the endpoint analysis in Section 4, followed by the midpoint analysis in Section 5. Finally, Section 6 presents the summary, limitations, and concluding remarks.

2. Literature Review

Several works have studied the LCA of hydrogen production pathways, water footprints of hydrogen, and desalination technologies’ life cycle assessment, offering diverse perspectives on energy efficiency, water usage, and environmental trade-offs. This section consists of three subsections, namely hydrogen LCA, water for hydrogen, and LCA on desalination.

2.1. Hydrogen LCA

Several studies have assessed the life cycle environmental impacts of hydrogen production pathways, offering diverse perspectives on energy efficiency, water usage, and environmental trade-offs. Some studies, such as Dufour et al. [9] highlight that while SMR with carbon capture and storage (CCS) pathways reduce greenhouse gas (GHG) emissions compared to conventional SMR, they may lead to increased burdens in other impact categories such as human toxicity, freshwater ecotoxicity, or resource depletion due to the additional infrastructure and energy required for carbon capture and compression. Therefore, while CCS offers climate mitigation potential, its broader environmental trade-offs warrant careful consideration.

Şahin [10] compared electrolyzer types, noting that proton exchange membrane (PEM) and anion exchange membranes offer high efficiency but rely on rare materials, while solid oxide electrolysis cells (SOECs) require high temperatures but are energy-efficient. These variations highlight the need to consider electrolyzer type in LCAs. This study specifically focuses on assessing water feed for electrolysis and SMR, without assessing the overall hydrogen production process itself.

Shaya and Glöser-Chahoud (2024) presented a comprehensive review of recent LCA studies on hydrogen via electrolysis, identifying gaps related to brine discharge, water source differentiation, and regionalized impact assessment [11]. This study directly addresses these gaps by incorporating partial BT and comparing water sourcing strategies across three regional electricity mixes and water availabilities.

Patel et al. [12] conducted an LCA to compare the climate change impacts of different hydrogen production pathways like SMR, SMR with CCS, thermal decomposition of methane (TDM), and PEM electrolysis, considering natural gas supplied via pipeline and LNG routes. The results demonstrate that hydrogen from PEM electrolysis using wind energy has the lowest emissions, while hydrogen from SMR has the highest, emphasizing the need to account for upstream emissions from natural gas, which are significant enough that only hydrogen from electrolysis meets Renewable Energy Directive (RED) II climate impact limits [12]. However, water sourcing and treatment were not included in their system boundaries, leaving a gap that our study addresses by evaluating the climate impacts of water provision for hydrogen production under different regional and technological contexts.

Kim and Kim [13] assessed electrolysis-based hydrogen in South Korea and found high emissions due to fossil-based electricity. They concluded that grid decarbonization is essential for clean hydrogen, supporting this study’s findings on the importance of regional electricity mixes. This study also assessed the impact of electricity mix from three different countries, to compare the differences in the environmental impacts from their electricity grids on water sourcing for hydrogen production.

Henriksen et al. [14] analyzed trade-offs between water use and GHG emissions across different hydrogen production pathways. They found that SMR with carbon capture increases water consumption due to additional energy requirements, while electrolysis, despite its lower emissions, requires more water overall.

The study by Nnabuife et al. [6] highlights the environmental trade-offs of various hydrogen production methods. It focuses on the scalability, cost-effectiveness, and technological advancements of methods like SMR, electrolysis, thermochemical water splitting, and biomass gasification. SMR is emphasized for its cost-effectiveness but critiqued for its high GHG emissions. Electrolysis is presented as a renewable energy-aligned method with high energy demands. The study identifies that integrating renewable energy sources can enhance the environmental sustainability of hydrogen production [6].

Coraça et al. [15] conducted a long-term integrated assessment of water use, GHG emissions, and costs for low-carbon hydrogen production in Canada, highlighting that hydrogen production could mitigate up to 162 million tons of CO₂ emissions but significantly increase water consumption, reaching 3815 million cubic meters depending on the energy mix and technology.

Moreover, Mehmeti et al. [16] assessed the environmental impacts of various hydrogen production technologies, including SMR, coal gasification, biomass gasification, water electrolysis (powered by both renewable and non-renewable sources), and thermochemical water splitting. Their study examined both specific midpoint impacts and endpoint impacts to provide a comprehensive understanding of water-related impacts and the total environmental footprint of each method. However, the studies by [10,13,14,15,16] did not evaluate water sourcing methods for hydrogen production.

2.2. Water for Hydrogen

The water footprint of hydrogen production was examined by Olaitan et al. [17], who quantified the significant differences in water demand between electrolysis and SMR. Their findings emphasize that water scarcity should be considered when scaling up hydrogen production, particularly in arid regions. However, while their study measured total water use, it did not assess the environmental burdens of specific water treatment processes. This study extends their work by evaluating how different water management strategies influence environmental impacts across various regional contexts.

Mika et al. [18] reviewed seawater treatment technologies for electrolysis, emphasizing the challenges of using untreated seawater due to membrane fouling, corrosion, and energy-intensive pre-treatment. They stress the importance of advanced treatment stages, aligning with this study’s focus on the environmental trade-offs of water sourcing. This study evaluates the energy intensity impacts from the required water treatment processes to provide pure water for hydrogen production.

Maddaloni et al. [19] evaluated treated wastewater as feedwater for SOECs, showing it can reduce environmental impacts and alleviate freshwater stress. Their work supports exploring alternative water sources in hydrogen systems, especially in arid regions. This insight aligns with the broader discussion in the current study regarding the need for context-specific water sourcing strategies and highlights treated wastewater as a promising pathway to reduce both water scarcity and environmental impacts.

Bonnail et al. [20] introduced a new paradigm in brine management by proposing disruptive Zero-Brine discharge technologies, highlighting the need for integrated solutions in regions under environmental stress [20]. However, their work does not quantify the environmental impacts of partial BT systems in hydrogen production scenarios, which this work addresses using LCA.

The paper by Kumar et al. [21] reviews the critical issue of freshwater demands in the hydrogen production process, emphasizing that hydrogen production by 2030 will create an additional demand for 2.1 billion cubic meters of freshwater. The authors discuss alternative solutions, such as treated wastewater, desalination, and direct seawater-fed electrolysis, as viable means to mitigate the pressure on freshwater resources, while identifying significant research gaps for enhancing sustainability and economic feasibility.

2.3. LCA on Desalination

Moreover, numerous studies have examined the LCA of desalination. For instance, Lee and Park [22] investigated a hybrid desalination process integrating reverse osmosis (RO) with hydrate-based desalination, demonstrating that such systems improve water recovery while reducing brine discharge. However, their study did not assess the broader environmental implications of conventional desalination in the context of hydrogen production.

Similarly, Skuse et al. [23] conducted an LCA comparing seawater desalination technologies incorporating graphene oxide (GO)-enhanced membranes for both RO and membrane distillation (MD). Their findings indicate that GO-enhanced membranes reduced environmental impacts by 3–7% in RO and by 27–34% in MD. While RO exhibited lower impacts than MD, the use of renewable heat sources for MD significantly reduced its environmental footprint.

In addition, a study by Lee and Jepson [24] conducted a systematic review of 38 LCA studies covering 295 scenarios to evaluate the environmental impacts of desalination technologies. The main findings indicate that the treatment process and energy sector are the primary contributors to environmental burdens, with potential trade-offs between reducing energy consumption and increasing chemical usage [24]. In contrast, this study evaluates the environmental impacts of seawater desalination and freshwater purification as a function of hydrogen production methods (SMR or electrolysis) and further assesses the implications of incorporating BT.

Khondoker et al. [25] performed a scoping review linking freshwater scarcity, salinity increase, and food production pressures to the urgent need for sustainable desalination methods [25]. Their focus is primarily on agricultural applications; this study complements this by applying LCA to industrial hydrogen pathways involving desalinated and purified water.

While the comparative studies by Nnabuife et al. [6] and Dufour et al. [9] both offer a broad evaluation of hydrogen production methods and their environmental impacts, especially in terms of GHG emissions, this work focuses specifically on the role of water resource selection and management in hydrogen production. This study adds an understudied perspective on the sustainability of hydrogen systems by analyzing the trade-offs between water demands, BT, and brine discharge in different geographical and technological contexts, which are not deeply explored in Nnabuife et al. [6]’s work.

While several studies have examined the environmental impacts of hydrogen production via electrolysis or SMR, few have explicitly assessed the environmental impacts of water supply for hydrogen, particularly desalination with and without BT, within a regionally tailored LCA framework. Moreover, no prior studies have integrated partial BT into comparative hydrogen production pathways using real-world energy and recovery assumptions. This gap may be attributed to multiple factors, including the complexity of modeling partial treatment configurations, limited operational data for intermediate brine treatment processes, and the absence of standardized life cycle inventory datasets for such technologies. Furthermore, research has largely focused on extreme scenarios, either conventional brine discharge or full zero liquid discharge (ZLD), which are easier to conceptualize and often aligned with policy objectives [26]. Additionally, since marine impacts are highly location-dependent and less standardized than global climate metrics, LCA studies often prioritize climate-related indicators over site-specific brine management considerations.

This study aims to fill that gap by quantifying the environmental burdens associated with different water sourcing configurations across three water-stressed regions (UAE, Australia, and Spain), using harmonized system boundaries and inventory assumptions in SimaPro.

