Belgium’s next offshore wind giant casts a long “shadow” over its neighbors, and scientists are only beginning to understand its reach
The southern North Sea is filling with wind turbines faster than any other sea basin on Earth — and the farms are running out of room to ignore each other. When a large wind installation rises close to existing ones, it casts a wake: a corridor of slower, more turbulent air stretching dozens of kilometers downwind, quietly draining energy from whatever lies in its path.
That interaction is far more complicated than standard planning tools suggest. A new year-long simulation focused on Belgium’s planned Princess Elisabeth wind farm cluster shows the wake behaves differently depending on wind direction, time of day, season, and atmospheric stability — sometimes in ways that run counter to expectations.
A crowded sea and a growing problem
The North Sea is no longer a place where wind farms can be sited in isolation. After the 2023 Ostend Declaration, Belgium, Germany, Denmark, and the Netherlands committed to delivering 65 gigawatts of offshore wind by 2030. Farms are going up faster than the sea has room for them, and gaps between clusters are shrinking.
A second finding: PE’s wake injects extra turbulence into the downstream cluster, speeding up wake recovery within it and partially offsetting the energy loss.
Every turbine slows the air behind it. When one farm’s wake reaches the next, power output drops — sometimes significantly. Standard engineering models handle simple cases reasonably well but struggle when multiple large clusters interact under shifting conditions.
The Princess Elisabeth (PE) zone sits at the center of this challenge. Belgium plans to build roughly 3.5 GW of capacity there, using 15 MW turbines with a rotor diameter of 236 meters. For prevailing southwesterly winds, PE sits directly upwind of the existing Belgian-Dutch cluster — a near-perfect case study in farm-to-farm interference. Researchers used the Weather Research and Forecasting (WRF) mesoscale model to simulate a full representative year, tracking how PE’s wake would affect 1,409 turbines across multiple neighboring farms.
Simulating a year of wind, wake, and weather
The team selected 2016 because its wind speed and direction distributions closely matched the 1990–2020 long-term average, making it a reliable stand-in for typical conditions. Three model runs were compared: no wind farms, existing farms as of 2022, and the planned PE cluster added. Comparing the second and third runs let researchers attribute impacts specifically to PE.
The model ran at 2 km resolution across the southern North Sea, and wind speed results were validated against five offshore measurement platforms, with biases staying within ±1 m/s.
To keep the accounting clean, the team split losses into two categories: “internal” losses the existing farms already inflict on each other, and “external” losses caused solely by adding PE. That distinction matters enormously.
Up to 5% less energy for the closest neighbor
The Belgian-Dutch cluster, sitting directly downwind of PE for prevailing winds, takes the hardest hit. Annual energy production drops by an average of 5% across the whole cluster; at specific turbine locations, the loss reaches 14%.
Those numbers sit alongside an even starker figure: internal wake losses within the densely packed existing cluster can reach 42% on a spatiotemporally averaged basis. PE adds to an already difficult situation. Farms farther away show losses near zero, with a few registering tiny apparent gains — likely numerical variability rather than real physics.
Stability, season, and the hour of day all matter
Wake losses shift with the atmosphere. Stable conditions, common in spring and summer when warmer air drifts over colder North Sea water, roughly double external wake losses compared with the unstable winter atmosphere. External losses peak during summer afternoons, when wind speeds are lower and southwesterly winds blow more frequently.
Internal and external losses also respond differently to time of day. Internal losses follow seasonal rhythms but show little hourly swing, while external losses vary more strongly through the day — a distinction that matters for grid operators trying to forecast output accurately.
An unexpected bonus: flow speedup and wake recovery
For certain southerly wind directions, PE doesn’t just steal energy — it briefly returns some. The cluster accelerates airflow around its flanks, giving neighboring turbines in lateral positions a small power boost. The study describes this as a lateral speedup effect linked to global blockage, noting it has received comparatively little attention in the literature.
A second finding: PE’s wake injects extra turbulence into the downstream cluster, speeding up wake recovery within it and partially offsetting the energy loss. Both effects point to the same underlying problem — simple additive models can’t capture these interactions, and the net result isn’t what individual loss estimates would predict.
What this means for the next wave of North Sea development
The year-long WRF dataset now provides a benchmark that faster engineering wake models can be tested against — the subject of a companion paper (Part 2), making this simulation a foundation rather than a standalone result.
The reach of effects also deserves attention. Even farms more than 100 km from PE showed statistically detectable changes in annual energy production, suggesting impact assessments must extend well beyond immediate neighbors. As offshore density rises, seasonal and diurnal variability in wake losses will need to feed into yield forecasts and revenue-sharing agreements between operators whose outputs are increasingly linked.
A full validation will only be possible once PE is built and real operational data become available. When that happens, this WRF run will serve as a baseline for refining models and better understanding the North Sea’s increasingly crowded future.
Reference: Porchetta, S., Munters, W., Lejeune, M., Borgers, R., Buckingham, S., and Howland, M. F.: Annual wake impacts in and between wind farm clusters – Part 1: WRF-simulated wake losses for different atmospheric conditions, Wind Energ. Sci., 11, 3273–3294, https://doi.org/10.5194/wes-11-3273-2026, 2026.
Daniel García is an Editor-in-Chief with strong expertise in structural work and engineering principles. He combines this technical foundation with deep knowledge of energy, spatial design, and emerging technologies, bringing a forward-thinking and analytical approach to editorial leadership.
