Life Cycle and Local Environmental Impacts of Floating Photovoltaic (FPV) Systems.

Life Cycle and Local Environmental Impacts of Floating Photovoltaic (FPV) Systems - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Search Log in Primary site navigation Search Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice Environ Sci Technol . 2026 May 18;60(21):14949–14960. doi: 10.1021/acs.est.5c17148 Life Cycle and Local Environmental Impacts of Floating Photovoltaic (FPV) Systems Daniela M Seitz †, * , Mark A J Huijbregts † , Sarian Kosten ‡ , Steef V Hanssen † Author information Article notes Copyright and License information † Department of Environmental Science, Faculty of Science, Radboud University, 6500GL Nijmegen, The Netherlands ‡ Department of Ecology, Faculty of Science, Radboud University, 6500GL Nijmegen, The Netherlands * Email: daniela.seitz@ru.nl . Received 2025 Nov 27; Accepted 2026 May 8; Revised 2026 May 7; Collection date 2026 Jun 2. © 2026 The Authors. Published by American Chemical Society This article is licensed under CC-BY-NC-ND 4.0 PMC Copyright notice PMCID: PMC13235550  PMID: 42150774 Abstract Water-floating solar photovoltaics (FPV) add to renewable energy capacity building without additional land requirements. We quantify the environmental impacts of FPV on inland water bodies with a focus on greenhouse gas (GHG) emissions and water consumption. We harmonize and compare all existing peer-reviewed life cycle assessments on FPV and add two case studies. We consider the full life cycle of FPV, including future (prospective) material recycling options, as well as changes in water evaporation and aquatic GHG emissions at the FPV location. On average, we computed an overall GHG footprint of 36 gCO 2 kWh –1 , similar to ground-based PV. Biogenic aquatic GHG emissions at the FPV location are expected to contribute a relatively minor share (1%–8%) to the GHG footprint of the FPV systems. Water savings due to prevented evaporation are much higher than the FPV systems’ water consumption. Collectively, our findings support the continued development and deployment of FPV technology to aid the transition toward renewable energy supply. Keywords: PV, floatovoltaics, solar energy, sustainability, GHG footprint, water scarcity 1. Introduction To limit climate change in accordance with the Paris Agreement, policymakers need to accelerate the transition from fossil fuels to renewable energy technologies such as solar photovoltaics (PV) at the global scale. Floating solar photovoltaics (FPV) – PV panels mounted on water-floating structures – increase utility-scale renewable energy capacity while avoiding trade-offs in land-use such as competition with agricultural food production. , FPV also demonstrates a higher efficiency compared to ground-based PV due to the cooling effect of water and wind. − At least 5.9 GWp of FPV capacity had been installed until the end of 2022 and global capacity is expected to increase, especially in Asian countries. , FPV is associated with a variety of environmental consequences. Two main advantages are the decarbonization potential compared to fossil electricity generation and water savings by preventing evaporation from lakes and reservoirs. Especially in arid regions, the water savings potential due to prevention of water evaporation is considered relevant, − reported at an average of 11 L per kWh electricity generated. Disadvantages include local impacts on water quality due to the decreased light availability and decreased water mixing. , Ray et al. measured an increase in aquatic greenhouse gas (GHG) emissions following FPV installation, primarily due to methane that is produced by microorganisms in anaerobic sediments and released across the water-atmosphere interface. An increase in aquatic methane emissions could reduce net emission mitigation achieved with FPV. Life cycle assessment (LCA) is a tool to quantify the various environmental burdens and benefits of an FPV system’s manufacture, use, and end-of-life phases. Previous LCA research on FPV technology shows GHG footprints of 1194 gCO 2 -eq per kWh. − Manufacture of PV panels has been reported as the largest contributor for many impact categories, including global warming, ecotoxicity, human toxicity, mineral resource depletion and particulate matter formation. Manufacture of the floating support structure can contribute between 17%–39% to the GHG footprint. , Comparison between existing case studies is, however, difficult due to differences in methodological choices regarding the system boundaries and impact assessment methods, as well as technological and spatial variability between the FPV installations, and the limited number of LCAs conducted on FPV. Furthermore, FPV evidently has on-site effects, , like preventing water loss by evaporation or altering local GHG dynamics, but the importance of local impacts in comparison to global life cycle impacts is unclear. Here, we aim to provide an understanding of the environmental sustainability of FPV electricity provision. We quantify the environmental impacts of FPV systems on inland water bodies per FPV system capacity unit (kW alternating current), with a focus on GHG emissions and water consumption. We consider the full life cycle of FPV, including future (prospective) recycling options for PV modules and floating support structures. We harmonize and compare all existing peer-reviewed LCAs on FPV and add LCAs for two FPV case studies in The Netherlands. In addition, we integrate the global GHG and water footprints with the on-site enhanced aquatic GHG emissions and prevented water evaporation. 2. Methods 2.1. Study Overview We assessed the environmental footprints of nine floating PV installations based on the functional unit of 1 kW alternating current (kWac) of FPV system capacity. Additionally, we analyzed the FPV systems per kilowatt hour (kWh) based on the reported simulated energy yields ( SI Table S1–1 ) to allow for a comparison between FPV and other (renewable) electricity sources. We distinguished the following main system components ( Figure ): PV modules, the floating structure (usually a type of plastic float), the panel mounting structure (usually metal), anchoring and mooring (mostly concrete and metal) and the electrical installation (cables and inverters). Collectively, the components that ensure the floating and stability of the PV modules (floaters, mounting and anchoring) are termed “support structure”. For all nine FPV systems, we included the following life cycle stages: manufacture of the system components, assembly and construction of the FPV installation on the water body, operation and maintenance (O&M) during the FPV use phase, expected end-of-life management, and transports within these life cycle stages ( Figure ). For comparability between FPV systems, we harmonized all systems to the construction year 2025 and the decommissioning year 2055, with a lifetime of 30 years. 1. Open in a new tab Overview of the system boundaries of this life cycle assessment study (manufacture of system components, assembly and construction, operation and maintenance, end-of-life) and the FPV system components (photovoltaic modules, electric installation, floating structure, mounting, mooring and anchoring). Environmental impacts such as greenhouse gas (GHG) emissions are shown as outward arrows and the use of resources such as water are shown as inward arrows into the system boundaries. 2.2. FPV Systems Description Table provides an overview of the studied FPV systems. FPV systems NL1 and NL2 are based on aggregated data provided by Solinoor B.V. and have not previously been included in published research. System NL1 has a capacity of 2.6 MWp (Megawatt peak) and is installed on a stormwater retention pond in The Netherlands. System NL2 is a 4 MWp floating park in planning, to be placed on a sand extraction pond in The Netherlands. The other FPV systems are sourced from published literature and are described in detail in their respective studies. Note that systems US, ES and TH are desk studies without actual plans for construction. 1. Overview of the FPV Systems Analyzed in This Study . Code Country Capacity (kWp) Year built Support structure (brand) Panel type Source CH Switzerland 406 2019 • HDPE floater; Aluminum mounting 290 Wp, glass–glass, monocrystalline-Si, bifacial Frehner et al. D Germany 1’479 2021 • HDPE floater 550 Wp, glass–glass PERC, monocrystalline-Si, bifacial Kester et al. NL1 Netherlands 2’576 2023 • HDPE floater; 550 Wp, glass-backsheet PERC, monocrystalline-Si, monofacial This study • aluminum mounting NL2 Netherlands 4’031 2025 • Copolymer floater 610 Wp, glass–glass PERC, monocrystalline-Si, bifacial This study • steel mounting US United States of America 10’000 - • LDPE foam, polyurethane sealant; 110 Wp, IBC module, flexible, monocrystalline-Si, monofacial Hayibo et al. • Concrete anchors, steel chains ES Spain 20’000 - • HDPE floater; ∼550 Wp, glass-backsheet PERC, monocrystalline-Si, monofacial Parascanu et al. • Steel mounting NL3 Netherlands 29’770 2021 • Steel/HDPE 550 Wp, glass–glass PERC, monocrystalline-Si, bifacial Kester et al. TH Thailand 150’000 - • HDPE floater; 270 Wp, polycrystalline-Si, monofacial Cromratie Clemons et al. • Steel mounting ID Indonesia 192’500 2023 • HDPE floater (Hydrelio Air); 560 Wp, glass–glass TOPCon, monocrystalline-Si, bifacial (Jinkosolar) Budi et al. • Aluminum mounting (Sungrow) Open in a new tab a We extracted their location, installed photovoltaic (PV) capacity in kilowatt peak, the year of construction and information about the support structure and PV module types as reported. FPV systems US, ES, and TH are based on desk studies and have not been built. The PV capacity values that are reported in kilowatt peak (kWp) were converted to kWac using a factor of 0.8 kWac kWp –1 , which accounts for inverter and other losses. 