3. Methodology

The methodology employed in this study follows an LCA approach, which is a systematic tool for evaluating the environmental impacts of products, processes, or systems across their entire life cycle. In this study, we evaluate the environmental implications of hydrogen production pathways with a focus on the water feed supply processes, including desalination, freshwater purification, and the management of brine and reject water.

3.1. LCA Framework

LCA is a standardized approach for evaluating the environmental impacts of products, services, or systems over their life cycle. It follows the ISO 14040 and ISO 14044 frameworks [27,28] and consists of four main steps:

• Defining Goal and Scope

This step outlines the purpose of the study, the system boundaries, and the functional unit used for analysis. In this work, the goal is to assess the environmental impacts of water supply strategies, namely seawater desalination with and without BT, and freshwater purification, used for hydrogen production via electrolysis or SMR. The system boundary is defined as cradle-to-gate, capturing upstream stages such as water sourcing, purification, and brine or reject water management. The functional unit (FU) is defined as 1 kg of hydrogen produced. This step is described in Section 3.3.1 and Section 3.3.2.

• Life Cycle Inventory (LCI)

The LCI phase involves compiling all relevant inputs and outputs of energy, water, and emissions for each process in the modeled system. In this study, inventory data included recovery rates, energy demands for desalination and freshwater purification, BT energy intensity, and regional electricity mix compositions. These data were obtained from peer-reviewed literature, industrial reports, and the Ecoinvent 3.10 background database [29]. The details of this phase are presented in Section 3.3, Section 3.4, Section 3.5, Section 3.6 and Section 3.7.

• Life Cycle Impact Assessment (LCIA)

This step translates the inventory data into environmental impact indicators. The IMPACT World+ (version 1.04) method was used to assess both midpoint and endpoint impacts. The midpoint indicators include climate change (expressed in kg CO₂-equivalent), marine eutrophication (kg N-equivalent), and water scarcity (m³ water-equivalent), while the endpoint indicators focus on human health (measured in disability-adjusted life years, DALYs) and ecosystem quality (expressed as potentially disappeared fraction, PDF·m²·yr). This phase is described in Section 3.8.

• Interpretation

This final phase involves the analysis and interpretation of the results in order to identify key contributors to environmental impacts, compare trade-offs, and draw conclusions regarding system sustainability. In this study, interpretation was carried out through comparison of 14 scenarios across three countries and two hydrogen production pathways. The results of this interpretation are presented in Section 4, Section 5 and Section 6.

3.2. Application of SimaPro Software for Scenario Modeling and Impact Assessment

The LCA was conducted using SimaPro software version 9.6.0.1 PhD and the IMPACT World+ impact assessment method (version 1.4), incorporating both midpoint and endpoint indicators to provide comprehensive insights. The Ecoinvent (version 3.10) system database was utilized for the LCA in SimaPro. Below is a summary of the steps used to apply SimaPro for this study.

• Scenario Setup and Process Modeling

Each of the fourteen scenarios analyzed in this study was modeled as a distinct process within SimaPro, representing a unique combination of hydrogen production method that the water source supports (electrolysis or SMR), water source (e.g., desalinated water, freshwater), BT configuration (with or without partial treatment), and regional electricity mix (UAE, Australia, or Spain). For each scenario, input flows from nature and the technosphere were entered manually based on region-specific assumptions and published literature.

Each scenario’s input was calculated or inserted manually in an Excel sheet arrangement for data keeping, which was based on assumptions inferred from literature findings or industry reports. Then, these inputs were used in the SimaPro scenario process inputs. For example, in the B2-EL-AU scenario (electrolysis using desalinated water with BT in Australia), inputs included 18 kg of seawater, 1.8 kg of tap water (RO), and 0.081 kWh of electricity—segregated for the electrolysis and BT subsystems. In contrast, the C2-SMR-AU scenario (SMR with freshwater purification) included 5.56 m³ of river water and 5.56 kg of RO-treated water as key material inputs, along with 0.012 kWh of electricity (from Australia’s electricity grid dataset) for the purification process. Background flows were linked to the Ecoinvent 3.10 database using region-appropriate electricity markets (e.g., “Electricity, low voltage [AU]”). No co-products or avoided burdens were modeled, for model simplification reasons. All processes were standardized to a functional unit of 1 kg of hydrogen.

• Impact Assessment Execution

All scenarios were then selected within SimaPro’s comparison interface. The IMPACT World+ (version 1.04) method was applied to evaluate both midpoint and endpoint indicators. The software executed the impact calculations for each scenario based on the complete inventory, generating absolute environmental burdens per functional unit (e.g., kg CO₂-eq, DALYs, PDF·m²·yr). For visualization and relative comparison purposes, normalized results were also generated automatically by the software, scaling impacts so that the highest value in each category was set to 100%.

• Result Export and Visualization

Upon completion of the impact assessment, results were exported as both tables (with absolute values) and charts (normalized comparisons). These outputs were used for post-processing, including the creation of comparative bar graphs shown in Section 4 and Section 5. Results were interpreted across regions and scenarios to examine trade-offs between water sourcing strategies and hydrogen production methods.

This structured use of SimaPro allowed for transparent, repeatable modeling of environmental impacts, aligning with the ISO 14040/14044 standards and consistent with best practices in published LCA literature [9,16,30].

3.3. Goal and Scope Definition

3.3.1. Goal

The goal of this study is to evaluate and compare the environmental impacts of water supply processes for hydrogen production using electrolysis or SMR. Specific emphasis is placed on energy consumption, brine, and freshwater reject disposal, and associated impacts on climate change, water scarcity, and ecosystem quality.

3.3.2. Scope

The study evaluates water feed scenarios for hydrogen production, with a cradle-to-gate system boundary. This includes seawater desalination, or freshwater purification, and management of brine from desalination or purification processes. The hydrogen production processes themselves (electrolysis and SMR) are excluded from the system boundary to focus on the water supply chain impacts.

The functional unit is defined as 1 kg of hydrogen produced. This unit allows for consistent comparisons across the scenarios while accounting for water and energy inputs required to produce 1 kg of hydrogen.

3.4. System Boundaries

As illustrated in Figure 1, the system boundaries of this study include several critical processes. First, the water feed values where water supply processes include seawater desalination for all regions, namely the UAE, Spain, and Australia, as well as freshwater intake and purification specifically for Australia. These processes form the foundation for ensuring an adequate water supply for hydrogen production.

Second, specific energy consumption (SEC) for water supply is considered, focusing on the energy required for desalination, measured in kilowatt-hours per kg of water (kWh/kg water), which varies across regions. Additionally, the energy needed for freshwater purification is included, capturing the variations in energy demand based on water source and treatment requirements.

Third, the system includes brine and reject water management in terms of energy use. This incorporates the energy required for brine discharge into seawater for scenarios involving desalinated seawater. For freshwater scenarios, the energy required for managing reject water through marine discharge into seawater is accounted for, and it ensures a comprehensive understanding of the environmental impacts of water management. In addition to that, the BT is included in one of the scenario sets. The BT dataset models the treatment activities from the point of brine availability at the treatment plant to its treatment and the release of emissions.

Certain processes are excluded from the system boundaries, such as direct emissions or energy consumption associated with electrolysis and SMR hydrogen production technologies, as well as the transportation, storage, and distribution of hydrogen. Moreover, the water requirements for cooling and water losses from evaporation are not considered for this study for both SMR and electrolysis. The energy for the extraction of freshwater and intake of seawater is also outside the scope of this study. These exclusions allow for a focused assessment of water supply processes and their associated environmental impacts.

This study is limited to a cradle-to-gate assessment of water sourcing and pre-treatment for hydrogen production and does not include the hydrogen production stage itself, such as SMR or electrolysis. As such, it does not represent a full LCA of hydrogen production pathways. While the environmental burdens of water provision are important, especially in water-stressed regions, they must be interpreted in the broader context of the total environmental impacts from hydrogen generation. To support this perspective, Section 5.5 provides a comparative overview between the climate change impacts of water sourcing (as modeled in this study) and those associated with hydrogen production based on recent literature. This comparison demonstrates that although the global warming potential (GWP) values from water sourcing are smaller in absolute terms, their relative contribution is more pronounced in low-carbon production pathways such as solar-powered electrolysis. Future research should aim to integrate both water provision and hydrogen generation within a full cradle-to-gate LCA to guide more holistic sustainability decisions.

3.5. Scenarios Analyzed

This study evaluates the environmental impacts of water supply processes for hydrogen production under three distinct scenario sets:

• Scenario Set A. Desalinated seawater for hydrogen production water supply without BT.
• Scenario Set B. Desalinated seawater for hydrogen production water supply with BT of 20% of rejected brine from the desalination process.
• Scenario Set C. Freshwater for hydrogen production water supply as a feedwater source (only analyzed for Australia).

Each scenario is regionally differentiated across the UAE, Australia, and Spain (except the third set of scenarios, which is only studied for Australia’s case) based on certain assumptions on recovery rates, energy requirements for desalination and purification, hydrogen production method, and water management practices. The use of different parameter settings may result in varying impacts, particularly due to differences in impacts on energy consumption. A sensitivity analysis could help understand these impacts from varying parameter settings, so it is planned to be addressed in future work. Due to constraints imposed by the length of this study, sensitivity analysis is excluded from the present study. This study will serve as the foundation for understanding the associated impacts.

