The Netherlands has a land problem. It’s one of the most crowded countries in Europe, a big chunk of it sits below sea level, and pretty much every flat acre that isn’t a town is already a farm, a greenhouse or a cow pasture.<\/p>\n
So when Dutch developers went looking for somewhere to put solar panels, a lot of them ended up on the water. Specifically, on old sand pits. Decades of digging sand out of the ground for construction left the country with deep, flooded holes that nobody farms and hardly anybody swims in. GroenLeven, one of the bigger Dutch solar developers, says on its website that more than half a million of its panels float on pits like these.<\/p>\n
Covering them in solar panels sounds like a pretty tidy solution. Free real estate, no farmland lost, and the neighbors are mostly ducks.<\/p>\n
But what happens to the water underneath?<\/p>\n
Deltares, the independent Dutch institute that does a lot of the country’s water research, spent just over three years trying to answer that at one lake in Friesland, in the north of the Netherlands. And the answer is arguably a bit less rosy than what developers tend to put on their websites.<\/p>\n
The lake is called De Dolten, and it sits just outside the village of Oudehaske, squeezed between the A7 highway, a landfill and a high-voltage substation. Not exactly prime vacation property. It covers 79.4 acres and gets down to 82 feet at its deepest point, according to Deltares.<\/p>\n
Two separate floating solar farms share it. The first, Oudehaske I, has about 17,000 panels and covers roughly 15% of the lake. Oudehaske II went in a couple of years later with about 38,800 panels and takes up around 36%. Together, they cover slightly more than half the water.<\/p>\n
Both arrays are anchored to the lake bed instead of being tied off to the banks, and GroenLeven’s project page<\/a> says they sit at least 66 feet from the shoreline. That gap is on purpose. In a pit this deep, most of the wildlife lives along the edges, so the setback leaves room for birds, bats and otters.<\/p>\n The company also built its floats with open lanes it calls light and air streets, so some sunlight and wind still reach the water between the rows. GroenLeven says the two farms make enough electricity for about 9,000 homes.<\/p>\n Dissolved oxygen is exactly what it sounds like. It’s oxygen gas mixed into the water, and it’s what fish, insects and pretty much everything else living in a lake breathes. If you’re a fish, it’s the only air you get. Scientists measure it in milligrams per liter, which is basically the same thing as parts per million. You’ll usually see it written as mg\/L.<\/p>\n A lake gets most of its oxygen from two places. Wind stirs air into the surface, and algae make it through photosynthesis, the same way the grass in your yard does. Put a roof over the water and you cut into both at once.<\/p>\n That’s the theory, anyway. Deltares and its partner TNO wanted real numbers, so they put two identical probes in De Dolten. One floated in open water, and the other hung under the older Oudehaske I array, both about 3.3 feet below the surface. They logged readings every hour and kept going until early September last year.<\/p>\n Over the whole stretch, the water under the panels averaged 7.65 milligrams of oxygen per liter, while open water averaged 8.73. By my math, that’s about 12% less under the solar farm.<\/p>\n Winter is where it gets a lot more lopsided. The report says oxygen under the panels ran up to 41% lower than in open water during the cold months. That’s the biggest gap Deltares recorded, not the average, so don’t read it as a typical day on the lake.<\/p>\n The algae took a hit too. Deltares used the oxygen data to work out how much the algae were producing, and found production under the panels was about 35% lower than in open water. Its computer model came up with a 43% drop, which the authors treat as a decent match.<\/p>\n Frankly, that algae number probably matters more than the oxygen one. Algae sit at the very bottom of a lake’s food chain, and the report itself says cutting their output could have serious consequences for everything that eats them.<\/p>\n One result I didn’t expect. At that 3.3-foot depth, the water under the panels was actually slightly warmer than open water, by about half a degree Fahrenheit on average. I’d have bet on the shade cooling it down. Deltares’ model does predict cooler surface water once panels go in, so the measurements and the simulations don’t quite agree on this one.<\/p>\n So are the fish down there gasping for air? Honestly, nobody can say from this report, because it doesn’t look at fish at all.<\/p>\n It’s also a side-by-side comparison, panels versus open water over the same hours and days. Deltares’ monitoring started after the first farm was already floating, so this isn’t a before-and-after study of the lake. And there was only one probe under the panels, hanging beneath the older and smaller of the two farms.<\/p>\n The lowest readings don’t settle much, either. The probe under the panels bottomed out at 0.72 milligrams per liter at one point, but the open-water probe dipped as low as 1.07, so low-oxygen spells happened on both sides of the lake.<\/p>\n Then there’s the open-water buoy. It runs on a small solar panel, and during the dark Dutch winter that panel sometimes couldn’t keep up, which left gaps in the buoy’s data in exactly the season where the biggest differences showed up. It’s a pretty ironic problem for a solar study to have.<\/p>\n Deltares does put a number on it. Based on its modeling, the report proposes capping floating solar at 15% of the surface on isolated quarry lakes like De Dolten, so the effects on the lake’s ecology don’t get out of hand.<\/p>\n The authors are careful about it, though. They say 15% is a cautious pick rather than something they’ve studied in depth, and that their simulations can’t tell you exactly where the tipping point sits.<\/p>\n Which is a bit awkward for this particular lake. Oudehaske I covers about 15% of De Dolten on its own, so the second farm pushed the lake well past the line Deltares is now suggesting. I’d guess Dutch planners are going to hear that number quite a bit when the next sand pit comes up for solar.<\/p>\n Other projects have gone a lot lighter. The Bavarian lake we covered in July<\/a> stood its panels upright and covered only 4.65% of the water. And if you want the other side of the argument, GroenLeven’s own Bomhofsplas farm near Zwolle<\/a> hung fish shelters under about 73,000 panels, and researchers counted more than 400 fish moving in.<\/p>\n GroenLeven says on its project page that it plans to measure water quality at De Dolten every year. The Dutch Ministry of Economic Affairs paid for the study, and on April 8, Deltares signed off on the final version of the 39-page report<\/a> it wrote with TNO.<\/p>\n","protected":false},"excerpt":{"rendered":" The Netherlands has a land problem. It’s one of the most crowded countries in Europe, a big chunk of it … <\/p>\nWhy does a lake need oxygen anyway?<\/h2>\n
So what did the sensors find?<\/h2>\n
There’s a lot this study can’t tell you<\/h2>\n
So how much of a lake should you cover?<\/h2>\n