2.3. Data Collection and Harmonization For all FPV systems, except the two FPV systems NL1 and NL2, we sourced the inventory data from the respective published literature as specified in Table . For FPV systems NL1 and NL2, we obtained aggregated data from the company Solinoor B.V. The data and our assumptions to derive the full life cycle inventories are documented in Supporting Information (SI) section A.1, B and Tables S2–1 and S2–2 . Next to the baseline end-of-life scenario, we examined recycling of all supporting materials of FPV systems NL1 and NL2. In the first recycling scenario, the system boundaries include dismantling and transports to recycling facilities. A second, full recycling scenario additionally includes processes to manufacture second-grade materials from the support structure components. None of the scenarios allocate any material benefits to FPV, following van der Hulst et al. For the FPV systems previously described in scientific literature, we remained close to the published information, harmonizing the inventories where necessary. Since the inventories for FPV systems ID, TH and CH were not fully available, we used the reported results (environmental impacts) directly, translated them to our functional unit and grouped the components and life cycle stages according to our definitions ( Figure ; SI sections A.6-A.8). Not all reported results were directly comparable with the other FPV systems, leading to gaps and limitations in the analysis, as described in the Results. We reconstructed the published inventories of FPV systems US, NL3, D and ES and linked them to the background processes from ecoinvent v.3.9.1 , ( SI sections A.3-A.5 ). The licensed ecoinvent database provides background life cycle data for materials, energy and processes. Components and life cycle stages were grouped according to our definitions ( Figure ). We included necessary inverter replacements in the O&M phase. We left the transformers out of our system boundaries as part of the grid-connection that is treated as background infrastructure. For all FPV systems except for system US, which has thin-film PV modules for application on a floating membrane, we replaced the originally modeled inventories for PV module manufacture and end-of-life with the state of the art glass–glass or glass-backsheet PERC (passivated emitter rear cell) module inventories according to van der Hulst et al. This harmonization allows focusing on the support structures of the FPV systems, i.e., the components that are characteristic of floating PV. While support structure materials such as bulky floating components or heavy anchoring and metal parts are typically manufactured regionally to reduce costs and logistic issues of transport ( SI section A.1 ), manufacture of the PV modules is specifically assigned to China, following the current production market where 96% of PV modules were produced in Asia in 2024, primarily in China (86%). , 2.4. Local Consequences of FPV Systems We included local consequences on biogenic methane emissions and prevention of water evapotranspiration at the site of installation for all nine FPV systems. Biogenic Methane Emissions We defined local greenhouse gas (GHG) emission impacts as the net changes in aquatic methane (CH 4 ) emissions caused at the installation site. We approximated the baseline (undisturbed) aquatic GHG emissions with CH 4 flux estimates from Rosentreter et al. We estimated the relative aquatic methane emissions increase after FPV installation as 59.2% (see SI section C ) based on Ray et al. During the months after installation, Ray et al. measured higher emissions in shallow ponds with FPV panels, where FPV covered 70% of the surface, compared to open ponds, mainly due to increased release of methane to the atmosphere via bubbles that are formed during anaerobic degradation of organic matter in sediments (ebullition). Given the lack of data on the impacts of FPV on aquatic CH 4 emissions, we applied these findings as the conservative default to all water bodies in our study. We assumed that the relative increase in methane emissions (59.2%) is attributed to the fraction of the surface covered by the FPV system and we scaled the increased emission flux with the FPV system area A FPV . We computed the net change in methane emissions as Δ C H 4 , i = j C H 4 , i · A F P V , i · 59.2 % 1 Where: ΔCH 4,i is the net increase in methane emissions for FPV system i (mgCH4 d –1 ), j CH4,i the baseline methane flux (mgCH 4 m –2 d –1 ) and A FPV,i the FPV installation area (m 2 ) at location i (see SI Table S1–3 ), 59.2% is the relative CH 4 emissions increase attributable to the FPV-covered area, computed based on Ray et al. ( SI section C ). We then integrated the daily median methane emissions linearly over 30 years as the worst-case scenario, and normalized to one kWac for each FPV system. To test the sensitivity of the approximated FPV-induced net aquatic methane gas emission changes and their contribution to the overall GHG footprints, we separately varied all main parameters of the computation. As the lower and upper estimates around the baseline methane fluxes j we computed the 2.5th and 97.5th percentiles based on Rosentreter et al. ( SI Table S1–6 ). We approximated 2.5th and 97.5th percentiles of the methane emission increase based on measurement uncertainties from a methane emission study ( SI–I.1 ). Furthermore, we tested a low estimate of the duration of increased emissions, i.e., 1 year, as the measured immediate emissions increase might not be a permanent change. Water Evaporation We defined local water savings as evaporation prevented by FPV cover of the water surface. We estimated the natural daily evaporation at the FPV locations with eq as reported in Valiantzas, using meteorological data from the National Solar Radiation Database (NSRDB), , and aggregated evaporation monthly ( SI section D ). E = 0.047 · R f · 9.5 + T d − 2.4 ( R f R e t ) 2 + 0.09 · ( 20 + T d ) · ( 1 − 0.01 R H ) 2 Where: E is evaporation (mm d –1 ), R f the radiation flux (MJ m –2 d –1 ), R et the extra-terrestrial radiation (in MJ m –2 d –1 ), T d the mean temperature (°C) and RH the relative humidity (in %) (see SI section D ). We determined the water savings as the difference between the natural evaporation E and the FPV-reduced evaporation over the covered area E c ( eq ). We derived the evaporation reduction relationship from Assouline et al., who showed that evaporation beneath partial surface cover can be approximated as proportional to the square root of the uncovered water fraction (α) (see SI section D ). w a t e r s a v e d , i = ( E i − E c , i ) · A F P V , i = E i ( 1 − α i 1 / 2 ) · A F P V , i 3 Where: water saved,i (L per month) is the suppressed evaporation at FPV location i, E i and E c,i the natural and FPV-covered area evaporation, respectively (mm per month), α i the fraction of the FPV-covered area where evaporation can occur and A FPV,i the area of the FPV installation at location i (m 2 ). We determined the evaporation area fractions (α i ) based on cross-sectional sketches or photos of the FPV systems or types ( Table S1–5 ). We summed the monthly water savings, linearly scaled to 30 years and normalized to one kWac for each FPV system. To test the sensitivity of the approximated FPV-induced evaporation prevention we separately varied the natural evaporation and our estimates for the evaporation area fractions. We used the 2.5th and 97.5th percentiles of the evaporation regression model uniformly as the prediction uncertainty of daily evaporation rates and we tested a range of ±50% for the evaporation area fractions (see SI Table S1–7 ). 