The UAE, Australia, and Spain have been chosen in this study for a strategic understanding of water-energy dynamics associated with hydrogen production in varying geographical and socio-economic perspectives. These three countries were chosen because their selection ensures the representation of a few distinguishable sets of scenarios on water resource availability, energy usage, and hydrogen production potential.

The UAE is a leading example of a water-scarce, arid region highly dependent on desalination for its freshwater supplies. The UAE has the second-largest desalination capacity in the world [8,31]. It has a well-developed desalination infrastructure and invests significantly in hydrogen production, making it an ideal case study in the assessment of environmental impacts of integrating desalination into the supply chain of hydrogen. Salinity levels along the UAE coast on the Arabian Gulf have been recorded to range from 36.5 practical salinity unit (psu) to 46 psu, with elevated salinity levels observed in the Ruwais coastal area, which is characterized by a concentration of industrial facilities, including petrochemical refineries [32,33]. Moreover, ongoing efforts in the UAE to switch to renewable energy provide a relevant context within which the compatibility of desalination-powered hydrogen production with sustainable energy objectives can be evaluated. Recently, the UAE electricity mix has been highly dominated by natural gas as an energy source for power generation, standing at around 81% of the total electricity mix [34]. On the other hand, nuclear energy accounts for approximately 13%, solar power at almost 5% and oil at less than 1% [34].

Meanwhile, with Australia making significant investments in renewable energy [35], its electricity mix in 2023 still relied heavily on coal (46.5%), while renewables accounted for nearly 33%, natural gas for 17.8%, and oil for 1.8%. [36]. Despite its available freshwater resources in certain regions, the country faces localized freshwater scarcity, making both desalination and freshwater purification relevant for hydrogen production. Australia maintains a position within the top 10 nations globally in terms of installed seawater desalination capacity [8,31]. Most of the desalination plants in Australia are operating in Western Australia [31]. Typical salinity levels in Western Australia coasts can range between 34.5 and 37 psu [37,38]. Moreover, Australia’s focus on exporting green hydrogen underscores the importance of evaluating the life cycle impacts of water inputs in its hydrogen production processes. Australia is the seventh region with the highest desalination capacity in the world [8], making it important to understand the impacts of desalination use in the country. The scenario of freshwater for hydrogen production water supply as a feedwater source was only analyzed for Australia’s case. This is due to more water scarcity issues present in the other two regions, especially for the case of the UAE, and hence, it means there will be more reliance on desalination.

Spain is also among the top 10 countries globally in desalination capacity [8,31]. The average salinity in the Spanish Mediterranean Sea is 37.7 psu, with fluctuations ranging from below 37 psu to above 38.2 psu [39]. For Spain’s case, around 70% of freshwater is utilized for agricultural use [40]. Increasing freshwater allocation for industrial use could intensify resource competition and exacerbate water scarcity in the region. Therefore, seawater desalination presents a viable alternative to reduce dependence on freshwater resources. Nevertheless, Spain policymakers could still utilize the freshwater scenarios’ results to understand the possible scale of impacts between freshwater and desalination for hydrogen production.

Contrary to the UAE and Australia, Spain’s electricity mix is largely renewable, accounting for 48.5% (of which 15.1% is by solar photovoltaic, 22.5% is by wind, and 10.9% is by hydropower), nuclear energy at about 20%, natural gas at 22.5%, and a small share from oil and coal [41]. Spain’s large and growing renewable energy sector and sustainable energy practices are another aspect that provides a different insight into the impact of different electricity mixes on the environmental impacts of water in hydrogen production.

Freshwater scenarios were not modeled for the UAE and Spain due to severe water scarcity. In these regions, limited freshwater resources are prioritized for domestic and agricultural use, making large-scale allocation for hydrogen production impractical [40,42].

While other countries, such as the United States, China, or Saudi Arabia, are highly relevant in the study context, they were excluded from this study for the purpose of minimizing complexity. Expanding the study to include additional countries would increase the complexity of data analysis and interpretation. The selection of the UAE, Australia, and Spain thus enables a focused analysis of three distinct yet complementary scenarios, thereby assuring that methodological consistency is realized and enabling actionable insights that could guide sustainable practices in hydrogen production across contexts.

3.5.1. Understanding Water Management in Hydrogen Systems

Desalination for Hydrogen Production

Desalination has emerged as a crucial technology for providing water inputs required for industry in arid and water-scarce regions such as the Middle East, North Africa, and parts of Australia [8]. This makes desalination a vital technology for supplying water inputs required for hydrogen production. Of all desalination technologies, RO is the most adopted so far due to its lower energy consumption and technological availability compared with thermal processes such as multi-stage flash distillation (MSF) and multiple effect distillation (MED). The SEC in seawater desalination depends on the level of technology, salinity of seawater, and regional efficiencies, hence ranging between 3.5 and 5.5 kWh/m³ [2,43]. Recovery rates from the desalination process are around 42% if RO is the technology used for desalinating seawater [8], where the rejected water is a highly saline brine that needs to be managed sustainably. According to Kim et al. [44], single-stage seawater reverse osmosis (SWRO) typically operates best at 40–50% recovery, while two-stage systems achieve an optimal 60% recovery, but the additional stage leads to a higher overall SEC compared with single-stage SWRO. For this study, a value of 50% recovery rate among all the seawater desalination scenarios is assumed [44].

Kim et al. [44] also reported that the SEC for SWRO desalination falls between 2.5 and 4.0 kWh/m³, while full-scale SWRO plants, accounting for both pre-treatment and post-treatment, typically require around 3.5 to 4.5 kWh/m³ in most cases for single-stage SWRO. For two-stage SWRO configurations, the SEC can be up to 6.7 kWh/m³ [43]. Martínez-Medina et al. [45] reported that SWRO desalination plants typically consume around 3.5 kWh/m³ of energy, though highly efficient facilities can reduce this to 3 kWh/m³ in certain cases. Park et al. [46] suggest that increasing the recovery rate in SWRO desalination is constrained by the maximum durable pressure that membranes can withstand. Higher recovery rates require elevated operating pressures, which can lead to excessive mechanical stress on membranes and increase the likelihood of scaling, particularly from divalent ions like calcium and magnesium, affecting system efficiency and longevity [46].

Desalination Brine Management

Brine is the highly concentrated saline by-product of desalination processes [47]. Brine discharge is one of the serious issues related to seawater desalination [48]. Brine, accounting for more than half of the feedwater volume based on the recovery rates, is often disposed of by marine discharge with dilution. Marine disposal into the sea is currently the most common disposal practice [2,43,47].

However, marine discharge, besides being practical, is posing environmental risks which include increased salinity, changing chemistry of the water, and even destruction of marine ecosystems. It can have localized impacts, breaking up biodiversity, especially in sensitive coastal environments [49]. The SEC for brine discharge pumping systems in desalination plants ranges from 0.01 to 0.03 kWh per m³ of water produced [50]. For this study, it is assumed that both brine from seawater desalination and freshwater purification are managed by marine discharge into the sea. The brine discharge SEC is assumed to be 0.03 kWh/m³ among all the scenarios with seawater desalination and 0.02 kWh/m³ for freshwater scenarios [50].

BT technologies are classified into membrane brine concentration (MBC) and evaporative brine concentration (EBC) methods. MBC technologies include ultra-high-pressure RO, osmotically assisted RO, low-salt rejection RO, forward osmosis, electrodeionization (EDI), and electrodialysis (ED), which use pressure, osmotic gradients, or electrical potential to separate water from brine [8,51]. On the other hand, EBC technologies, such as membrane distillation, rely on thermal energy to evaporate water and leave behind concentrated brine, but they are typically more energy-intensive and have higher operational costs compared to membrane-based methods [51].

In this study, the brine concentration method to treat part of the reject brine in scenario set B is assumed to be ED, as it is an innovative technology that has high recovery rates and can manage high inlet salinity [47]. The study by Tong and Elimelech [26] discusses ED as a salt-concentrating technology used in ZLD systems and highlights that the SEC of ED for treating high-salinity feedwater (above 15,000 mg/L) ranges from 7 to 15 kWh per cubic meter of feedwater [26]. For this study, a middle value of 10 kWh/m³ is assumed for SEC for BT for all the scenarios with BT [26]. While a value of 10 kWh/m³ was selected as a scenario-based assumption, it is important to note that this does not generalize all BT technologies. Other options, such as advanced oxidation processes or chemical softening, may involve different energy intensities and environmental outcomes. These alternatives could be considered in future comparative analyses.

While BT can increase costs and energy consumption, it is essential to mitigate the severe environmental impacts associated with untreated brine discharge, such as harm to marine ecosystems through increased salinity, habitat degradation, and biodiversity loss [49].

Freshwater Purification

A substitute for desalination, when there are suitable freshwater supplies in the region, is freshwater purification. The major inputs from sources would be rivers, lakes, and municipal water supplies.

There are many treatment methods that could help reach the desired purity of water for hydrogen production. These include processes such as RO, MSF, MED, nanofiltration, ED, EDI, and electrodialysis reversal, which are used for desalination and water purification through membrane filtration, thermal distillation, and electrochemical separation [8].

