Review on Near-Surface Meteorological Variations Induced by Wind Turbines

Boumendil, Paul; Joulin, Pierre-Antoine; Rodier, Quentin; Masson, Valéry

doi:10.5194/wes-2026-118

Preprints

https://doi.org/10.5194/wes-2026-118

Preprints

20 Jul 2026

| 20 Jul 2026

Status: this discussion paper is a preprint. It has been under review for the journal Wind Energy Science (WES). The manuscript was not accepted for further review after discussion.

Review on Near-Surface Meteorological Variations Induced by Wind Turbines

Paul Boumendil, Pierre-Antoine Joulin, Quentin Rodier, and Valéry Masson

Abstract. This review synthesizes more than two decades of research on near-surface meteorological variations induced by wind turbines and wind farms, combining evidence from field measurement campaigns, satellite observations, mesoscale and microscale simulations. Across these different approaches, the literature shows that wind turbines influence the atmospheric boundary layer through blockage effects, momentum extraction, turbulence generation, and rotor-induced vertical motions. Near-surface meteorological responses are strongly controlled by atmospheric stability. Under stable nighttime conditions, wind turbines generally induce near-surface warming, reduced relative humidity, modifications of turbulence characteristics, and changes in sensible and latent heat fluxes. In contrast, effects during unstable daytime conditions are typically weaker and may even exhibit slight cooling. Through their influence on surface moisture and turbulent exchanges, wind turbines can also enhance evaporation. These microclimatic modifications may subsequently have an impact on ecosystems, with some studies reporting impacts on vegetation activity and crop productivity. However, the magnitude and even the sign of these responses remain highly variable across sites. Wind farm size, layout, topography, atmospheric conditions and surface characteristics further modulate the spatial extent and intensity of the observed effects through cumulative wake interactions and flow redistribution processes. Overall, the meteorological perturbations induced by wind turbines are generally small relative to the natural variability of the atmospheric boundary layer, although locally significant effects can occur under favorable atmospheric conditions. Their persistence, cumulative effects within large wind farms, and continued growth in turbine size and installed capacity warrant further observational and modeling efforts to improve understanding of turbine-atmosphere-surface interactions.

Received: 03 Jul 2026 – Discussion started: 20 Jul 2026

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.

Download & links

Paul Boumendil, Pierre-Antoine Joulin, Quentin Rodier, and Valéry Masson

Status: closed

RC1: 'Comment on wes-2026-118', Anonymous Referee #1, 14 Sep 2026

The comment was uploaded in the form of a supplement: https://wes.copernicus.org/preprints/wes-2026-118/wes-2026-118-RC1-supplement.pdf

Citation: https://doi.org/10.5194/wes-2026-118-RC1
RC2:
'Comment on wes-2026-118', Anonymous Referee #2, 18 Oct 2026
Review of wes-2026-118: “Review on Near-Surface Meteorological Variations Induced by Wind Turbines” by Boumendil et al.

General:
This manuscript addresses an important topic of recurring popular interest, bringing together a useful collection of field observations, satellite analyses, and numerical studies of wind-turbine-induced changes near the surface. A careful synthesis in this area could help explain why reported responses differ among the many sites, stability regimes, and modeling approaches. The distinction between surface temperature and air temperature, and the discussion of observational attribution problems, are useful starting points.
However, the manuscript is not yet a sufficiently reliable or critical review of this literature. It devotes considerable space to describing instruments and numerical methods, but provides less assessment of what actual insight the available evidence provides. Findings from individual turbines, extensive wind farms, land and ocean environments, and very different measurement heights are frequently combined into general conclusions. Several explanations are internally inconsistent, some studies are mischaracterized, and the quantitative summaries lack the information needed to judge whether their results are comparable. The authors state that “Several studies included in this review do not explicitly focus on near-surface variations, and the reported variations are therefore derived from our interpretation of the figures presented by the authors,” but do not provide details about how these inferences are made or from which figures. The strongest claims, particularly those concerning evaporation and vegetation, exceed the evidence presented.
Although the topic is appropriate for Wind Energy Science, I cannot recommend publication of the manuscript in its present form. The authors have assembled a substantial body of literature on a challenging topic, but the breadth of the review makes it difficult to provide the careful, thorough, critical assessment that this subject requires. My concerns are based on the accuracy and traceability of the synthesis, and addressing all of these concerns would require substantial reassessment of the source studies, reconstruction of the quantitative summaries, and revision of the physical interpretation and conclusions. I therefore recommend rejection, while encouraging the authors to narrow the focus to a more manageable review. For example, a review of near-surface air-temperature responses at onshore wind farms could critically compare observational evidence and numerical results, distinguish air temperature from land-surface temperature, and assess the mechanisms underlying differences among studies. Such a narrower scope would offer a manageable path toward a rigorous and useful contribution.
My detailed comments about the present manuscript are below, organized into major issues and minor corrections.
Major:
Scope and organization: What exactly is meant by “near-surface” (title. lines 56, most of sections 204, and lines 703-706)? Please be precise and quantitative. For boundary-layer meteorologists, this label could mean the land or sea skin, the lowest 2–10 m, the surface layer, or all heights below the rotor. These heights are not interchangeable. The long methodological introduction summarizing modeling setups delays the physical synthesis until page 16, and the variable-by-variable treatment separates processes that should be interpreted together. I suggest a shorter methods section followed by a physical framework linking momentum extraction, pressure adjustment, turbulent transport, mean vertical advection, and surface exchange. Then, organize the evidence by response, with explicit distinctions between individual turbines and wind farms and between land and ocean. (A more limited scope would focus on either onshore or offshore, as recommended above.) End with a synthesis identifying well-supported findings, conditional findings, and unresolved questions. Retain detailed model descriptions only where these model descriptions explain differences in reported near-surface responses.