2.5. Prospective Approach Since LCA databases like ecoinvent inventories usually lag behind the technological state-of-the-art and changes in energy mixes, we used a prospective approach to more accurately represent the background economy, including decarbonization efforts, for the current situation (2025) and the future at FPV end-of-life (2055). We aligned the ecoinvent v3.9.1 inventories with the integrated assessment model IMAGE using the python package premise v.2.2.9. The IMAGE model considers (future) developments in electricity generation, transport and other sectors based on socio-economic and environmental dynamics. We chose the “middle of the road” shared socioeconomic pathway 2 (SSP2). We used the baseline projection (SSP2-Base) of the year 2025 for the manufacture, construction and O&M phases. For the end-of-life phase in 2055, we implemented the representative concentration pathway (RCP) 2.6 (SSP2-RCP26). RCP2.6 represents a radiative forcing of 2.6 W m –2 , resulting in global mean surface temperature increases of 1.6–1.8 °C by 2100 which aligns with global climate agreements. We used this prospective approach for all FPV systems except for the support structures and on-site impacts of FPV systems CH, ID and TH, for which the inventory data were not fully available and the reported results used conventional LCA modeling (see SI sections A.6-A.8 ). We tested two other prospective scenarios for the end-of-life phase to check the sensitivity of the overall (climate) impacts to the assumption of future pathways. We used the SSP2-RCP1.9 scenario (2055) as a more ambitious mitigation pathway that aims to limit radiative forcing to only 1.9 W m –2 , and the SSP2-Base scenario (2055) as the business-as-usual scenario where economic activities and emissions follow historical trends. 2.6. Life Cycle Impact Assessment We assessed the greenhouse gas footprint for a time horizon of 100 years (GWP100), including hydrogen (H) and biogenic carbon dioxide (CO 2 ), following the IPCC2021 methodology. Additionally, we assessed the greenhouse gas footprint for a time horizon of 20 years (IPCC2021 GWP20, including H and biogenic CO 2 ) to enable comparison with previous assessments of aquatic CH 4 emissions. The water footprint was quantified via the total amount of water consumed over the full life cycle (ReCiPe2016 impact indicator), including the water savings via prevention of water evaporation at the site of the FPV installation. Other environmental impacts at midpoint level, such as fine particulate matter formation, land use and acidification, were also assessed with the ReCiPe2016 methodology. The results for these environmental impacts can be found in SI section F . We computed the environmental impacts of the FPV systems using the python-based LCA software Brightway2 v2.4.7, with the brightway GUI Activity Browser v.2.11.2 and the web-based IDE JupyterLab. 3. Results 3.1. Greenhouse Gas Footprint Climate impacts vary across FPV systems ( Figure ), with the FPV systems TH and CH standing out as the upper and system US as the lower extremes. The membrane-based foam-float FPV system US has a much smaller GHG footprint than systems with a conventional support structure of polyethylene (PE) floaters and metal mounting components, which we attribute mainly to the smaller material use of the lightweight design of FPV system US. The system’s thinner PV modules also have a lower GHG footprint than the other PV modules. At the other end of the GHG footprint range are FPV systems CH and TH, which stand out due to the large contribution of their support structures that contain aluminum mounting systems. We could not obtain the amounts and processes for aluminum production of systems CH and TH, which constrains a comparison with the much smaller GHG footprints of the aluminum mounting structures in systems NL1 and ID. The shares of primary and secondary aluminum used is relevant for the environmental impact, since secondary aluminum reduces the high energy demand of primary aluminum production by 95%. Furthermore, FPV system CH is a relatively small power plant located at high altitude, possibly requiring sturdier and also disproportionately more materials per unit capacity than larger installations, reflecting nonlinear scaling. 2. Open in a new tab Life cycle greenhouse gas (GHG) footprints of the FPV systems. Climate change impacts were modeled with a (a) 100-year and (b) 20-year time horizon with the IPCC2021 life cycle impact assessment methodology. All manufacture and on-site impacts were based on the SSP2-Base background database for the year 2025 and the end-of-life impacts on the SSP2-RCP2.6 background database, except for the support structure and on-site impacts of FPV systems CH, ID and TH, for which the inventories were not fully available and the reported results follow conventional LCA (see SI sections A.6-A.8 ). A comparison with impacts modeled with conventional LCA (see SI section H ), shows that conventional LCA overestimates production and end-of-life GHG emissions of the support structure and electrical installation in all cases by a maximum of 8% vis-à-vis our prospective LCA setting. Across FPV systems, the manufacture causes 54%–84% of overall greenhouse gas (GHG) emissions, followed by the end-of-life phase (6%–37%) and the on-site impacts (4%–12%), which include the local biogenic GHG emissions (1%–8% of overall GHG footprint) ( Figure ). The biogenic GHG emission estimates are linearly dependent on all input parameters and may vary from below zero for a negative factor of relative increase in methane emissions to more than double of the original estimates with separate variation of the input parameters ( Figure S1–28 ). We found that the relative increase factor affects the overall GHG footprints up to ±10%, the baseline emissions change the overall GHG footprints by −1 to −5% for the low and up to +14% for the high estimates, while changing the duration of increased emissions from 30 years to 1 year decreases overall GHG footprints by 1 to 8% ( Figure ). Using a time horizon of 20 years increases GHG emissions across all categories and raises the share of biogenic emissions to 3%–18% ( Figure ). This 20-year horizon gives more weight to shorter-lived GHGs such as biogenic CH 4 , which dominates on-site aquatic GHG emissions during FPV operation. 3. Open in a new tab Sensitivity of overall GHG footprints to the uncertainty ranges of the net changes of aquatic methane emissions. We separately tested sensitivity to the input parameters “baseline aquatic methane emissions”, “relative methane emissions change factor” and the “duration of changed methane emissions”. Within manufacture, PV panels cause the majority of the GHG emissions for most FPV systems (24%–88%), owing to energy-intensive and high-temperature refinement and production processes, like solar-grade silicon cell manufacture. HDPE floaters usually make up the largest part of the support system manufacture GHG emissions, while copolymer floaters as applied in the NL2 installation might offer a possibility for a lower-impact floating system, although this result bears a large uncertainty due to insufficient availability of inventory data. Within the end-of-life phase, PV modules are the major GHG emission, similarly to the manufacture phase. PERC PV module end-of-life impacts are large compared to FPV system US ( Figure ) due to the inclusion of silver and silicon recovery (without awarding material benefits). The overall GHG footprints modeled with our prospective approach are lower than the GHG footprints of FPV with the previously reported PV modules when modeled with the conventional LCA approach that does not take into account future developments in the background economy ( SI Figure S1–26 ). The cleaner energy mix (less carbon-intensive) of the prospective background database for the end-of-life impacts leads to smaller end-of-life contributions to the global warming potential than for other impact categories (see SI Figure S1–9 ). The GHG footprints of FPV systems, except for the system US, range between −11% to −6% for the high mitigation scenario and +9% to +17% for the low mitigation scenario. Due to the minor contribution of the end-of-life impacts to the overall GHG footprint of FPV system US, the choice of the prospective scenario changes the overall GHG impacts by less than 0.1%. The largest difference between the two end-of-life scenarios (baseline and recycling) is due to the floating structure: Polyethylene (PE) (NL1) and copolymer material (NL2, approximated as synthetic rubber) are incinerated in the baseline scenario, leading to direct CO 2 emissions. Avoiding those emissions through recycling decreases the GHG footprint of the support structure and electrical waste by at least a factor of 10 ( SI Figure S1–25 ). Notably, the incineration of PE and rubber also causes much higher GHG emissions than producing recycled HDPE granulate ( SI Figure S1–25-a ) and synthetic rubber ( SI Figure S1–25-b ), respectively. We compared the GHG footprint of FPV to other (renewable) energy technologies on a per kWh basis. Floating PV performs within the same range as ground-mounted utility scale PV, although we found a large range (9–66 gCO 2 -eq kWh –1 , see Figure ) and added GHG emissions due to the manufacture of HDPE floaters and other supporting elements. We also need to assume that the environmental impacts of the energy technologies shown in Figure would be lower if assessed in 2025 than in the reference year 2015. FPV system US has the lowest GHG footprint out of all FPV systems with 9 gCO 2 -eq per kWh. It has low GHG emissions per installed capacity ( Figure ), and is also in a location with high irradiance and, therefore, high electricity yields. Generally, FPV systems located in areas with higher irradiance (US (Arizona), ES (Murcia, Spain) and ID (West Java, Indonesia)), report higher estimations for the energy yield (kWh per kWp per year, see SI Table S1–1 ), resulting in lower GHG footprints per kWh than the FPV systems in central Europe (Netherlands, Germany, Switzerland). Overall, FPV’s GHG footprint falls within the same order of magnitude as ground-based PV and wind energy, meaning that FPV energy could help mitigate GHG emissions by replacing fossil sources like coal and gas in the energy transition. 