Freshwater treatment processes, such as RO or deionization, achieve recovery rates ranging from 70% to 81% depending on the feedwater quality and treatment technology [2,8]. Unlike seawater desalination, which faces higher salinity challenges and lower recovery rates, freshwater typically contains fewer dissolved solids, enabling higher recovery efficiencies.

Unlike desalination, freshwater purification is less power-intensive, using from 0.1 to 2.5 kWh/m³ as the minimum energy requirement for conventional water supplies [2]. Recovery rates for freshwater purification systems are about 81% if RO is used for purification, while they can reach up to 97% if we use EDI and 90% if ED is used [8]. For this study, RO is assumed to be used for the freshwater purification process as it requires significantly less energy consumption than the other methods [51].

3.6. Scenario Sets

This section will explain the three main sets of scenarios that are studied in this work: Desalination without BT, desalination with partial BT, and freshwater purification as a feedwater source.

3.6.1. Desalinated Seawater for Hydrogen Production Water Supply Without BT–Scenario Set A

These scenarios represent cases where seawater serves as the primary feedwater source. The seawater undergoes RO desalination to produce ultrapure water required for hydrogen production. This study assumes a single-pass, single-stage configuration for SWRO as the primary desalination method. Brines are discharged directly into marine environments with dilution as the management strategy, and no BT is implemented. The electricity mix database in Ecoinvent (version 3.10) utilized the data from the International Energy Agency World Energy Balances Report [52], which refers to the combination of energy sources used to generate electricity.

The SEC for desalination is assumed to be 3.5 kWh/m³ [45]. The energy consumption for brine discharge into the sea is assumed to be 0.05 kWh per m³ of water disposed, which is insignificantly higher than the literature values range [50]. The key regional differences lie in the recovery rates and energy requirements for desalination.

Table 1. Scenario set A’s assumed values for main parameters.

Scenario	Water Source	Recovery Rate (%)	Desalination SEC (kWh/m3)	Brine Management	Brine Discharge SEC (kWh/m3)
A1-EL-AE: Electrolysis in UAE without BT	Seawater	50	3.5	Marine discharge	0.03
A2-EL-AU: Electrolysis in Australia without BT	Seawater	50	3.5	Marine discharge	0.03
A3-EL-ES: Electrolysis in Spain without BT	Seawater	50	3.5	Marine discharge	0.03
A4-SMR-AE: SMR in UAE without BT	Seawater	50	3.5	Marine discharge	0.03
A5-SMR-AU: SMR in Australia without BT	Seawater	50	3.5	Marine discharge	0.03
A6-SMR-ES: SMR in Spain without BT	Seawater	50	3.5	Marine discharge	0.03
References		[2,43,44,46]	[45]		[50]

describes the main assumptions for the first scenario set, which are denoted as A1-EL-AE, A2-EL-AU, A3-EL-ES, A4-SMR-AE, A5-SMR-AU, and A6-SMR-ES.

3.6.2. Desalinated Seawater for Hydrogen Production Water Supply with BT of 20% of Brine—Scenario Set B

These scenarios introduce partial BT for 20% of the total rejected brine to mitigate the impacts of brine discharge.

BT increases the overall energy demand but offers potential reductions in marine ecosystem impacts. The treated brine is assumed to be recirculated back into the desalination process intake loop. This means that the ED-treated brine was assumed to be blended with the incoming seawater feed to the desalination plant. This assumption was made to reflect a potential reuse pathway that utilizes the treated stream instead of discharging it. As ED treatment typically reduces salinity, mixing it with raw seawater is not expected to raise the overall feed salinity; in fact, it may slightly reduce it, thereby not affecting the assumed energy demand of the RO desalination process significantly. This indicates that the BT is applied to reduce the impacts of seawater quality without any other intentional benefit to our desalination system. Like the first set of scenarios, for the second set we also assume a value of 0.05 kWh/m³ for brine discharge and the case of marine disposal as a management method.

Full BT and ZLD systems are known to significantly increase operational energy requirements, often 2 to 10 times higher than conventional desalination, due to additional thermal concentration or crystallization stages [2,26]. While ZLD can enhance environmental performance by eliminating marine discharge [53], its high energy footprint may outweigh benefits in certain regions unless supported by low-carbon energy sources. Therefore, the 20% brine treatment rate was chosen as a practical and near-term mitigation strategy rather than a full ZLD configuration, which can significantly increase energy consumption and operational costs. ED was selected as the brine concentration method due to its proven applicability for high-salinity brines, moderate recovery potential, and relatively lower energy demand compared to EBC treatment methods. This makes ED a suitable technology for partial treatment scenarios in large-scale desalination plants.

The following scenarios in

Table 2. Scenario set B’s assumed values for main parameters.

Scenario	Water Source	Recovery Rate (%)	Energy for Desalination (kWh/m3)	SEC for BT (kWh/m3)	Brine Treated (% of Brine)	Brine Discharge SEC (kWh/m3)
B1-EL-AE: Electrolysis in UAE with 20% BT	Seawater	50	3.5	10	20	0.03
B2-EL-AU: Electrolysis in Australia with 20% BT	Seawater	50	3.5	10	20	0.03
B3-EL-ES: Electrolysis in Spain with 20% BT	Seawater	50	3.5	10	20	0.03
B4-SMR-AE: SMR in UAE with 20% BT	Seawater	50	3.5	10	20	0.03
B5-SMR-AU: SMR in Australia with 20% BT	Seawater	50	3.5	10	20	0.03
B6-SMR-ES: SMR in Spain with 20% BT	Seawater	50	3.5	10	20	0.03
References		[2,43,44,46]	[45]	[26]		[50]

represent BT scenarios, denoted as B1-EL-AE, B2-EL-AU, B3-EL-ES, B4-SMR-AE, B5-SMR-AU, and B6-SMR-ES.

3.6.3. Freshwater for Hydrogen Production Water Supply as a Feedwater Source (Only Analyzed for Australia)—Scenario Set C

Freshwater scenarios assume surface water (e.g., rivers) as the feedwater source, requiring minimal energy for purification. Reject water from the purification process is managed through discharge into the sea, implying that the RO plant assumes access to both riverine and coastal discharge points. The freshwater recovery rate is assumed to be 81% based on [8].

The energy required for reject water disposal is assumed to be 0.02 kWh/m^3, as it is less saline than rejected brine from seawater desalination.

Table 3. Scenario set C’s assumed values for main parameters.

Scenario	Water Source	Recovery Rate (%)	SEC for Freshwater Purification (kWh/m3)	Reject Water Management	SEC for Reject Water Disposal (kWh/m3)
C1-EL-AU: Electrolysis—Australia with Freshwater	Freshwater	81	1	Surface discharge	0.02
C2-SMR-AU: SMR—Australia with Freshwater	Freshwater	81	1	Surface discharge	0.02
References		[8]	[2]		[4]

below represents freshwater scenarios, denoted as C1-EL-AU and C2-SMR-AU.

3.7. Life Cycle Inventory (LCI)

The life cycle inventory includes key inputs and outputs for each scenario based on data from literature, industrial reports, and SimaPro databases (including Ecoinvent database version 3.10 [48]).

Table 4. Key parameter assumptions for the study in general.

Assumption	Values Assumed	References
Water Requirements for Electrolysis	9 kg of ultrapure water per 1 kg of hydrogen	[5,6]
Water Requirements for SMR	4.5 kg of ultrapure water per 1 kg of hydrogen	[5,6]
Recovery Rates for Desalination—Australia, Spain, and UAE	50%	[2,43,44,46]
Recovery Rates for Freshwater Purification—Australia	81%	[2,43,44,46]
SEC for Desalination (RO)—Australia, Spain, and UAE	3.5 kWh/m3	[45]
SEC for Water Supply for Freshwater Purification	1 kWh/m3	[2]
SEC for Marine Disposal (Brine)	0.03 kWh/m3	[50]
SEC for Marine Disposal (Reject Water)	0.02 kWh/m3	[50]

, as shown below, summarizes the key overall parameters’ assumptions related to the LCI.

These values form the foundation for evaluating the environmental impacts and trade-offs in hydrogen production. Regional seawater quality differences normally would affect these values but as they are much smaller values than the energy values for desalination, the differences can be considered negligible.

3.8. Impact Assessment Method

The environmental impacts were assessed using IMPACT World+ Endpoint and IMPACT World+ Midpoint in SimaPro software version 9.6.0.1. IMPACT World+ is a life cycle impact assessment (LCIA) method that integrates both midpoint and endpoint indicators to assess environmental impacts comprehensively [30,54].

3.8.1. Endpoint Indicators

In IMPACT World+ (version 1.04), the endpoint indicators are categorized under two main areas of protection: human health and ecosystem quality [30,54].

Human Health

This endpoint measures the adverse effects of environmental stressors on human well-being. It captures impacts such as increased morbidity and mortality caused by air pollution, climate change, or contaminated water. Human health impacts are expressed in DALYs (disability-adjusted life years), which quantify the loss of healthy life years due to environmental degradation [30,54].