Literature selection and strength of evidence: The manuscript does not describe how studies were identified or selected. Please state the search sources, search terms, date cutoff, inclusion criteria, and treatment of conference papers, preprints, and other sources. (A nice example of this is in Fischereit et al. (2022).) The selection must be transparent. The review should evaluate whether each study controls for wind direction, operational status, terrain, land cover, season, and background variability. In particular, the Henschen et al. (2011) case study is repeatedly used to support relatively large temperature, humidity, and evaporation responses. However, Rajewski et al. (2014) explicitly discuss the lack of undisturbed measurements outside that wind farm and insufficient characterization of field variability that undermine the conclusions of Henschen et al. (2011). Whenever those results are discussed, those limitations need to accompany the reported values; a spatial difference is not automatically a turbine-induced effect.

Quantitative synthesis and comparability: Lines 377–387 and Figs. 8, 9, 14, 16, 18, and 19. The summary tables are central to the review, but they combine means, extrema, selected cases, and different spatial and temporal averages, as well as measurements with modeled diagnostics, “temperature” with LST, and relative humidity with water-vapor content. (Because relative humidity relies on temperature, distinguishing any moistening or drying effect is complicated by the accompanying temperature changes unless results are reported in terms of water vapor content.) These differences should be acknowledged within the tables rather than relying on a general caveat located elsewhere in the manuscript that can be easily overlooked. Include measurement height, downstream distance, reference condition, averaging period, stability condition, and uncertainty or significance. The authors state that “Several studies included in this review do not explicitly focus on near-surface variations, and the reported variations are therefore derived from our interpretation of the figures presented by the authors,” but do not provide details about how these inferences are made or from which figures. Instead, please identify values digitized or inferred from figures and give the original figure or table number and extraction method. Use a consistent difference convention, such as turbine-influenced minus reference. Separate neutral conditions from missing information; crossed cells and blank cells must have unambiguous meanings (i.e. Fig. 18 Rajewski et al. unstable?). (Consider replacing the cells marked with a cross with “not considered”.) Add a table to supplementary information describing which “reported variations are therefore derived from our interpretation of the figures” and which figures from which papers were considered so that synthesis is traceable/reproducable/justified.

Mischaracterization of modeling approaches. Lines 31–35, 190–220, 249–265, and Fig. 1. The historical and methodological account needs substantial and thorough corrections.
Baidya Roy et al. (2004) used elevated energy extraction and a TKE source, and should not be grouped with surface-roughness representations like those of Keith et al. (2004) (lines 31-35). The manuscript itself describes this correctly at lines 205–209, creating an inconsistency with the discussion at lines 190–196. Explain the distinct scales and turbine representations used by Baidya Roy et al. (2004) and Keith et al. (2004). Fischereit et al. (2022) again provides a good example.

Figure 1 incorrectly labels Mirocha et al. (2014) as AL; that paper implemented a GAD in WRF-LES. The AL implementation in WRF by Marjanovic et al. (2017) should appear. Wu and Porté-Agel (2011), cited at lines 263–265 as AL, actually compared nonrotating and rotating actuator-disk models. The authors must carefully review and correct the model classifications throughout the text and tables. Also distinguish the Archer et al. (2020) bug fix from their empirical adjustment of the TKE source: the latter is not a universally established correction factor (Larsén et al. 2024; Quint et al. 2025). Assess older studies by their actual configuration rather than issuing a blanket warning based only on WRF version.