4. Open in a new tab GHG emission footprints of FPV systems compared to other energy sources. GHG footprints were determined in gCO 2 -equivalents per kWh for the harmonized FPV systems based on the IPCC2021 “GWP100, including H and biogenic CO 2 ” (horizontal bars) and contrasted to GHG footprints of other energy sources as determined by Hertwich et al. All manufacture and on-site impacts were based on the SSP2-Base background database for the year 2025 and the end-of-life impacts on the SSP2-RCP2.6 background database, except for the support structure and on-site impacts of FPV systems CH, ID and TH, for which the inventories were not fully available and reported results follow a conventional LCA. Hydro (low) and (high) refer to a low and high GHG footprint found for two hydropower reservoirs, wind represents the range between off-shore and on-shore wind turbines, PV CdTe, PV CIGS and PV poly-Si are the ranges for ground-mounted and rooftop-based Cadmium Telluride, Copper Indium Gallium Selenide and polycrystalline Silicon PV cell technologies, respectively. CSP represents the range between trough and tower concentrated solar power. Natural gas refers to a natural gas combined cycle power plant, and coal to the range for coal power plants (subcritical pulverized coal, integrated gasification combined cycle power plant (IGCC), supercritical pulverized coal). Fossil with CCS comprises natural gas and coal with carbon capture and storage. 3.2. Water Footprint The FPV water consumption impacts (ReCiPe2016 indicator) amount to 7 to 14 m 3 kWac –1 ( Figure a), stemming largely from cooling and other activities during the manufacture phase. Within that, PV panel production (water cooling in the energy-intensive PV manufacture process) makes up at least 50% of the water consumption across FPV systems. Prevented water evaporation, presented as negative values in Figure b, is in a range of 20 to 637 m 3 kWac –1 , and in relative terms from 2 to 90 times the life cycle water consumption. The majority of the water consumption is depriving a different location than where water can be saved locally, so prevented evaporation is not considered a direct mitigation of water consumption. 5. Open in a new tab Life cycle water footprints of FPV systems: (a) water consumption of the FPV systems; (b) prevented water evaporation on-site and sum of water consumption. Positive values correspond to the life cycle water consumption potential of FPV as determined with the LCA method ReCiPe2016. Negative values represent the evaporation prevented due to FPV at the installation sites. All manufacture and on-site impacts were based on the SSP2-Base background database for the year 2025 and the end-of-life impacts on the SSP2-RCP2.6 background database, except for the support structure and on-site impacts of FPV systems ID and TH, for which the inventories were not fully available, and the reported results follow a conventional LCA. Compared to our prospective setting, conventional LCA underestimates production and end-of-life water consumption of the support structure and electrical installation in all cases on average by about 20% (see supporting data). (*) water consumption impacts of FPV system CH were not reported. The savings achieved with prevented evaporation depend on geographic location (irradiance and temperature), but also on the assumed evaporation area fraction. For example, the prevented evaporation for FPV systems NL1, NL2 and NL3, which are all located at a similar latitude within The Netherlands differs by an order of magnitude (from 20 m 3 kWac –1 for NL2 to 180 m 3 kWac –1 for NL1). Varying the evaporation area fraction affects the evaporation prevention estimates of conventional FPV systems inversely by −100% to +250%, driven by the relatively large uncertainty range of ±50% ( Figure ). For FPV systems with low water contact, an increase of the evaporation area fraction to up to 1 (i.e., no water contact) reduces the computed water savings to zero (i.e., −100% evaporation prevention), e.g., for NL1, NL2 and CH. A variation of the natural evaporation with the 95% prediction uncertainty of the daily evaporation causes changes of ±17% ( Figure ). 6. Open in a new tab Sensitivity of prevented evaporation estimates to the uncertain input parameters. We separately tested the uncertainty ranges for the parameters “evaporation from the uncovered water body” and the “evaporation area fraction per FPV area”. 4. Discussion 4.1. FPV in the Context of Other Energy Sources Conventional photovoltaic energy has a substantial greenhouse gas (GHG) mitigation potential. With reported emission factors of 40 to 50 gCO 2 per kWh generated electricity, PV electricity can significantly reduce the electricity mix’ GHG emissions, which averaged globally at 445 gCO 2 kWh –1 in 2024. For FPV systems with PERC solar panels using crystalline Silicon, namely all FPV systems except system US (thin-film PV), we found an average GHG footprint of 40 gCO 2 -eq kWh –1 (20 to 66 gCO 2 -eq kWh –1 ), exceeding ground-based and rooftop-based polycrystalline Silicon GHG footprints in only two cases ( Figure ). Thin-film PV technologies (CIGS and CdTe) halve the average GHG footprint of FPV. FPV systems do require more materials than rooftop and ground-mounted systems due to the floating support structure. On the other hand, FPV systems require less land than ground-based utility PV and thus, eliminate competition with agriculture, reducing land-use change associated emissions and lowering impacts on terrestrial biodiversity. From that perspective, roof-based PV installations would be preferrable over both ground- and water-based PV. As transformers and associated grid-connection infrastructure were excluded from the LCA system boundaries, strict comparability with utility-scale ground-based PV is limited. A consistent comparison would require expanding the system boundaries and defining the functional unit as AC electricity delivered to the grid. Despite such limitations, our study indicates that FPV is a viable addition to other PV technologies and wind energy in terms of GHG mitigation potential of the energy mix, even when adding aquatic GHG emissions. Furthermore, FPV ensures low land-use impacts and water savings benefits. Further assessment of (on-site) aquatic biodiversity could strengthen the comparison of biodiversity effects between PV technologies. Previous research has estimated that installing FPV on 10% of reservoir surfaces in the EU would generate approximately 140 TWh of energy, covering 7% of the EU’s electricity demand. Renewable energy sources such as FPV cannot yet continuously replace fossil energy in the grid. We used kWac as the functional unit to compare the environmental impacts of FPV technologies. The functional unit of kWh is needed for system-level decision-making which also requires careful consideration of intermittency and temporal integration. Globally, there is large potential and demand to install FPV on hydropower reservoirs, creating synergies like availability of electrical infrastructure and the mitigation of intermittency issues by using the solar energy locally to pump water. Furthermore, reduced water evaporation due to FPV increases hydropower capacity. , Upscaling the environmental impacts computed for the FPV systems in this study to a global scale would require additional LCA studies that appropriately represent the FPV market, as the sector’s growth is most prominent in Asian countries, particularly China, often with large-scale FPV systems. 