Ecosystem Quality

This endpoint evaluates the loss of biodiversity and the degradation of ecosystem functions caused by human activities. It considers factors such as habitat destruction, eutrophication, and pollution. The impacts are measured using the potentially disappeared fraction (PDF) of species per square meter per year, reflecting the extent and severity of biodiversity loss [30,54].

3.8.2. Midpoint Indicators

Midpoint indicators in IMPACT World+ serve as proxies for specific environmental mechanisms or stressors within an LCIA. They are located at intermediate stages in the cause-and-effect chain of environmental damage, providing a detailed understanding of potential impacts before they are aggregated into endpoint indicators [30,54].

Based on the IMPACT World+ Midpoint method (version 1.04), the 18 midpoint indicators analyzed in this study include climate change (short-term), climate change (long-term), fossil and nuclear energy use, mineral resources use, photochemical oxidation formation, ozone layer depletion, freshwater ecotoxicity, human toxicity (cancer), human toxicity (non-cancer), freshwater acidification, terrestrial acidification, freshwater eutrophication, marine eutrophication, particulate matter formation, ionizing radiation, land transformation (biodiversity), land occupation (biodiversity), and water scarcity.

All 18 midpoint indicators were assessed using SimaPro with the IMPACT World+ method, but this study focuses on highlighting the three most affected indicators: water scarcity, marine eutrophication, and climate change (long-term), as they showed the highest relevance across the analyzed scenarios.

Marine Eutrophication

Marine eutrophication measures the potential for nutrient enrichment in marine ecosystems, primarily driven by nitrogen emissions from agricultural runoff, brine discharge, and industrial processes. This enrichment leads to excessive algal blooms, depleting oxygen levels in water, and causing harm to aquatic life. In IMPACT World+, marine eutrophication is quantified using kg N-equivalents [30,54].

Climate Change

Climate change evaluates the potential for GHG emissions to contribute to global warming, primarily through the trapping of heat in the Earth’s atmosphere. This indicator plays a crucial role in understanding long-term environmental impacts of GHG emissions. In IMPACT World+, it is expressed in kg CO₂-equivalents [30,54].

Water Scarcity

Water scarcity assesses the impact of freshwater consumption on regional water availability. In IMPACT World+, water scarcity is evaluated using water stress metrics such as m³ equivalents [30,54].

3.9. Assumptions and Limitations

The following assumptions were made for the LCA inventory for this study:

• Electricity Mix: The regional electricity mixes for the UAE, Australia, and Spain were assumed based on IEA reports [52]. This is due to it being the data available in the latest update of SimaPro 9.6.0.1.
• Technological Uniformity: RO technology was assumed for all seawater desalination scenarios and freshwater purification scenarios.
• Brine Management: Marine discharge with dilution was considered standard for desalination scenarios.
• Freshwater Reject Management: Marine discharge was also assumed for freshwater purification scenarios.

The study does not account for technological variability within desalination and purification plants, and regional variability in brine management infrastructure is not included. It also does not include the impacts from electrolysis and SMR processes themselves. Moreover, it does not include the downstream impacts, like hydrogen transport or storage. This study considers only the stoichiometric water requirement for hydrogen production (9 L/kg for electrolysis and 4.5 L/kg for SMR [5,6]) as feedwater. Auxiliary water uses, such as cooling, are excluded from the system boundary. However, literature indicates that cooling can account for the majority of total water consumption. For electrolysis, cooling water demand is approximately 14.4 L/kg H₂ for PEM systems and 18 L/kg H₂ for alkaline systems, representing about 60–70% of total water use [3,50]. For SMR, cooling can add approximately 38 L/kg H₂, contributing more than 85% of total water requirements [3,50]. If desalination were required to supply this additional demand, the associated environmental burdens would significantly increase. Future research should expand system boundaries to include these auxiliary uses for a more comprehensive assessment.

4. Results and Discussion for Endpoint Analysis in Impact World+

This section shares the LCA results and discussion for endpoint indicators, beginning with an analysis of the impacts on human health and ecosystem quality across all scenarios, followed by a comparison of scenario sets to highlight regional and methodological variations. The analysis was performed using IMPACT World+ Endpoint methodologies implemented in SimaPro.

The results highlight the two impact categories available in IMPACT World+ Endpoint Version 1.04 methodology: Human Health and Ecosystem Quality. Figure 2 illustrates the comparative impacts visually.

The vertical axis displays normalized percentages, where 100% represents the scenario with the highest environmental impact for each indicator, providing a baseline for comparison. Lower percentages reflect lower relative impacts compared with the highest value. The chart uses blue bars to represent human health impacts and orange bars for ecosystem quality. The chart is divided into three sets of scenarios, each representing distinct water sourcing and treatment strategies that are investigated in this study.

4.1. Human Health

Impacts on human health are quantified using disability-adjusted life years (DALYs), a metric that represents the loss of healthy life years attributed to environmental degradation [30,54]. Scenario B2-EL-AU, representing scenario of seawater desalination for electrolysis with BT in Australia, registers the highest human health impacts at 100%, attributed to high energy consumption and reliance on coal. Conversely, scenario A6-SMR-ES in Spain, which integrates desalination for SMR without BT, and renewable energy in the electricity mix, reduces human health impacts to 35.7%, demonstrating the effectiveness of sustainable electricity mixes. Scenario C2-SMR-AU in Australia, utilizing freshwater for SMR, achieves the lowest human health impacts (26.9%), reflecting the advantages of freshwater purification over energy-intensive desalination.

The desalination process is highly energy-intensive, driven by the high salinity of seawater in the region. For the case of the UAE in scenario A2-EL-AU, and Australia in scenario B2-EL-AU (using seawater desalination for electrolysis), mark the highest two results due to the high energy demand from desalination and BT. This energy demand is compounded by the reliance on a fossil fuel-dominated electricity mix, resulting in significant GHG emissions that contribute to climate change. Climate change then feeds into the contributors to the impacts on human health in the LCIA.

Spain exhibited significantly lower impacts compared to other regions due to Spain’s electricity mix that incorporates a substantial share of renewable energy sources, such as wind and solar power [55]. This reduces the carbon intensity of the processes involved, as can be seen in Figure 2.

Figure 2 shows that Scenarios using electrolysis had almost double the impacts compared with scenarios using SMR. This is due to the inherent energy consumption from the desalination process or the freshwater purification process. For example, in scenario A1-EL-AE, human health was impacted at a rate of 80.4%, compared to 40.3% in scenario A4-SMR-AE.

4.2. Ecosystem Quality

Impacts on ecosystem quality are quantified using the PDF of species per square meter per year, a metric that represents the extent and severity of biodiversity loss [30,54]. Like the trends observed in human health impacts, scenario B2-EL-AU, representing desalination with BT in Australia, demonstrated the highest impacts on ecosystem quality (100%).

Using desalination involves significant brine discharge into marine environments. This discharge introduces high salinity and chemical contaminants that disrupt marine ecosystems, leading to biodiversity loss and habitat degradation. Furthermore, it involves energy consumption, which contributes to more GHG emissions, amplifying impacts on ecosystem quality. This explains why using SMR had almost only half of the impacts for both ecosystem quality and human health, as can be seen in Figure 2. As an example, for Australia, scenario A2-EL-AU using desalination for electrolysis without BT, the ecosystem quality impacts reduced from a very high percentage of 84.5% to 43.8% when the system was for SMR (in A5-SMR-AU). This is attributed to the significant increase in pure water demand for electrolysis compared to water demand for electrolysis, which are assumed to be 9 kg of pure water and 4.5 kg of pure water, respectively [5,6]. A decrease in water demand leads to a subsequent reduction in the necessity for seawater desalination or freshwater purification. This analysis is applicable across all scenarios in Australia, Spain, and the UAE.

Australia’s energy supply, while transitioning to lower carbon-intensive sources, still relies significantly on coal, unlike the other two regions, which exacerbates environmental pressures on ecosystems through emissions and associated impacts. The unique biodiversity and sensitive marine environments in Australia further amplify the adverse effects of disturbances caused by brine discharge, making ecosystem impacts more pronounced in this region. These combined factors explain why scenario B2-EL-AU resulted in the highest ecosystem quality impacts compared to other scenarios.

4.3. Endpoint Results Comparison Between Scenario Sets

Comparing the scenario sets reveals distinct patterns in endpoint impacts. The second scenario set, desalination with BT (scenario set B), shows consistently high environmental impacts due to the energy-intensive nature of desalination, the environmental burden of direct brine discharge, and the additional energy required to treat brine. This is the case for both human health and ecosystem quality impacts. While BT reduces the direct environmental toxicity of brine discharge, the added energy demands partially offset the benefits. That pattern is observed even when renewable energy is dominant in the case of Spain. For example, scenario B3-EL-ES, utilizing electrolysis in Spain with BT, demonstrated a 79.5% impact on human health, compared to 71.3% in scenario A3-EL-ES, which operated without BT. Similarly, ecosystem quality impacts decreased from 69.1% in B3-EL-ES to 60.5% in A3-EL-ES.

Freshwater scenarios (C1-EL-AU and C2-SMR-AU) outperform most of the other scenarios, as using freshwater for hydrogen leads to significantly reduced impacts on the environment, especially in terms of GHG emissions that affect both human health and ecosystem quality. Scenario C2-SMR-AU (utilizing SMR to produce hydrogen) demonstrates the lowest impacts across all categories, due to lower water demand compared with electrolysis. These comparisons underscore the role of energy consumption and regional resource availability in determining environmental performance.