Temperature mechanisms require a consistent budget(lines 484–523). The contrast between “enhanced vertical mixing” and “vertical convergence and divergence of turbulent heat fluxes” is presented as a change in explanation, although turbulent-flux divergence is precisely how turbulent transport changes mean temperature. Similarly, a reduction in local TKE does not by itself establish a reduction in turbulent heat transport: the heat flux depends on a covariance, not TKE alone. Introduce a mean potential-temperature budget and distinguish mean advection, turbulent-flux divergence, and diabatic terms. Explain which terms have actually been quantified in each study (recognizing that not all terms are represented in all studies). The statement (lines 510–513) that coherent wake motions are represented as subgrid TKE in large-scale parameterizations is far too strong. Parameterized TKE does not reproduce the directional structure or scalar transport of rotor-scale coherent motions. Differences in resolution may contribute to differing interpretations, but should not be asserted to resolve them without budget evidence.

Sensible heat flux signs and the surface energy balance (Lines 619–640,719–722, Fig. 18). This discussion is internally inconsistent throughout the manuscript.
With the stated upward-positive convention, enhanced downward heat transport makes a negative flux more negative and increases its magnitude. It cannot explain a reduction in flux magnitude, as stated at lines 637–639.

The stable-condition Quint et al. entry in Fig. 18 reports a more negative flux inside the farm, while lines 634–636 describe a general reduction in magnitude within farms. Recheck the original results and report H, the perturbation ΔH, and changes in |H| separately.

Distinguish the surface flux from a model from a flux measured at 10m or further aloft and from its vertical divergence.

Identify whether simulations prescribe surface temperature, prescribe heat flux, or allow the surface to respond interactively. Those choices constrain which temperature and flux responses can occur and are absolutely critical to comparing studies.

Humidity and evaporation are not interchangeable (lines 525–560,642–659).
The discussion correctly acknowledges that relative humidity can decrease while atmospheric moisture content increases, but subsequently calls these changes “drying signals” and then uses reduced RH to support enhanced evaporation. The discussion must distinguish a thermodynamic reduction in RH caused by warming from an actual decrease in specific humidity or mixing ratio.

State whether percentage changes in RH are percentage points or relative percentages.

Evaporation depends on moisture availability, surface resistance, exchange coefficients, and the surface-to-air vapor gradient: reduced RH alone does not demonstrate an increase in evaporation. The 1.4 and 3.2 mm totals at lines 649–651 do not yield a 52% increase using 1.4 mm as the baseline: the increase would be about 129% (and please see my comments above regarding the limitations of Henschen et al. (2011).

Verify the original denominator, periods, and locations rather than simply changing the percentage.

Clarify if the 1.0–2.0 mm/day of Rajewski et al. (2014) is total evaporation or its perturbation and how that integration relates to the cited 60 W/m² (is this over 24 hours or just over daytime hours?).

Vegetation and agricultural responses: Lines 671–677, 690–701, and 721–725.
Separate measured scalar exchange, remotely sensed vegetation indices, modeled biophysical quantities, and harvested crop yield. Changes in these quantities cannot be equated without additional evidence. (I do realize this is difficult to do with some of these studies, but if the studies are limited, that should be reflected in the review.)

The conclusion that the meteorological alterations “have been shown to inhibit local vegetation growth and productivity” is far too strong, especially given the opposing findings discussed immediately afterward.

Evaluate land clearing, roads and turbine pads, agricultural management, irrigation, terrain, and background climate as alternative or interacting causes.

The explanation at lines 675–677 attributes differences during nonoperation to turbine structures without adequately distinguishing that factor from spatial heterogeneity and other confounding factors.

The CO2 percentages also need an explicit citation, sign convention, baseline, and conditioning; a large relative flux change need not represent an equally large change in plant carbon uptake or yield.

Fundamentally, please acknowledge that reported ecological responses remain site dependent and incompletely attributed unless stronger causal evidence is provided.

Generalization about wake and turbulence responses. Lines 393–450, 566–608, and 709–730. Several conditional findings are incorrectly presented as general rules.
Faster inflow does not necessarily produce a larger normalized velocity deficit because turbine thrust coefficients and control regimes vary with speed; the review must distinguish absolute and relative deficits.

The label of “Venturi effects” should be replaced by a careful explanation of pressure-driven flow redistribution around a permeable rotor, with evidence for the proposed mechanism.

“Wake touchdown” distances are case dependent and for interpretation, the rotor clearance, stability, ambient turbulence, and definition of touchdown should be incorporated.

Similarly, cumulative wakes, wake overlap, and vertical transport of TKE are not interchangeable explanations.

Do not infer a transport mechanism solely from a downstream TKE increase.

Define TKE and TI separately, since TI can increase when mean speed decreases even if velocity variance does not increase.