4.2. Life Cycle Environmental Impacts of FPV We identified the PV module (manufacture and waste management) as the major contributor to many impact categories, causing higher environmental impacts than the floating support structures for most FPV systems. Due to the prospective background databases for current and future impacts, the total climate impacts reported in our study are on average lower than the impacts computed with conventional LCA. Our quantitative assessment of alternative prospective background scenarios for the end-of-life stage showed that the policies and actions taken to mitigate climate change and environmental pollution will likely impact the overall GHG footprint of long-lived renewable energy sources such as FPV. Other environmental impacts such as human toxicity and ecotoxicity become more relevant during end-of-life management than in production and are not directly influenced by the prospective LCA approach. There are a number of technical uncertainties of FPV that we did not consider in the present study. For example, the same conversion efficiency is assumed for FPV and ground-based PV, while Naushad et al. reported a higher conversion efficiency for FPV due to the cooling effect of the water body. Another difference we did not consider is the potentially accelerated degradation of FPV panels and support structures due to biofouling caused by algal growth and bird droppings, which may reduce conversion efficiency and decrease the system’s lifetime. We expect that assumptions regarding degradation rates and material lifetimes have less of an effect for the environmental impacts per kWac than inventory choices regarding processes and material and energy intensity. We recognize combined impacts of these choices implicitly in the large inter-system variability that dominates the environmental impacts per kWac even for similar FPV designs. While the degradation rate and material intensity influence the environmental impacts per kWh to a limited extent, environmental impacts per kWh scale inversely proportional with the material lifetime. Assuming 40 instead of 30 years of FPV lifetime and energy generation would decrease the environmental impacts of production by about 25%. More long-term research and practical experience is needed to accurately consider these effects and achievable material lifetimes for FPV. Another source of uncertainty in the life cycle assessment methodology is the choice of the geographical regions, which determines the electricity mix and, therefore, the carbon intensity of a process. In our study, we followed current market trends and realistic engineering choices. Alternative modeling choices regarding the geography would change transport means and distances, and, more importantly, could alter the composition of environmental impacts: e.g., manufacture of PV modules in a greener energy mix has the potential to significantly change the overall GHG footprint due to its large share. However, choosing to model PV module manufacture in Europe instead of China would entail more detailed analysis, since China controls the vast majority of supply chain components. Our results point to the technical challenges in recycling that need to be solved to comply with aspired circular economy goals, e.g., with the EU’s circular economy action plan. On the support structure level, a significant difference in GHG emissions stems from incineration vs recycling of floating compounds, showing that an increased HDPE recycling rate could lower GHG emissions. Our full recycling scenario does not award recycled material benefits to the FPV life cycle. Nevertheless, it implies that environmental impacts of the FPV support structure could be significantly reduced by using more recycled materials in manufacture. Whether, and to what extent, recycled materials can replace primary materials in FPV support structures or would be diverted to other markets depends on the material quality achievable with available recycling technologies, as recycled materials are typically downcycled and of lower quality. Our findings indicate that PV module waste management needs to be tackled to reduce overall environmental impacts of PV energy provision, such as human toxicity impacts. PV module recycling will become increasingly relevant as the amount of decommissioned solar panels grows exponentially over the next years and decades. 4.3. Local Impacts of FPV Systems We found that local evaporation reductions are significantly larger than the life cycle water consumption for all FPV systems. The mean evaporation savings under consideration of the average yearly energy yields of the FPV systems are 3.5 L kWh –1 with a standard deviation of 2.8 L kWh –1 , somewhat lower than previously reported FPV-induced evaporation savings at 11 L kWh –1 . Our estimates are highly sensitive to the visually determined evaporation area fraction, emphasizing the importance of reporting the dimensions of FPV installations. Furthermore, our approach likely underestimates the magnitude of evaporation reductions, as reduced wind speeds and reduced net radiative input due to shading would lead to lower evaporation rates from the water surface underneath the FPV panels. ,,− Indirect heat trapping effects could partially offset these other thermodynamic effects. Physical (process-based) modeling might improve the estimate of prevented evaporation by computing the altered evaporation from FPV-covered water bodies based on altered thermodynamic parameters, rather than relying on the evaporation area fraction (α). For FPV systems that cover a significant share of the water surface and obstruct wind and radiation to a larger extent – with a membrane-based FPV system such as system US as an extreme example – our computation approach might compare better to modeling evaporation in the FPV-induced microclimate than for FPV systems with a more open support structure and low α values. FPV offers a multi-output alternative to other evaporation-preventing measures. , In water scarce regions, reduced evaporation could be important for downstream water availability as well as for hydropower provision. − In contrast, manufacture of FPV systems contributes to water scarcity at other locations than the FPV installation location. In future work, regionalized LCA could determine these water scarcity (prevention) impacts, e.g., using the AWARE (Available WAter REmaining) method which weighs the absolute water consumption based on the likeliness of water deprivation in specific regions between 0.1 and 100. Across our FPV system locations, prevented evaporation in The Netherlands (NL1-NL3) is weighted the lowest with average annual factors as low as 0.9, while the prevented evaporation in the water-scarce region of FPV system ES (Spain) would be weighted with up to 100. This range of weighting factors over 2 orders of magnitude may alter the ranking of water savings benefits. Regionalized assessment of the evaporation prevention and the water consumption may also shift the net water footprints. A water scarcity-based assessment could help policymakers and practitioners evaluate where to deploy FPV and from where to source materials by providing a regionally specific interpretation of the water stress caused or avoided by the supply chain of an FPV installation. Our approach linearly scales FPV-induced changes in biogenic GHG emissions to the FPV area and may not reflect long-term effects over 30 years. The immediate FPV-induced increase in aquatic methane emissions observed by Ray et al. might be caused primarily by plant dieback and decomposition, after which methanotroph-driven attenuation of diffusive methane emissions could stabilize the new baseline emissions at a lower level. On the other hand, a loss of macrophytes in shallow ponds could signify a lack of rooting in the sediment, a reduction in radial oxygen loss and a lower CH 4 oxidation, leading to a buildup of CH 4 in the sediment given an organic carbon influx exists, which might sustain higher CH 4 ebullition. , Our sensitivity analysis shows the possible extent of such uncertainties. Further uncertainty is introduced by lack of data quantifying FPV-induced GHG emission dynamics in deep vs shallow water bodies or different climates. Deep water bodies would likely be affected differently by FPV systems than shallow lakes. FPV systems might affect stratification, but the direction and extent of these effects depend on factors such as FPV coverage and reservoir depth. , Prolonged stratification and limited oxygen replenishment might enhance hypolimnetic CH 4 accumulation, potentially increasing total CH 4 emissions, particularly via ebullition, and upon mixing. , To strengthen the methodology, more evidence of long-term FPV-induced biogenic CH 4 emission dynamics is needed. Given projected increases in CH 4 emissions from inland water bodies, resolving how FPV alters light, oxygen dynamics and microbial CH 4 cycling becomes critical for constraining the aquatic GHG implications of FPV systems and, consequently, for increasing their overall GHG emission mitigation potential. Two key local effects of an FPV installation are the (partial) reduction of wind mixing and of sunlight intensity in water, which can affect aquatic plants and other ecological aspects, ,, including mitigation of algal blooms in eutrophic aquatic environments. This may affect local eutrophication, acidification, toxicity and biodiversity. Quantifying these local environmental impacts and comparing them to the respective impacts determined through LCA could further clarify FPV’s overall environmental sustainability, but requires more practical research at FPV installation sites, since FPV field measurement research is scarce. Future work could translate measured physical changes (e.g., light availability, oxygen dynamics) and resulting ecological responses (e.g., primary production) into causeeffect relationships, enabling inclusion of FPV’s influence on eutrophication and aquatic biodiversity in an LCA context. Supplementary Material es5c17148_si_001.pdf (3MB, pdf) es5c17148_si_002.xlsx (388.7KB, xlsx) Acknowledgments