It is also worth mentioning that scenario C1-EL-AU, utilizing electrolysis and freshwater as feed, had higher impacts than all the SMR scenarios in scenario sets A and B, even when freshwater purification is significantly less energy-consuming than desalination. For example, scenario C1-EL-AU has 22.8% higher impacts than scenario A5-SMR-AU, even though they are both in Australia. This is due to the impact on water availability from the freshwater purification process, which withdraws freshwater from rivers. Water availability reduction contributes to human health impacts in LCIA. Seawater withdrawal for desalination does not impact water availability in the assessment method used (IMPACT World+ Endpoint).

5. Results and Discussion for Midpoint Analysis in Impact World

This section focuses on the results and discussion for key midpoint indicators using IMPACT World+ methodology in SimaPro, starting with a detailed analysis of climate change (long term), marine eutrophication, and water scarcity impacts, followed by a comparison of scenario sets to explore regional and technological differences, and concluding with a trade-off analysis to evaluate the implications of the studied water management strategies.

The three midpoint indicators represent critical facets of environmental sustainability, capturing GWP, nutrient pollution impacts on marine environments, and pressures on freshwater resources. By evaluating these indicators, we also gain insights into the trade-offs and benefits of using different hydrogen production methods under different regional and technological contexts. Figure 3 illustrates the results across all scenarios, highlighting significant regional and process-based variations.

Similar to Figure 2, the vertical axis in Figure 3 displays normalized percentages, where 100% represents the scenario with the highest environmental impact for each indicator, establishing a baseline for comparison. The indicators are color-coded within each bar to distinguish their contributions to the total impact. The blue bars represent climate change impacts, the orange bars represent marine eutrophication, and the green bars represent water scarcity impacts.

5.1. Climate Change

Scenario B2-EL-AU, representing seawater desalination for electrolysis with BT in Australia, exhibits the highest midpoint impact for climate change at 100%. This is primarily driven by the high energy demand of desalination processes and the BT process, coupled with Australia’s coal-dominated electricity mix. The reliance on non-renewable energy sources significantly increases GHG emissions, contributing to global warming. In contrast, scenario C2-SMR-AU with freshwater in Australia achieves the lowest climate change impacts at 26.4%, due to less energy intensity of the freshwater purification process (assumed to be 1 kWh/m³ of water [2] compared with 3.5 kWh/m³ of water [45] for desalination in AU).

Renewable energy integration in Spain in scenarios A6-SMR-ES and B6-SMR-ES also demonstrates significant reductions in climate change impacts, highlighting the critical role of electricity mix decarbonization and reduction of energy consumption in SMR cases. They are both lower the any of the other scenarios in scenario sets A and B. For example, scenario B6-SMR-ES has an impact of 34.7% on climate change, while B4-SMR-AE and B5-SMR-AU had impacts of 44.2% and 50.2%, respectively.

5.2. Marine Eutrophication

Marine eutrophication impacts, driven by nutrient loading into aquatic environments, show the highest levels in scenarios with BT, such as B2-EL-AU, which also scored the highest impact for marine eutrophication (100%). This increase is primarily due to the additional energy demand of the BT process, which intensifies fossil fuel combustion in regions like Australia, where coal remains a significant energy source, and the UAE, with fossil fuel accounting for the main portion of the electricity mix. Higher energy use leads to increased atmospheric nitrogen oxide emissions, which can deposit into aquatic environments, exacerbating nutrient pollution, along with other related emissions that deteriorate the quality of marine water. Additionally, the BT process itself often involves chemical treatments and nutrient removal that may produce byproducts or residuals contributing to eutrophication when not managed effectively. In regions with sensitive marine ecosystems, such as Australia, these effects are magnified due to the vulnerability of local aquatic environments to nutrient imbalances.

This counterintuitive result highlights a critical policy insight: partial BT, while conceptually intended to mitigate marine discharge impacts, can increase global environmental burdens, including marine eutrophication, when energy demand is significant and electricity grids are carbon-intensive. However, this finding is limited to the scope of energy-related and water-related impacts assessed in this study. Future research should incorporate the influence of chemical additives and effluent composition in order to provide a more comprehensive evaluation of marine ecosystem risks.

In contrast, scenarios using freshwater as feedwater, such as C1-EL-AU for electrolysis in Australia, demonstrate significantly lower marine eutrophication impacts (53.0%), compared with 86.0% for scenario A2-EL-AU of Australia using electrolysis with desalinated seawater as feedwater without BT. This emphasizes the positive impact of reduced energy consumption due to freshwater purification requiring only around a quarter of the energy consumption of that of seawater desalination process.

This pattern highlights the environmental tradeoffs between managing brine discharge through BT and the additional energy burden it imposes. Moreover, Australia’s sensitive marine ecosystems amplify the impacts of brine discharge, even with treatment, compared to regions like Spain, which benefit from both lower salinity levels and renewable energy adoption.

While this study evaluates marine eutrophication based on indirect emissions associated with energy use, it is important to acknowledge that several direct environmental risks of brine discharge, particularly those related to marine toxicity and site-specific stressors, are not fully captured by the LCIA indicators applied. Existing literature highlights significant physical, chemical, and biological impacts of brine discharge, including localized salinity stratification, accumulation of toxic pretreatment chemicals, reduced dissolved oxygen, and damage to benthic organisms and sensitive ecosystems such as coral reefs and seagrass beds. To complement the trade-off analysis, a summary table (

Table 5. Brine discharge impacts on marine environment.

Category	Impact	Examples/Notes
Physical	Increased salinity and density of discharged brine alter water column and seabed ecosystems	Brine layers travel kilometers; disrupt benthic circulation
Physical	Temperature changes from warm brine discharge can disrupt local marine temperature balance	Especially relevant in thermal desalination plants
Chemical	Presence of chemical additives (e.g., antiscalants, coagulants) may be toxic to marine life	Polyphosphonates, ferric chloride common in brine
Chemical	Reduced dissolved oxygen from decomposing organic matter can create hypoxic zones	Hypoxic zones affect respiration and biodiversity
Biological	Benthic ecosystem disruption (osmotic stress, species composition changes)	Polychaetes, mollusks, bacteria show stress symptoms
Biological	Damage to seagrass beds and coral reefs due to hypersalinity and temperature changes	Seagrass die-off and coral bleaching observed
Biological	Impaired development and survival of plankton and fish larvae from salinity shock	Larval mortality reported in brine exposure studies
Biological	Bioaccumulation of toxic chemicals through marine food chains	Observed in mollusks and small fish near outfalls

) is provided outlining these effects drawn [49,56,57,58,59,60,61]. These non-LCA environmental burdens, while not quantified in this study, further justify the exploration of BT and set the foundation for future research that integrates brine discharge composition directly into the system boundary [49,56,57,58,59,60,61].

5.3. Water Scarcity

Water scarcity impacts, as measured in the IMPACT World+ Midpoint framework, are exclusively linked to the consumption of freshwater resources, such as groundwater or surface water [30]. As a result, only scenarios utilizing freshwater purification, C1-EL-AU and C2-SMR-AU, are affected by this indicator. In these scenarios, freshwater extraction directly contributes to water scarcity impacts, with scenario C1-EL-AU utilizing freshwater for electrolysis in Australia showing the highest impacts (100%) compared with 50% in scenario C2-EL-AU of freshwater used for SMR. This is due to the marginal difference in pure water demand between electrolysis and SMR, where electrolysis requires almost double the amount theoretically, as shared in Table 4.

This highlights the importance of freshwater resource management in determining water scarcity impacts, particularly in regions where groundwater or surface water is a primary input for hydrogen production processes. This also indicates how different hydrogen production methods can determine the impacts on water scarcity in the regions of interest. Even though SMR reduces the water scarcity impacts by 50%, it uses fossil fuels compared with electrolysis, which avoids using fossil fuels. Therefore, combining CCS with SMR is favorable to reduce associated GHG emissions for avoiding climate change impacts [9].

Scenarios relying on seawater desalination, which are in scenario set A and scenario set B, do not contribute to scarcity impacts, as seawater extraction is not accounted for under this indicator.

While seawater-based scenarios do not register water scarcity impacts under the current LCIA method, this does not imply that seawater extraction is environmentally neutral. The IMPACT World+ water scarcity indicator exclusively addresses freshwater withdrawals and does not account for site-specific ecological effects of seawater intake. These include potential impacts such as impingement and entrainment of marine organisms, alterations to local ecosystems, and thermal or salinity gradients caused by intake systems [8,48]. As these effects are not captured in standard LCIA models, future assessments should incorporate site-specific marine intake impacts to complement water scarcity evaluations and provide a more holistic perspective.

5.4. Midpoint Results Comparison Between Scenario Sets

Comparing the scenario sets reveals distinct patterns in midpoint impacts. Scenario set B (desalination with BT) consistently shows the highest impacts across all midpoint categories when it comes to marine eutrophication and climate change, with Australia scenarios, such as B2, utilizing desalination for electrolysis, performing the worst due to the combination of fossil fuel dependence.

