What is claimed, and what holds up
Floating solar panels, mounted on rafts on lakes and reservoirs, are often described as cutting evaporation by half and adding a 5% jump in efficiency from being cooled by the water underneath. Neither number appears in the 5 peer reviewed studies read for this article. What holds up is smaller, and it depends entirely on how much of the reservoir the panels actually cover.
Evaporation, measured on a real reservoir
At the Passauna reservoir in Brazil, a working floating array was measured directly, not modeled. Over the water the panels actually shade, evaporation dropped 60 to 70%. But that shaded patch is only 1,265.14 square meters of an 8.5 square kilometer reservoir, about 0.01% of its surface, so the saving barely moves the reservoir total.
Coverage decides the reservoir wide result
At Lake Nasser in Egypt, researchers modeled evaporation at several coverage levels across an 11 year weather record. The lake now loses about 12 billion cubic meters of water to evaporation a year, close to 22% of Egypt current Nile water allocation.
Covering 25% of the lake would save 2.1 billion cubic meters a year, 50% would save 4.2 billion, 75% would save 6.3 billion, and the whole lake would save 8.4 billion, a relationship close to a straight line. Covering only the shallow edge, just 15% of the surface, would save 2.66 billion cubic meters a year, more than covering 25% anywhere else, because shallow water evaporates fastest.
The shallow edge option covers only the shallowest 15% of the lake surface, chosen because shallow water evaporates fastest. Every other bar covers that share of the lake in any location.
Show the numbers
| Shallow edge, 15% | 2.66 |
| 25% of the lake | 2.1 |
| 50% of the lake | 4.2 |
| 75% of the lake | 6.3 |
| 100% of the lake | 8.4 |
Turned into a share of the total loss, each percentage point of surface covered removes about 0.7 percentage points of total evaporation, a calculation made for this article from those same figures, not stated by the paper itself.
This share of the total loss is a calculation made for this article, dividing each saved amount above by the current total loss of about 12 billion cubic meters a year. It is not a figure the paper itself states.
Show the numbers
| 25% covered | 17.5 |
| 50% covered | 35 |
| 75% covered | 52.5 |
| 100% covered | 70 |
The case this technology actually fits
In lower income farming regions the case looks different. Researchers modeled a 1.14 megawatt floating array covering 30% of the Vaigai reservoir in Tamil Nadu, India, an irrigation reservoir. It would generate 1.9 gigawatt hours of electricity a year, save 42,731.56 cubic meters of water a year that would otherwise evaporate before reaching crops, and avoid 44,734.62 tons of modeled carbon dioxide a year.
The cooling claim does not hold up either
The other benefit often claimed, that water underneath keeps the panels cooler and more efficient, does not survive as a flat number. Researchers in the Netherlands measured 2 real floating systems for up to 24 months and found the cooling effect was small, in places close to nothing, because reservoir water there often matched or exceeded air temperature. A modeled figure of 11% exists for Lake Nasser, but it describes a hot, dry climate, the paper did not measure it directly, and it only applies in that setting. A separate review of the published literature on floating panels over water confirms the cooling effect is not settled science.
Coverage share, not the panels themselves, decides whether floating solar meaningfully cuts evaporation from a reservoir. A demonstration covering 0.01% of a reservoir, like the one in Brazil, changes almost nothing reservoir wide. A plan covering 25% or more, or aimed at the shallowest 15% of the surface, is what turns the technology into a real water saving tool.