The authors thank Rutger Schreur from Solinoor B.V. for providing aggregated data from two FPV projects, which enabled the life cycle assessment conducted in this study. We also thank the NWO SPARKLES project consortium, particularly Sven Teurlincx, for forming the consortium and sparking FPV discussions. All data sets generated during this life cycle assessment study and source data for figures are openly available in DANS (Data Station Physical and Technical Science) at 10.17026/PT/LMABM6 . The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.est.5c17148 . Life cycle inventory harmonization steps, end-of-life scenarios explanation, biogenic methane emission estimation and assumptions, water evaporation reduction calculations, additional life cycle impact assessment results ( PDF ) Life cycle inventories of the floating photovoltaic systems and solar panels analyzed in this study ( XLSX ) Research reported in this study was funded by the Dutch Research Council (NWO) as part of the SPARKLES project. The authors declare no competing financial interest. References United Nations . Paris Agreement; United Nations Framework Convention on Climate Change, Paris, France, 2015. https://unfccc.int/sites/default/files/english_paris_agreement.pdf . [ Google Scholar ] Cagle A. E., Armstrong A., Exley G., Grodsky S. M., Macknick J., Sherwin J., Hernandez R. R.. The land sparing, water surface use efficiency, and water surface transformation of floating photovoltaic solar energy installations. Sustainability (Switzerland) 2020;12(19):8154. doi: 10.3390/su12198154. [ DOI ] [ Google Scholar ] Sahu A., Yadav N., Sudhakar K.. Floating photovoltaic power plant: A review. Renewable and Sustainable Energy Reviews. 2016;66:815–824. doi: 10.1016/j.rser.2016.08.051. [ DOI ] [ Google Scholar ] Dörenkämper M., Wahed A., Kumar A., de Jong M., Kroon J., Reindl T.. The cooling effect of floating PV in two different climate zones: A comparison of field test data from the Netherlands and Singapore. Sol. Energy. 2021;214:239–247. doi: 10.1016/j.solener.2020.11.029. [ DOI ] [ Google Scholar ] Cazzaniga R., Cicu M., Rosa-Clot M., Rosa-Clot P., Tina G. M., Ventura C.. Floating photovoltaic plants: Performance analysis and design solutions. Renewable and Sustainable Energy Reviews. 2018;81:1730–1741. doi: 10.1016/j.rser.2017.05.269. [ DOI ] [ Google Scholar ] Liu H., Krishna V., Lun Leung J., Reindl T., Zhao L.. Field experience and performance analysis of floating PV technologies in the tropics. Progress in Photovoltaics: Research and Applications. 2018;26(12):957–967. doi: 10.1002/pip.3039. [ DOI ] [ Google Scholar ] Wu R., Ma C., Xu Z., Deng Z., Liu Z.. Heat dissipation effect of the floating photovoltaic for an inland water test site in China. Energy Conversion and Management. 2024;322:119177. doi: 10.1016/j.enconman.2024.119177. [ DOI ] [ Google Scholar ] Rodríguez Gallegos C. D., Gandhi O., Huixuan S., Paton C., Ji Z., Yacob Ali J. M., Alvarez-Alvarado M. S., Zhang W., Rodriguez Gallegos C. A., Chua L. H. C., Reindl T.. Global Floating PV Status and Potential. Progress in Energy. 2025;7:015001. doi: 10.1088/2516-1083/ad9074. [ DOI ] [ Google Scholar ] Tiismus H., Maask V., Astapov V., Korõtko T., Rosin A.. State-of-the-Art Review of Emerging Trends in Renewable Energy Generation Technologies. IEEE Access. 2025;13:10820–10843. doi: 10.1109/ACCESS.2025.3528640. [ DOI ] [ Google Scholar ] Niccolai A., Grimaccia F., Lorenzo G. D., Araneo R., Ughi F., Polenghi M.. A Review of Floating PV Systems With a Techno-Economic Analysis. IEEE Journal of Photovoltaics. 2024;14(1):23–34. doi: 10.1109/JPHOTOV.2023.3319601. [ DOI ] [ Google Scholar ] Bontempo Scavo F., Tina G. M., Gagliano A., Nižetić S.. An assessment study of evaporation rate models on a water basin with floating photovoltaic plants. International Journal of Energy Research. 2021;45(1):167–188. doi: 10.1002/er.5170. [ DOI ] [ Google Scholar ] Rosa-Clot M., Tina G. M., Nizetic S.. Floating photovoltaic plants and wastewater basins: An Australian project. Energy Procedia. 2017;134:664–674. doi: 10.1016/j.egypro.2017.09.585. [ DOI ] [ Google Scholar ] Ranjbaran P., Yousefi H., Gharehpetian G. B., Astaraei F. R.. A review on floating photovoltaic (FPV) power generation units. Renewable and Sustainable Energy Reviews. 2019;110:332–347. doi: 10.1016/j.rser.2019.05.015. [ DOI ] [ Google Scholar ] Jin Y., Hu S., Ziegler A. D., Gibson L., Campbell J. E., Xu R., Chen D., Zhu K., Zheng Y., Ye B., Ye F., Zeng Z.. Energy production and water savings from floating solar photovoltaics on global reservoirs. Nature Sustainability. 2023;6(7):865–874. doi: 10.1038/s41893-023-01089-6. [ DOI ] [ Google Scholar ] Liu Z., Ma C., Li X., Deng Z., Tian Z.. Aquatic environment impacts of floating photovoltaic and implications for climate change challenges. Journal of Environmental Management. 2023;346:118851. doi: 10.1016/j.jenvman.2023.118851. [ DOI ] [ PubMed ] [ Google Scholar ] Ziar H., Prudon B., Lin F. Y., Roeffen B., Heijkoop D., Stark T., Teurlincx S., de Senerpont Domis L., Goma E. G., Extebarria J. G., Alavez I. N., van Tilborg D., van Laar H., Santbergen R., Isabella O.. Innovative floating bifacial photovoltaic solutions for inland water areas. Progress in Photovoltaics: Research and Applications. 2021;29(7):725–743. doi: 10.1002/pip.3367. [ DOI ] [ Google Scholar ] Ray N. E., Holgerson M. A., Grodsky S. M.. Immediate Effect of Floating Solar Energy Deployment on Greenhouse Gas Dynamics in Ponds. Environ. Sci. Technol. 2024;58:22104. doi: 10.1021/acs.est.4c06363. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Cromratie Clemons S. K., Salloum C. R., Herdegen K. G., Kamens R. M., Gheewala S. H.. Life cycle assessment of a floating photovoltaic system and feasibility for application in Thailand. Renewable Energy. 2021;168:448–462. doi: 10.1016/j.renene.2020.12.082. [ DOI ] [ Google Scholar ] Hayibo K. S., Mayville P., Pearce J. M.. The greenest solar power? Life cycle assessment of foam-based flexible floatovoltaics. Sustainable Energy & Fuels. 2022;6(5):1398–1413. doi: 10.1039/D1SE01823J. [ DOI ] [ Google Scholar ] Budi, I. D. ; Widodo, S. ; Sitohang, N. S. ; Sulistyo, H. D. ; Kaharudin, D. . Comparative Life Cycle Assessment of Floating and Ground-Mounted Utility PV System in Cirata, Indonesia. In 2024 International Conference on Technology and Policy in Energy and Electric Power (ICTPEP), 2024. [ Google Scholar ] Frehner A., Stucki M., Itten R.. Are alpine floatovoltaics the way forward? Life-cycle environmental impacts and energy payback time of the worlds’ first high-altitude floating solar power plant. Sustainable Energy Technologies and Assessments. 2024;68:103880–103880. doi: 10.1016/j.seta.2024.103880. [ DOI ] [ Google Scholar ] Kester, J. ; Liu, J. ; Binani, A. . Carbon Footprint of Floating PV Systems; International Energy Agency (IEA), 2024. 10.69766/JGAZ9626. [ DOI ] [ Google Scholar ] Parascanu M. M., Keiner D., Breyer C., Macé P., Lizasoain-Arteaga E.. Comprehensive prospective environmental assessment of innovative photovoltaic technologies: Integration into electricity grids in Finland, Germany, and Spain. Energy. 2025;315:134454. doi: 10.1016/j.energy.2025.134454. [ DOI ] [ Google Scholar ] Gadzanku S., Mirletz H., Lee N., Daw J., Warren A.. Benefits and Critical Knowledge Gaps in Determining the Role of Floating Photovoltaics in the Energy-Water-Food Nexus. Sustainability. 2021;13(8):4317. doi: 10.3390/su13084317. [ DOI ] [ Google Scholar ] Frischknecht, R. ; Stolz, P. ; Heath, G. ; Raugei, M. ; Sinha, P. ; de Wild-Scholten, M. . Methodology Guidelines on Life Cycle Assessment of Photovoltaic Electricity; International Energy Agency (IEA) Photovoltaic Power Systems Programme, 2020. [ Google Scholar ] Poralu Marine . Plateforme solaire flottante sur le Lac des Toules. Mayville P., Patil N. V., Pearce J. M.. Distributed manufacturing of after market flexible floating photovoltaic modules. Sustainable Energy Technologies and Assessments. 2020;42:100830. doi: 10.1016/j.seta.2020.100830. [ DOI ] [ Google Scholar ] van der Hulst M. K., Magoss D., Massop Y., Veenstra S., van Loon N., Dogan I., Coletti G., Theelen M., Hoeks S., Huijbregts M. A. J., van Zelm R., Hauck M.. Comparing Environmental Impacts of Single-Junction Silicon and Silicon/Perovskite Tandem Photovoltaics-A Prospective Life Cycle Assessment. ACS Sustain Chem. Eng. 2024;12(23):8860–8870. doi: 10.1021/acssuschemeng.4c01952. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] ecoinvent . ecoinvent database (version 3.9.1) [Allocation cut-off by classification]. 3.9.1 ed.; Zurich, Switzerland, 2022. [ Google Scholar ] Wernet G., Bauer C., Steubing B., Reinhard J., Moreno-Ruiz E., Weidema B.. The ecoinvent database version 3 (part I): overview and methodology. International Journal of Life Cycle Assessment. 2016;21(9):1218–1230. doi: 10.1007/s11367-016-1087-8. [ DOI ] [ Google Scholar ] Philipps, S. ; Warmuth, W. . Photovoltaics Report; Fraunhofer Institute for Solar Energy Systems ISE, 2025. https://www.ise.fraunhofer.de/en/publications/studies/photovoltaics-report.html . [ Google Scholar ] Wirth, H.