Scenario set C (freshwater purification) outperforms the other sets in most midpoint categories, including marine eutrophication (highly affected by nitrogen oxides emissions) and climate change (highly impacted by CO₂ emissions), due to a significant reduction in energy demand.

The trade-off analysis underlines the intrinsic complexity of the optimal design of water and energy systems aimed at hydrogen production. While desalination provides a stable supply of water in extremely water-scarce countries like the UAE, its high energy intensity and environmental burdens call for more efficient energy use and brine management technologies. Similarly, BT was expected to lower marine pollution but it was outweighed by the additional energy demands and associated environmental impacts.

It is important to note that this conclusion is based on the specific assumption of 20% BT using ED with an SEC of 10 kWh/m³. Other BT technologies or different recovery percentages may yield different environmental results and are not assessed within the scope of this study.

5.5. Impact on Climate Change in Overall Hydrogen LCA Context

This study focuses on the environmental impacts of water sourcing and pre-treatment for hydrogen production, rather than the hydrogen production process itself. However, to contextualize the climate relevance of our findings, we compare the results of this study’s water pathway scenarios with the GWP of full hydrogen production chains reported in [12], as shown in

Table 6. Climate change impact of water feed vs. hydrogen production.

Scenario		Unit	Electrolysis Water Feed (This Study’s Results)	SMR Water Feed (This Study’s Results)	PEM Electrolysis—Solar [12]	SMR [12]	SMR + CCS [12]
GWP from H2 production based on [12]	GWP results from [12]	kg CO2-eq/kg H2			2.5	12.3	7.6
GWP from water feed based on this study results	A1-EL-AE	kg CO2-eq/kg H2 F.U.	0.129
	A2-EL-AU		0.145
	A3-EL-ES		0.104
	A4-SMR-AE			0.065
	A5-SMR-AU			0.073
	A6-SMR-ES			0.052
	B1-EL-AE		0.149
	B2-EL-AU		0.169
	B3-EL-ES		0.117
	B4-SMR-AE			0.075
	B5-SMR-AU			0.085
	B6-SMR-ES			0.059
	C1-EL-AU		0.089
	C2-SMR-AU			0.045

.

A recent study by Patel et al. [12] provides a comprehensive LCA of hydrogen production methods across various regions, including SMR, SMR with CCS, and solar-powered PEM electrolysis. Their results highlight the dominant climate impacts from the hydrogen generation stage, with reported GWP values of 2.5 kg CO₂-eq/kg H₂ for PEM electrolysis powered by solar, 12.3 kg CO₂-eq/kg H₂ for conventional SMR, and 7.6 kg CO₂-eq/kg H₂ for SMR + CCS [12].

In the Impact World+ method used in our study, the equivalent midpoint indicator for GWP is titled “Climate Change”, and this study specifically reports values under the “Climate Change, long-term” category, reflecting GWP from long-term GHG emissions.

These values confirm that hydrogen generation remains the dominant contributor to climate change impacts, particularly in SMR pathways. Nevertheless, the climate change impact from water sourcing and pre-treatment is not negligible, especially for desalination and BT scenarios. In electrolysis, the GWP from water sourcing reaches 5–7% of the total hydrogen LCA emissions, while in SMR it accounts for a smaller share, typically below 1%.

Although the absolute values from water treatment appear small, their contribution becomes increasingly significant at the scale of large hydrogen production plants. As global hydrogen capacity expands, particularly in arid regions relying on desalination, the relative and cumulative climate impacts of water sourcing will become more pronounced. This underscores the importance of evaluating water supply strategies as part of a holistic sustainability assessment for hydrogen production systems.

6. Summary, Limitations, and Conclusions

This section provides a results summary of the regional comparisons and discusses key highlights of the trade-offs, advantages, and challenges associated with different water management strategies and hydrogen production methods across the studied regions. Then, it is followed by identifying limitations and directions for future research, and finally, the conclusion.

6.1. Summary

The study results show large variability in impacts across regions and scenarios. The differences in regional characteristics in terms of electricity mixes and recovery rates resulted in differences in the impacts on the studied endpoint and midpoint indicators. Moreover, there were trade-offs when introducing BT to the systems and when different methodologies were used to produce hydrogen (i.e., SMR or electrolysis).

6.1.1. Summary of Regional Comparisons

Australia and the UAE are highly fossil fuel-dependent countries. Australia showed the highest impacts for most categories, including impacts on human health, ecosystem quality, climate change, and marine eutrophication (especially in scenario B2-EL-AU utilizing desalination for electrolysis with BT). The UAE’s desalination scenario for electrolysis (B1-EL-AE) showed similarly high impacts on human health, marine eutrophication, and climate change. The two mentioned scenarios of Australia and the UAE in desalination for electrolysis dominated all the impact bars overall, contributing to values hovering around 50% more than the other scenarios studied. For instance, in the case of climate change impacts, Scenario B1-EL-AE (UAE, desalination for electrolysis with BT) recorded 88%, while Scenario B4-SMR-AE (UAE, freshwater purification for SMR without BT) demonstrated 44.2%. This means B1-EL-AE contributed double the impact of B4-SMR-AE on climate change.

Equally, for marine eutrophication, Scenario B2-EL-AU (Australia, desalination with BT for electrolysis) showed impacts of 100%, while Scenario C1-EL-AU (Australia, freshwater purification for electrolysis) was at 52.7%, resulting in a nearly 50% higher contribution by B2-EL-AU. These comparisons highlight how desalination-based scenarios for electrolysis, especially in regions like the UAE and Australia, consistently dominate the impact categories compared to freshwater purification scenarios in Australia. Comparing Spain and Australia, scenario B2-EL-AU (Australia, desalination for electrolysis with BT) contributed to almost 30% more ecosystem quality impacts, and about 20% more to human health impacts compared to Scenario B3-EL-ES (Spain, desalination for electrolysis with BT). That reflects the higher reliance on coal in Australia’s electricity mix, which contributes to more GHG emissions and their associated impacts.

As for the scenarios with the lower impacts, Spain showed the least effects for most studied endpoint and midpoint indicators in all scenarios in scenario set A and scenario set B. Spain benefits from a greater share of renewable energy, significantly reducing its impacts on human health, ecosystem quality, climate change, and marine eutrophication. If scenario set C (freshwater scenarios) is included in this comparison, the lowest endpoint and midpoint impacts in general (except for water scarcity) were seen in scenario C2-SMR-AU with freshwater purification for the production of hydrogen via SMR in Australia. For example, for climate change, it recorded 26.4%, which is comparatively 38.9% lower than A5-SMR-AU (43.2%) and 47.4% lower than B5-SMR-AU (50.2%). In marine eutrophication, C2-SMR-AU recorded 26.5%, showing relative reductions of 38.6% compared to A5-SMR-AU (43.2%) and reduction of 47.2% compared to B5-SMR-AU (50.2%). For human health impacts, C2-SMR-AU recorded 23.0%, which is comparatively 47.5% lower than A5-SMR-AU (43.8%) and 54.2% lower than B5-SMR-AU (50.2%). Lastly, for ecosystem quality, C2-SMR-AU recorded 26.1%, demonstrating comparative reductions of 38.7% compared to A5-SMR-AU (42.6%) and about 48.0% less compared to B5-SMR-AU (50.0%). These comparisons highlight the substantial environmental advantages of using freshwater purification with SMR in C2-SMR-AU, as opposed to the more energy-intensive desalination systems employed in the first and second scenario sets. However, utilizing freshwater is not a privilege for all regions. Many regions around the world face water stress, making desalination the only viable solution for freshwater distribution.

6.1.2. Summary of Trade-Offs

The comparison between the three countries shows that the SMR hydrogen production method is about 50% less impactful than electrolysis, for both midpoint and endpoint indicators. This highlights the significant influence of water consumption impacts on the studied indicators in all regions, regardless of the electricity mix. Their findings show that while SMR is cheaper, it generates higher emissions than electrolysis, especially with renewable energy. For cutting carbon intensity in desalination, freshwater purification, or BT, renewable-energy infrastructures should become core investments in the hydrogen systems. This is implied by the scenarios of Spain, which show that it has much less impact compared with the UAE and Australia scenarios.

On the other hand, introducing BT into desalination systems presents distinct trade-offs. While scenario set B (desalination with BT) might support reducing the impacts of brine on the marine ecosystem [49,53], it leads to increased human health and ecosystem quality impacts. For instance, in Spain, ecosystem impacts rose from 60.5% (A3-EL-ES) to 69.2% (B3-EL-ES) due to the added energy intensity of BT, which outweighs its targeted benefits. Introducing BT also shows trade-offs in midpoint impacts.

Midpoint indicators revealed a 3–15% increase in climate change and marine eutrophication impacts when comparing scenarios within the same country, with and without BT. This is due to the added energy demand for BT, which amplifies nutrient-related emissions, such as nitrogen oxides, contributing to eutrophication and added CO₂ emissions attributed to climate change. These results underscore the need for more efficient desalination technologies to minimize brine discharge and balance the environmental trade-offs introduced by BT. The study also indicates that some impacts, such as ecosystem quality, increase despite the potential benefits of BT. Water scarcity in the midpoint assessment is affected only by the freshwater scenarios (scenarios C1-EL-AU and C2-SME-AU). The results show that using freshwater for SMR (C2-SMR-AU) had an almost 50% lower impact compared with the scenario using electrolysis (C1-EL-AU). This is due to SMR requiring almost half of the pure water requirement values for electrolysis.

