Spatial distribution of local burdens from onshore wind energy deployment in Germany

The expansion of onshore wind energy is central to decarbonizing electricity systems, yet it creates localized burdens, such as visual intrusion and noise, that are spatially heterogeneous and unevenly distributed. These patterns raise concerns of spatial distributional justice. Previous analyses of spatial burden distributions face two main limitations. First, local burdens are often approximated using simple infrastructure-based measures. However, experienced impacts also depend on distance to turbines as well as the density and value of affected populations and assets. These dimensions can be captured more directly through changes in residential property values as an impact-based burden measure. Second, spatial assessments typically rely on a single selected approach to distributional justice, although multiple valid approaches exist.

Here, we assess spatial distributive justice by combining two measures of localized burdens, infrastructure-based and impact-based, with multiple approaches to spatial distributional justice. We hypothesize that the diagnosed degree of spatial distributive justice depends critically on both the burden measure and the justice approach applied.

We analyse wind-energy-related burdens in Germany using two complementary measures. The first is an infrastructure-based measure based on installed wind turbines. The second is an impact-based measure derived from spatially modelled property value losses associated with turbine proximity. The impact-based measure uses a multi-arm causal forest to estimate distance-based, heterogeneous price effects at a 1 km² resolution. Treatment is defined for distances of 0–1 km, 1–2 km, and 2–3 km, with locations beyond 3 km serving as the control group. Estimation relies on an unconfoundedness assumption supported by AIPW diagnostics. Effects are extrapolated using GAM smoothing to obtain continuous spatial coverage for aggregation. Both burden measures are related to five variables, the number of inhabitants, electricity demand, land area, energy potential, and gross domestic product. These variables represent different approaches to spatial distributive justice. Spatial distributive patterns are evaluated using Lorenz curves and Gini coefficients at the federal state (NUTS-1) and district (NUTS-3) levels.

Results show that both the chosen burden measure and the distributive justice approach materially affect inferred spatial disparities. Infrastructure-based measures foreground deployment intensity. Impact-based measures emphasize locations where exposure overlaps with dense and high-value housing markets, resulting in larger absolute economic losses. Rural districts tend to appear more burdened under infrastructure-based measures. Urban districts account for a larger share of impact-based burdens. Turbine counts exhibit only a very weak linear correlation with modelled property value losses, with Pearson r close to zero at the NUTS-3 level. This indicates that infrastructure intensity and monetized local impacts capture distinct dimensions of burden. The resulting distributive patterns vary systematically across justice approaches. Relating burdens to the number of inhabitants, electricity demand, or land area yields broadly North–South contrasts. Relating burdens to gross domestic product or energy potential emphasizes West–East differences.

Overall, the results demonstrate that assessments of spatial distributional justice in wind energy deployment are highly sensitive to both the burden measure and the distributive justice approach applied. We provide a transferable workflow for integrating impact-based burden surfaces into spatial planning metrics. This enables more transparent and robust interpretations of regional burden distributions.