Recent Facts about Photovoltaics in Germany; Fraunhofer Institute for Solar Energy Systems ISE, Freiburg, Germany, 2025. https://www.ise.fraunhofer.de/en/publications/studies/recent-facts-about-pv-in-germany.html . [ Google Scholar ] Rosentreter J. A., Borges A. V., Deemer B. R., Holgerson M. A., Liu S., Song C., Melack J., Raymond P. A., Duarte C. M., Allen G. H., Olefeldt D., Poulter B., Battin T. I., Eyre B. D.. Half of global methane emissions come from highly variable aquatic ecosystem sources. Nature Geoscience. 2021;14(4):225–230. doi: 10.1038/s41561-021-00715-2. [ DOI ] [ Google Scholar ] Vroom R. J. E., Kosten S., Almeida R. M., Mendonça R., Muzitano I. S., Barbosa I., Nasário J., Oliveira Junior E. S., Flecker A. S., Barros N.. Widespread dominance of methane ebullition over diffusion in freshwater aquaculture ponds. Frontiers in Water. 2023;5:1256799. doi: 10.3389/frwa.2023.1256799. [ DOI ] [ Google Scholar ] Valiantzas J. D.. Simplified versions for the Penman evaporation equation using routine weather data. Journal of Hydrology. 2006;331(3–4):690–702. doi: 10.1016/j.jhydrol.2006.06.012. [ DOI ] [ Google Scholar ] NREL . National Solar Radiation Database (NSRDB) Data Viewer. National Renewable Energy Laboratory (NREL), 2025. https://nsrdb.nrel.gov/data-viewer (accessed 2025 July 28 2025). [ Google Scholar ] Sengupta M., Xie Y., Lopez A., Habte A., Maclaurin G., Shelby J.. The National Solar Radiation Data Base (NSRDB) Renewable and Sustainable Energy Reviews. 2018;89:51–60. doi: 10.1016/j.rser.2018.03.003. [ DOI ] [ Google Scholar ] Assouline S., Narkis K., Or D.. Evaporation suppression from water reservoirs: Efficiency considerations of partial covers. Water Resour. Res. 2011;47(7):W07506. doi: 10.1029/2010WR009889. [ DOI ] [ Google Scholar ] Stehfest, E. ; Vuuren, D. ; Kram, T. ; Bouwman, A. ; Alkemade, R. ; Bakkenes, M. ; Biemans, H. ; Bouwman, A. ; Elzen, M. ; Janse, J. ; van Minnen, J. ; Müller, M. ; Prins, A. . Integrated Assessment of Global Environmental Change with IMAGE 3.0. Model description and policy applications; 2014. [ Google Scholar ] Sacchi R., Terlouw T., Siala K., Dirnaichner A., Bauer C., Cox B., Mutel C., Daioglou V., Luderer G.. PRospective EnvironMental Impact asSEment (premise): A streamlined approach to producing databases for prospective life cycle assessment using integrated assessment models. Renewable and Sustainable Energy Reviews. 2022;160:112311. doi: 10.1016/j.rser.2022.112311. [ DOI ] [ Google Scholar ] Hausfather, Z.

Explainer: How ‘Shared Socioeconomic Pathways’ explore future climate change. 2018. Forster, P. ; Storelvmo, T. ; Armour, K. ; Collins, W. ; Dufresne, J. L. ; Frame, D. ; Lunt, D. J. ; Mauritsen, T. ; Palmer, M. D. ; Watanabe, M. ; Wild, M. ; Zhang, H. . The Earth’s Energy Budget, Climate Feedbacks, and Climate Sensitivity. In Climate Change 2021: The Physical Science Basis, Masson-Delmotte, V. , Zhai, P. , Pirani, A. , Connors, S. L. , Péan, C. , Berger, S. , Caud, N. , Chen, Y. , Goldfarb, L. , Gomis, M. I. , Huang, M. , Leitzell, K. , Lonnoy, E. , Matthews, J. B. R. , Maycock, T. K. , Waterfield, T. , Yelekçi, O. , Yu, R. , Zhou, B. , Eds.; Cambridge University Press, 2021; pp 923–1054. [ Google Scholar ] Bastviken D., Johnson M. S.. Future methane emissions from lakes and reservoirs. Nature Water. 2025;3:1397. doi: 10.1038/s44221-025-00532-6. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Huijbregts M. A. J., Steinmann Z. J. N., Elshout P. M. F., Stam G., Verones F., Vieira M., Zijp M., Hollander A., van Zelm R.. ReCiPe2016: a harmonised life cycle impact assessment method at midpoint and endpoint level. International Journal of Life Cycle Assessment. 2017;22(2):138–147. doi: 10.1007/s11367-016-1246-y. [ DOI ] [ Google Scholar ] Mutel C.. Brightway: an open source framework for life cycle assessment. Journal of Open Source Software. 2017;2(12):236. doi: 10.21105/joss.00236. [ DOI ] [ Google Scholar ] Steubing B., de Koning D., Haas A., Mutel C. L.. The Activity Browser  An open source LCA software building on top of the brightway framework. Software Impacts. 2020;3:100012. doi: 10.1016/j.simpa.2019.100012. [ DOI ] [ Google Scholar ] Project Jupyter . JupyterLab. GitHub, 2025. https://github.com/jupyterlab/jupyterlab (accessed 2025 August 20 2025). [ Google Scholar ] WEF . Net-Zero Tracker 2023: Aluminium Industry; World Economic Forum, Geneva, Switzerland, 2023. https://www.weforum.org/publications/net-zero-industry-tracker-2023/in-full/aluminium-industry-net-zero-tracker/ . [ Google Scholar ] Hertwich E. G., Gibon T., Bouman E. A., Arvesen A., Suh S., Heath G. A., Bergesen J. D., Ramirez A., Vega M. I., Shi L.. Integrated life-cycle assessment of electricity-supply scenarios confirms global environmental benefit of low-carbon technologies. Proc. Natl. Acad. Sci. U.S.A. 2015;112(20):6277–6282. doi: 10.1073/pnas.1312753111. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] NREL . Life Cycle Greenhouse Gas Emissions from Electricity Generation: Update Life Cycle Assessment of Energy Systems; Golden, U.S., 2021. https://data.nrel.gov/submissions/171 . [ Google Scholar ] IEA . Electricity 2025; International Energy Agency (IEA), Paris, 2025. https://www.iea.org/reports/electricity-2025 . [ Google Scholar ] van de Ven D. J., Capellan-Perez I., Arto I., Cazcarro I., de Castro C., Patel P., Gonzalez-Eguino M.. The potential land requirements and related land use change emissions of solar energy. Sci. Rep. 2021;11(1):2907. doi: 10.1038/s41598-021-82042-5. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Kakoulaki G., Gonzalez Sanchez R., Gracia Amillo A., Szabo S., De Felice M., Farinosi F., De Felice L., Bisselink B., Seliger R., Kougias I., Jaeger-Waldau A.. Benefits of pairing floating solar photovoltaics with hydropower reservoirs in Europe. Renewable and Sustainable Energy Reviews. 2023;171:112989–112989. doi: 10.1016/j.rser.2022.112989. [ DOI ] [ Google Scholar ] Kakoulaki, G. ; Szábo, S. ; Jäger-Waldau, A. . Powering Up to Cool Down Climate Change: Tripling Renewable Energy to Meet COP28 Pledge in the EU. In 2024 IEEE 52nd Photovoltaic Specialist Conference (PVSC), 2024. [ Google Scholar ] Lima A. G. G., de Mello Junior H. D., Quirino T. M., Mendoza L. A. F., Nunes G. M. S., Muller G. M.. Hydro-floating photovoltaic systems as virtual batteries: A case study of Brazil’s Furnas hydropower plant. Renewable Energy. 2026;256:124036. doi: 10.1016/j.renene.2025.124036. [ DOI ] [ Google Scholar ] Haas J., Khalighi J., de la Fuente A., Gerbersdorf S. U., Nowak W., Chen P. J.. Floating photovoltaic plants: Ecological impacts versus hydropower operation flexibility. Energy Conversion and Management. 2020;206:112414. doi: 10.1016/j.enconman.2019.112414. [ DOI ] [ Google Scholar ] Naushad N. M., Rimon G. R., Khan M. R.. Local temperature and humidity aware prediction of degradation, lifetime, and long-term yield of mono- and bifacial floating photovoltaic systems. Sol. Energy. 