![]() |
| Fig. 1: Floating solar in Spain. (Image Source: M. Popp, through license CC BY-SA 2.0) |
Solar panels do not just make electricity. Most of the sunlight that hits a panel turns into heat in the silicon and glass. On a hot day a dark module behaves more like a small black roof than a sleek high-tech device. Crystalline silicon modules typically lose about 0.4 to 0.5 percent of power for every degree Celsius the cells warm above the test condition of 25°C. [1]
Floating photovoltaic (FPV) systems change the thermal environment.(See Fig. 2.) Instead of sitting above a hot, dry surface, the panels hover a short distance over water that can absorb heat and evaporate. Here we examine two linked questions: why FPV systems run cooler and produce more electricity than land based systems, and what the global numbers look like if we scale that idea up to real reservoirs.
Only a fraction of the incoming solar power, typically 15 to 20 percent, leaves a PV module as electricity. [2] The rest becomes heat, which raises the cell temperature until losses to the environment balance the absorbed power. Over bare soil or concrete, the air just above the panel is already warm and often still. Over water, two extra cooling channels open up:
The water surface is usually a few degrees cooler than nearby land, especially in hot seasons.
Evaporation from the reservoir continuously removes heat from the air layer bathing the back of the modules.
Sukarso and Kim used satellite data over West Java and found that lake surfaces were on average about 8°C cooler than the surrounding ground. [3] If PV modules above the reservoir track that temperature difference, the cells would also be roughly 8°C cooler. Silicon cells lose around 0.4 percent of output per degree, so an 8°C temperature drop should raise power by about 8 × 0.4% ≈ 3 percent.
This simple line of reasoning matches field data fairly well. In a small experiment in Malaysia, Majid and co-workers mounted 80 W panels on a pond and on land and found that the floating panel produced about 5.93 percent more power on average than an identical land based panel. [4] Other studies on larger FPV systems report gains of roughly 0.6 to 4.4 percent depending on climate, wind speed and mounting details. [5]
The same shade that cools the panels also shades the reservoir. Evaporation from open water is driven by sunlight, wind and dryness of the air. Covering even part of the surface with opaque modules cuts the solar term almost entirely beneath them and also slows the wind at the water surface.
Field measurements in hot, dry regions suggest that FPV can cut evaporation by tens of percent. Studies report reductions on the order of 40 to 70 percent under the covered area, depending on local climate and array layout. [6]
Recent work by Jin and co-authors combined global reservoir maps with a detailed PV system model to estimate how much electricity FPV could realistically produce. Assuming that at most 30 percent of each reservoir is covered, and that individual FPV fields do not exceed 30 km2, they find a practical global FPV generation potential of 9,434 ± 29 TWh per year, which is about 3.4 × 1019 joules per year. [7]
At that coverage level the United States could generate about 1,900 TWh per year from FPV, India about 770 TWh, and Brazil about 860 TWh, with many developing countries having FPV potentials several times larger than their current electricity demand. [7]
Floating PV is not simply land PV placed on pontoons. Once systems move from kilowatt pilots to multi-megawatt fields, several physical and economic constraints become dominant:
Structures and loads. Waves, currents, and seasonal water-level changes introduce time-varying forces into floats and moorings. Large arrays must survive storms without breaking apart or drifting into shore or the dam.
Water quality and ecology. Wide FPV coverage alters the light and heat flux at the surface. This can suppress evaporation but may also modify stratification, oxygenation, and algae dynamics in ways that remain only partly understood.
Corrosion and maintenance. A humid, constantly wetted environment is harsher on metals, connectors, and wiring than most land sites. Reservoirs with high mineral content or treated wastewater can accelerate wear.
Cost and access. Floats, walkways, moorings, and boat-based maintenance add material and labor costs that land PV avoids. The modest efficiency gain from cooling must offset this additional complexity, and in many markets it does notlikely the main reason the world is not saturated with FPV installations.
The physical mechanism behind floating solar's performance boost is straightforward: cooler cells make more power, and water is an excellent heat sink. Measurements across climates confirm a real, reproducible efficiency gain. Whether that gain justifies the greater installation and maintenance costs is a site-specific economic question, not a physical one. This remains the central barrier to floating PV's widespread adoption.
[1] T. Hooper, A. Armstrong, and B. Vlaswinkel, "Environmental Impacts and Benefits of Marine Floating Solar," Sol. Energy 219, 11 (2021).
[2] K. R. C. Lakshmi and G. Ramadas, "Dust Deposition's Effect on Solar Photovoltaic Module Performance: An Experimental Study in India's Tropical Region," J. Renew. Mater. 10, 2133 (2022).
[3] A. P. Sukarso and K. N. Kim, "Cooling Effect on the Floating Solar PV: Performance and Economic Analysis on the Case of West Java Province in Indonesia," Energies 13, 2126 (2020).
[4] Z. A. A. Majid et al, "Study on Performance of 80 Watt Floating Photovoltaic Panel," J. Mech. Eng. Sci. 7, 1150 (2014).
[5] C. J. Ramanan et al., "Towards Sustainable Power Generation: Recent Advancements in Floating Photovoltaic Technologies," Renew. Sustain. Energy Rev. 194, 114322 (2024).
[6] L. W. Farrar et al., "Floating Solar PV to Reduce Water Evaporation in Water Stressed Regions and Powering Water Pumping: Case Study Jordan," Energy Convers. Manage. 260, 115598 (2022).
[7] Y. Jin et al., "Energy Production and Water Savings From Floating Solar Photovoltaics on Global Reservoirs," Nat. Sustain. 6, 865 (2023).