A fish pond and a solar array share a piece of water and then want opposite things from it. The panels make electricity in the middle of the day. The fish are most likely to run short of oxygen before dawn, when photosynthesis has stopped and the water is warm. Any “fishery-solar” project that sizes the array and forgets the 3am aerator has designed a power station and hoped the pond would cope.
Updated October 2026. Yield figures below are from one published trial. They are not a promise that covering a pond raises the harvest.
Two ways to put modules over water
One design lifts the modules on piles or frames above a pond that people still farm. Boats can pass, feed can be thrown, and a planned share of the surface stays open to the air. The other design floats the modules on the water. Floating cover changes how much light and wind the surface gets, and it changes how you harvest. Both can be called a hybrid. They are different construction jobs, different oxygen regimes, and different answers to “can the farmer still work this pond”.
Notice 406, issued by China’s National Energy Administration on 2 September 2014, named fish ponds and lakes among the places a locally consumed distributed project might sit, beside barren hills and agricultural greenhouses. That sentence allowed a location. It did not specify pile height, coverage, or dissolved oxygen. Those are the design.
The load is pumps and aerators, and the hard hour is night
NCAT’s primer on photovoltaics in aquaculture starts from closed systems: pumps and aerators move water and keep oxygen in it. The array is sized from flow and from total dynamic head, which is lift plus the friction in pipes, filters and fittings. NCAT’s practical advice is to use a DC pump when a DC pump can meet that head and flow, because an inverter adds a conversion and another device to fail. An AC pump is the fallback when no DC pump fits.
A system that must run all night, and has no utility, needs a battery and a charge controller. NCAT says a grid connection is the more reliable way to run a 24-hour farm, and that batteries are what you add when the grid is not there. Their worked case is Lashto Fish Farm in Haiti: six 12,000-gallon tanks, at least 90,000 tilapia fingerlings a year, 63 modules up to 14,490 W, and 24 flooded lead-acid batteries totalling 3,232 ampere-hours. That is one farm, in lead-acid equipment, published as an example of scale. It is not a template you copy onto a pond of a different volume.
A review in Energies (2021) makes the same electrical point across a wider set of sites: photovoltaics already show up in aquaculture as the supply for aeration and hatchery loads, including off-grid. The biology is a separate question from whether the panels can run a motor.
Shade changes the water, and one trial is not a rule
Covering the surface cools and calms the pond, and it also cuts the light that phytoplankton use to make oxygen. A paper in Aquaculture Research followed floating photovoltaic ponds at about 40% surface cover in three places in Taiwan. The covered ponds had lower biological oxygen demand and less plankton, and higher oxidation-reduction potential. At two of the sites, pH, temperature and dissolved oxygen were lower than in the open ponds. Harvests in that study were higher: about 1.1 times for giant freshwater prawn, 1.2 times for tilapia and 1.4 times for milkfish, without a reported harm to growth.
Read the sentence in that order. Oxygen and temperature moved. Yield rose in this trial, at this coverage, for these species. Forty percent is the coverage they tested, not a number to stamp on every pond. A denser float, a warmer climate, or a species that needs the phytoplankton can land on the wrong side of the same mechanism. Measure dissolved oxygen before dawn. If it falls, the aerator runtime has to rise, and the battery or the grid has to supply that rise. The panels will be asleep.
What to write down before anyone quotes a “hybrid”
- Is the pond still fished, and do boats need a clear lane? That chooses piles or a partial float, and it caps the cover.
- What is the night load in watts, for how many hours, and what is the worst morning oxygen you will accept?
- Is there a grid? If there is not, the battery is sized from the night load, not from the module nameplate.
- Who records dissolved oxygen, and what happens when it drops?
A brochure that leads with a cooler pond and a second crop of electricity, and never mentions the aerator at 3am, has skipped the hour that kills fish.
Papers and the primer
Pumps, head, DC where it fits, and the Haiti tanks: NCAT, photovoltaic applications in aquaculture. Aeration and hatcheries, including off-grid: Pringle and colleagues, Energies, 2021. The 40% cover trial: Aquaculture Research, doi 10.1111/are.15665. The 2014 Chinese note that named fish ponds as a site type: National Energy Administration, notice 406, 2 September 2014.