Although Scenario Set B was developed with the conceptual aim of reducing direct marine environmental risks through partial BT, this benefit remains qualitative in the current study. The LCA results, based on midpoint indicators from the IMPACT World+ method, do not reflect direct marine discharge characteristics, such as salinity, temperature, or brine toxicity. As such, the observed increase in marine eutrophication in BT scenarios reflects upstream energy burdens rather than localized marine relief, highlighting the need for more targeted indicators and boundary expansions in future studies.

While the present study focuses on environmental impacts, it is important to acknowledge that energy consumption associated with desalination BT also translates into significant economic implications. High energy requirements can substantially affect the operational cost of water supply systems, especially in energy-intensive processes like seawater desalination and ED-based BT. These considerations are critical for decision-making and project feasibility in large-scale hydrogen production. Although a full techno-economic assessment is beyond the scope of this work, future research should integrate economic evaluations alongside environmental metrics to support more comprehensive sustainability assessments [62,63].

6.2. Limitations and Future Directions

While this study highlights various environmental impacts, several limitations warrant further investigation:

• Dynamic recovery rates and energy requirements were not explored, which may vary with technological advancements.
• Although the ecosystem quality endpoint in IMPACT World+ encompasses several categories relevant to ecological health, it may not comprehensively capture site-specific impacts of large-scale freshwater withdrawals, such as disruption to riverine flow regimes, aquatic habitat degradation, or local biodiversity loss. These localized ecological consequences, especially in water-scarce or sensitive regions, are important considerations and should be incorporated in future studies using more spatially resolved or hybrid LCA approaches.
• Emerging technologies such as direct seawater electrolysis are gaining attention as a potential future pathway that could bypass the need for extensive pretreatment and brine discharge. While promising, such technologies are still in developmental stages and fall outside the scope of this study but merit exploration in future assessments.
• This study does not include downstream hydrogen purification processes, such as membrane-based separation using palladium (Pd)-based or graphene oxide (GO)-based technologies, which may be required depending on the application and gas purity specifications. These processes, while important for the full hydrogen life cycle, were excluded to maintain focus on the upstream water supply chain. Future studies could expand the system boundary to incorporate purification stages and evaluate their associated environmental burdens.
• This study focused on the theoretical stoichiometric water demand for hydrogen production (9 L/kg H₂ for electrolysis, for example), excluding additional water use for system cooling, pre-treatment, or other operational processes. This exclusion likely underestimates the overall water footprint and associated environmental impacts, particularly in scenarios where cooling water also requires desalination. Future studies should expand the system boundaries to include these auxiliary water demands or conduct sensitivity analyses to capture their influence on impact rankings.
• While the present study focuses on environmental impacts, it is important to acknowledge that energy consumption associated with desalination and BT also translates into significant economic implications. High energy requirements can substantially affect the operational cost of water supply systems, especially in energy-intensive processes like seawater desalination and ED-based BT. These considerations are critical for decision-making and project feasibility in large-scale hydrogen production. Although a full techno-economic assessment is beyond the scope of this work, future research should integrate economic evaluations alongside environmental metrics to support more comprehensive sustainability assessments [62,63].
• Future work could assess full brine recovery options such as ZLD, which were not included here due to their high energy demands and complexity. While this study focused on a feasible 20% ED scenario, upcoming work will compare alternative BT technologies and recovery levels to better capture environmental and operational trade-offs.
• This study assumes a uniform SEC of 3.5 kWh/m³ for SWRO based on typical operational values [45]. However, regional variability exists, with some regions like Spain achieving values below 2 kWh/m³ [64]. Future work will investigate this by conducting a sensitivity analysis on SEC values to capture the impact of technological differences and regional efficiency levels on LCA outcomes. Sensitivity analysis of other parameters, like SEC for BT and desalination recovery rates, will also be investigated, which can enhance understanding of the associated impacts from the system parameters.
• In this study, potential marine benefits of BT were based on qualitative assumptions and not explicitly quantified in the LCA, as the composition and fate of discharged brine were outside the system boundary. Future research will expand the system boundary to include brine effluent characteristics, enabling a more accurate assessment of marine environmental impacts using spatially resolved or composition-specific marine indicators.

This comprehensive analysis emphasizes the need for context-specific solutions and highlights the potential for innovation in hydrogen production feedwater systems to achieve sustainable energy transitions.

Midpoint indicators highlight the nuanced trade-offs in hydrogen production, with freshwater scenarios excelling in energy efficiency but posing challenges for water resource sustainability.

The results of this study highlight the contribution of this research in evaluating hydrogen production through a combined lens of endpoint and midpoint indicators. By integrating brine and reject water management strategies, this analysis provides a comprehensive view of the environmental trade-offs associated with different water sourcing and management strategies. The study’s focus on regional variations and the inclusion of both desalination and freshwater scenarios could represent a significant contribution to the literature on sustainable hydrogen production [5,6].

6.3. Policy Implications

The findings of this study highlight that policies promoting green hydrogen must go beyond the choice of production technology to include water sourcing strategies. In water-stressed regions, incentives and subsidies should encourage the use of low-carbon electricity for desalination and account for the environmental trade-offs associated with brine treatment. Regulatory frameworks should also support technological innovation, such as coupling desalination with renewable energy and developing sustainable brine management solutions, to minimize unintended life-cycle impacts. These measures are critical for aligning hydrogen development with broader sustainability objectives.

Beyond environmental considerations, the selection of hydrogen production systems is influenced by technical and economic factors such as cost-effectiveness, carbon footprint, energy demand, water availability, technology maturity, infrastructure requirements, and local regulatory priorities.

Table 7. Key technical and economic factors influencing hydrogen production strategy selection.

Factor	Description
Levelized cost of hydrogen (LCOH)	Total cost per unit of hydrogen produced, factoring in capital, O&M, water, and energy costs.
Energy intensity—H2 production	Energy required for electrolysis or SMR processes (kWh/kg H2), affecting cost and emissions.
Energy intensity—water supply	Energy needed to supply pure water (e.g., desalination vs. freshwater treatment).
Water footprint	Quantity of freshwater required per kg of hydrogen; critical in water-scarce regions.
Carbon intensity	GHG emissions associated with hydrogen production; critical for Net Zero and climate targets.
Brine or effluent management	Environmental risks associated with waste streams (e.g., brine discharge into marine ecosystems).
Technology readiness	Maturity of electrolysis vs. SMR, and associated water or BT systems.
Grid mix dependence	Sensitivity to regional electricity source (renewable vs. fossil), especially for electrolysis.
Infrastructure availability	Existing capacity and compatibility of water supply and hydrogen production infrastructure.
Regulatory or strategic goals	National or regional targets (e.g., renewable integration, energy independence, water security).

summarizes these key factors, which typically guide policymakers and investors when evaluating hydrogen options in arid regions or countries undergoing energy transitions. While this study focused on environmental impacts and did not include a techno-economic analysis (TEA), policy and investment decisions often hinge on cost elements such as the levelized cost of hydrogen, capital and operational expenses, and infrastructure availability. Future work should integrate LCA with TEA to enable more holistic decision-making in water-stressed and energy-transitioning regions.

6.4. Conclusions

In conclusion, this study has shown that regional electricity mixes and hydrogen production method choices, in addition to BT, source of feedwater, can all have varying impacts on the environmental performance of water sourcing for hydrogen production systems in terms of water scarcity, climate change, marine eutrophication, human health, and ecosystem quality. The results suggest that environmental impacts of water supply for hydrogen production could be reduced by approximately 30% if the system relied more on renewable energy, as demonstrated in the different cases of Spain (scenarios A3-EL-ES, A6-SMR-ES, B3-EL-ES, and B6-SMR-ES). As this study’s LCA system is mostly relying on energy consumption data, it means that a system with a cleaner electricity mix (more reliance on renewable energy, for example) can severely reduce the environmental impacts from water intake systems for hydrogen production. Overall, the findings indicate that in the three studied regions, scenarios employing SMR resulted in approximately 50% lower impacts compared to those utilizing electrolysis.

BT scenarios yielded mixed results. Impacts on human health, ecosystem quality, climate change, and marine eutrophication increased by 10% to 18% due to reduced brine discharge into the ocean, and due to the additional energy demands. It outweighed the benefits of treating the brine due to increased nutrient and CO₂ emissions, contributing to eutrophication and climate change. A key finding of this study is that brine treatment, though designed to address marine discharge concerns, can paradoxically worsen certain environmental impacts, such as marine eutrophication, because of its high energy demand. The current work suggests a base for further research and subsequent policy efforts toward sustainable transition in water and energy within the context of the hydrogen economy. It supports informed decision-making to balance water security, renewable energy integration, and marine life protection, reducing the environmental impacts of hydrogen production. It also clarifies that investing in renewable energy makes water feed processes for hydrogen production less burdensome on the environment. Ultimately, this study underscores that truly ‘green’ hydrogen must account for sustainability across the entire supply chain, from the ocean to the electrolyzer or the methane reformer, and that overlooking the environmental costs of water sourcing risks undermining the goals of decarbonization.