Observing Wake Impacts in Offshore Europe: Part 2

With publicly available power data, we can get a key look on long-term energy production impacts.

Picking up from our last post on wake impacts at the Walney 1 and 2 wind farms, we now shift focus to Belgium. Prior to 2017, only three wind farms operated off the coast near Bruges—Thornton Bank, Belwind, and Northwind. But between 2017 and 2021, the landscape changed dramatically. A wave of new developments emerged in close proximity, culminating in the full commissioning of the four-phase Borssele cluster in 2021. Although the original wind farms benefit from southwest flow and relatively wake-free upstream conditions, the sheer density of nearby turbines suggests that wake effects should still be detectable. Moreover, the collective presence of so many wind farms effectively forms offshore Europe’s first “supercluster,” introducing additional blockage effects that are unlikely to be negligible.

Our second case study looks at the Thornton Bank, Belwind, and Northwind wind farms in Belgium (circled in purple), which since 2017 have seen extensive development of several wind farms in close proximity. A wind rose taken from Vortex shows predominant winds from the southwest direction.

Continuing our analysis with the ENTSO-E 16.1A power production dataset, we now examine long-term trends in net capacity factor (NCF) for the Belgian offshore wind farms. Unlike the Walney 1/2 case study, this dataset offers full hourly reporting for each wind farm dating back to 2015, providing greater confidence in the reliability of the production data.

As in the previous analysis, we calculate long-term corrected NCF on a rolling 12-month basis by linearly correlating reported monthly production with monthly mean 100-meter wind speeds from the ERA5 reanalysis dataset. We then apply this regression model to ERA5 wind speeds spanning a 25-year period (2000–2024) to assess long-term trends.

As illustrated in the figure below, the linear correlation between each farm’s production and ERA5 wind speeds is consistently strong. Notably, we observe a clear decline in production per unit wind speed from the time of commissioning (shown in purple) through to late 2024 (shown in yellow).

Relationship between monthly capacity factor and monthly mean 100-m wind speeds from the ERA5 reanalysis for each of the Belgian case study wind farms.

In the final figure below, we present the trends in long-term corrected NCF for each of the case study wind farms. While the commissioning of Nobelwind had minimal impact, a more noticeable decline in NCF begins with the commissioning of Rentel and Norther—and becomes especially pronounced following the addition of Seamade and the Borssele cluster.

Setting aside the temporary dip in production across all farms in 2022—which was followed by a recovery—we observe a consistent downward shift in performance. Prior to the neighboring build-out, Northwind operated at an NCF of approximately 42–43%, but has since declined to around 35–37%. Belwind’s NCF dropped from roughly 39–40% to 32–34%, while Thornton Bank fell from about 39% to 32–33%.

Long-term trend in NCF for the three Belgian case study wind farms. Shaded regions mark the commissioning commissioning of neighbouring wind farms which, based on the rolling 12-month calculation of NCF, occur over a 12-month period.

Of course, we cannot definitively attribute the observed decline in NCF to wake and blockage effects alone. These wind farms have been operating for over a decade, and other performance-related factors—such as aging infrastructure, maintenance practices, or operational curtailment—could also contribute. However, the clear alignment between the timing of significant NCF drops and the commissioning of nearby wind farms strongly suggests that wake and blockage effects are the primary drivers of the trend.

This analysis highlights that power losses from wake and blockage effects are not only measurable but clearly identifiable using publicly available datasets like ENTSO-E and ERA5. As we enter a new era of multi-gigawatt offshore “megaclusters,” the ability to assess inter-farm wake impacts will become increasingly critical. Continued access to high-quality, transparent data will be essential for understanding the full magnitude of these losses—and for guiding smarter planning, design, and operational strategies as offshore wind scales up globally.