2025;299:113639. doi: 10.1016/j.solener.2025.113639. [ DOI ] [ Google Scholar ] European Commission . A new Circular Economy Action Plan: For a cleaner and more competitive Europe; COM(2020) 98 final; European Commission, Brussels, 2020. https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A52020DC0098 . [ Google Scholar ] Golkaram M., Mehta R., Taveau M., Schwarz A., Gankema H., Urbanus J. H., De Simon L., Cakir-Benthem S., van Harmelen T.. Quality model for recycled plastics (QMRP): An indicator for holistic and consistent quality assessment of recycled plastics using product functionality and material properties. Journal of Cleaner Production. 2022;362:132311. doi: 10.1016/j.jclepro.2022.132311. [ DOI ] [ Google Scholar ] Xia S., Yang Y., Poon J. P. H.. How to tackle the looming challenge of solar PV panel recycling. Proc. Natl. Acad. Sci. U. S. A. 2025;122(27):e2417921122. doi: 10.1073/pnas.2417921122. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Aminzadeh M., Lehmann P., Or D.. Evaporation suppression and energy balance of water reservoirs covered with self-assembling floating elements. Hydrol. Earth Syst. Sci. 2018;22(7):4015–4032. doi: 10.5194/hess-22-4015-2018. [ DOI ] [ Google Scholar ] Mady B., Lehmann P., Or D.. Evaporation Suppression From Small Reservoirs Using Floating CoversField Study and Modeling. Water Resour. Res. 2021;57(4):e2020WR028753. doi: 10.1029/2020WR028753. [ DOI ] [ Google Scholar ] Ilgen K., Schindler D., Wieland S., Lange J.. The impact of floating photovoltaic power plants on lake water temperature and stratification. Sci. Rep. 2023;13(1):7932. doi: 10.1038/s41598-023-34751-2. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Abdelgaied M., Kabeel A. E., Zeleňáková M., Abd-Elhamid H. F.. Floating Photovoltaic Plants as an Effective Option to Reduce Water Evaporation in Water-Stressed Regions and Produce Electricity: A Case Study of Lake Nasser, Egypt. Water (Switzerland) 2023;15(4):635. doi: 10.3390/w15040635. [ DOI ] [ Google Scholar ] Abdelal Q.. Floating PV; An assessment of water quality and evaporation reduction in semi-arid regions. International Journal of Low-Carbon Technologies. 2021;16(3):732–739. doi: 10.1093/ijlct/ctab001. [ DOI ] [ Google Scholar ] Padilha Campos Lopes M., de Andrade Neto S., Alves Castelo Branco D., Vasconcelos de Freitas M. A., da Silva Fidelis N.. Water-energy nexus: Floating photovoltaic systems promoting water security and energy generation in the semiarid region of Brazil. Journal of Cleaner Production. 2020;273:122010–122010. doi: 10.1016/j.jclepro.2020.122010. [ DOI ] [ Google Scholar ] da Costa L. C. A., da Silva G. D. P.. Save water and energy: A techno-economic analysis of a floating solar photovoltaic system to power a water integration project in the Brazilian semiarid. International Journal of Energy Research. 2021;45(12):17924–17941. doi: 10.1002/er.6932. [ DOI ] [ Google Scholar ] Seitfudem G., Berger M., Schmied H. M., Boulay A. M.. The updated and improved method for water scarcity impact assessment in LCA, AWARE2.0. Journal of Industrial Ecology. 2025;29(3):891–907. doi: 10.1111/jiec.70023. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Ray N. E., Aredas S., Grodsky S. M., Canino A., Cardoso S. J., Holgerson M. A., Theus M., Schmidt M. L.. Higher methanotroph abundance and bottom-water methane in ponds with floating photovoltaic arrays. bioRxiv. 2026 doi: 10.1101/2025.07.24.666521. [ DOI ] [ Google Scholar ] Aben R. C. H., Velthuis M., Kazanjian G., Frenken T., Peeters E., Van de Waal D. B., Hilt S., de Senerpont Domis L. N., Lamers L. P. M., Kosten S.. Temperature response of aquatic greenhouse gas emissions differs between dominant plant types. Water Res. 2022;226:119251. doi: 10.1016/j.watres.2022.119251. [ DOI ] [ PubMed ] [ Google Scholar ] Davidson T. A., Audet J., Jeppesen E., Landkildehus F., Lauridsen T. L., Søndergaard M., Syväranta J.. Synergy between nutrients and warming enhances methane ebullition from experimental lakes. Nature Climate Change. 2018;8(2):156–160. doi: 10.1038/s41558-017-0063-z. [ DOI ] [ Google Scholar ] Bredeweg E. M., Arismendi I., Murphy C. A., Henkel S. K.. Modeling diverse environmental responses of reservoirs to floating photovoltaic systems. Limnologica. 2025;115:126293. doi: 10.1016/j.limno.2025.126293. [ DOI ] [ Google Scholar ] Montes-Pérez J. J., Marcé R., Obrador B., Conejo-Orosa T., Díez J. L., Escot C., Reyes I., Moreno-Ostos E.. Hydrology influences carbon flux through metabolic pathways in the hypolimnion of a Mediterranean reservoir. Aquatic Sciences. 2022;84(3):36. doi: 10.1007/s00027-022-00867-2. [ DOI ] [ Google Scholar ] Vachon D., Langenegger T., Donis D., McGinnis D. F.. Influence of water column stratification and mixing patterns on the fate of methane produced in deep sediments of a small eutrophic lake. Limnology and Oceanography. 2019;64(5):2114–2128. doi: 10.1002/lno.11172. [ DOI ] [ Google Scholar ] Exley G., Armstrong A., Page T., Jones I. D.. Floating photovoltaics could mitigate climate change impacts on water body temperature and stratification. Sol. Energy. 2021;219:24–33. doi: 10.1016/j.solener.2021.01.076. [ DOI ] [ Google Scholar ] Song J., Choi Y.. Analysis of the potential for use of floating photovoltaic systems on mine pit lakes: Case study at the Ssangyong open-pit limestone mine in Korea. Energies. 2016;9(2):102. doi: 10.3390/en9020102. [ DOI ] [ Google Scholar ] Exley G., Hernandez R. R., Page T., Chipps M., Gambro S., Hersey M., Lake R., Zoannou K. S., Armstrong A.. Scientific and stakeholder evidence-based assessment: Ecosystem response to floating solar photovoltaics and implications for sustainability. Renewable and Sustainable Energy Reviews. 2021;152:111639. doi: 10.1016/j.rser.2021.111639. [ DOI ] [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials es5c17148_si_001.pdf (3MB, pdf) es5c17148_si_002.xlsx (388.7KB, xlsx) Data Availability Statement All data sets generated during this life cycle assessment study and source data for figures are openly available in DANS (Data Station Physical and Technical Science) at 10.17026/PT/LMABM6 . Articles from Environmental Science & Technology are provided here courtesy of American Chemical Society ACTIONS View on publisher site PDF (3.0 MB) Cite Collections Permalink RESOURCES Similar articles Cited by other articles Links to NCBI Databases Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top

Related documents

• Evaluating the sustainability of pv technologies and their cooling systems: a life cycle assessment review#1146423
• Assessing the Spatial Suitability, Economic Viability, and Community Capacity for Solar Photovoltaic Technology in Urban Informal Settlements#625703
• Past Field Experiments of Solar Radiation Modification: A Review of the Scientific and Technical Aspects for Governance#246793
• Photovoltaic arrays reshape multitrophic biodiversity through spatial heterogeneity in Qinghai Tibet Plateau alpine meadows#642446
• Environmental Consequences in Coastal Wetland Following World’s largest Lithium-Ion Battery Storage Facility Fire: Vegetation and Soil Influences#684120