Coverage and accuracy of references.
The manuscript should situate its contribution relative to existing reviews (Fischereit et al. 2022; Porté-Agel et al. 2020; Adkins and Sescu 2022; Zhou et al. 2022) and explain clearly what additional insights this paper provides.

Lines 51–53 should recognize existing implementation and evaluation in NWP systems, including Fischereit et al. (2024, 2026) instead of implying representation in operational models has not been addressed.

Lines 27–28: Please review the references regarding sound, visual and light pollution, bats/birds and include references that are not more than nine years old.

Ensure that claims are correctly related to their citations: Gao et al. (2023) focuses on plant biomass production and does not comment on the blade-icing power-loss claim at lines 665–667. This is just one example of incorrect citations

Minor:
Lines 31–35: Numerical studies demonstrate simulated responses; “measurable” should not imply that these two modeling papers directly measured atmospheric changes.

Line 33: quantify “large-scale”. Similarly “small-scale” in line 38.

Line 38: unclear why Xie and Archer (2017) is the representative citation here?

Line 42: Calling Armstrong et al. (2014) an opinion article is not itself objectionable, but the review should complete the discussion by mentioning Armstrong et al. (2016) and the concurrent studies (Rajewski et al. 2013, 2014; Smith et al. 2013)). A neutral revision would describe the 2014 paper as identifying uncertainties that motivated subsequent field investigation.

Line 51: be careful not to imply that this manuscript will address wind farm impacts on wildlife, rather wind farm impacts on meteorology that might affect wildlife.

Line 52: at least two EU projects are already implementing wind farms in operational forecast models (Fischereit et al. 2024, 2026).

Figure 1 has several errors and requires very careful review and revision. Correct “Baydia Roy,” “Measurments,” “Comparaison,” “Biomasse,” and “Spectroradiometerv.” Apply a consistent author–year convention. Specify whether timeline positions denote publication dates or campaign dates; if campaign dates are intended, use the actual observation periods. Correct the AL and GAD errors noted above. Perhaps the dates for measurement campaigns should consider the dates of the campaigns, not the dates of the publications describing the measurements. So Baidya Roy could be 1989, Rajewski et al. would become 2010/2011, Smith et al. would become 2012. And CWEX had three resulting publications?

Lines 128–131: Vertical gradients of humidity or mixing ratio do not by themselves define thermal stability. Distinguish these from stability metrics and use potential or virtual potential temperature to define stability where possible.

Lines 154–160: Subtracting a regional mean does not by itself isolate a causal wind-farm signal. Replace “thereby isolating” with language describing the reduction of regional background variability.

Lines 287–288: A spin-up period does not ensure physical consistency, and a typical 24-hour duration is not universal. Explain exactly what is spun up and how adequacy is evaluated. A reference to a mesoscale modeling textbook or similar would be appropriate.

Lines 314–316: Rajewski et al. (2014) is an observational flux study, not an example of an LES setup. Its mention of z0 = 0.05 m describes the Smith et al. measurement environment. Check this attribution.

Figures 7 and 15: State the order of subtraction explicitly and consistently. Retained panel letters such as a, c, e or b, d, f should be explained or relabeled.

9 contains a duplicated Smith et al. (2013) row and doesn’t explain the asterisk for Rosencrans et al. (2025).

Figure 10: Use K, not °K . Does the vertical profile represent actual temperature or potential temperature? Define the lapse-rate sign convention and ensure it is consistent between the reproduced panels and the discussion.

Line 590: The friction-velocity expression is dimensionally incorrect (exponent should be ¼, not ½). Further, some studies use a rotated coordinate system such that the crosswind stress is 0. If that applies here, state that clearly.

Lines 596–597: The text states CWEX turbulence measurement heights as 8–10 m, while Fig. 16 says 6.5 m and Rajewski et al. (2014) reports sonic anemometers at 6.5 m. Check carefully and do not confuse flux measurement heights from temperature measurement heights.

Lines 664–670: Give a source for to blade icing for the 80% claim. Also distinguish freezing sea spray, the subject of Rosencrans et al., from other turbine-icing mechanisms, and again note the difference between simulatiosn and measurements.

Lines 679–684: which citation for the CWEX dew analysis? Clearly define the sampled conditions and comparison underlying the 120-minute difference

Lines 762–766: The comparison with other anthropogenic surface modifications needs quantitative comparisons including matched variables, scales, and averaging periods. Otherwise, remove the comparison. Difficulty of detection does not quantify its magnitude relative to other perturbations.

Throughout: Correct incomplete sentences and agreement errors, including lines 93–95, 283–286, 596–597, and 719–720. Standardize near-surface terminology, units, turbine-model abbreviations, and citation formatting.

References:
Adkins, K. A., and A. Sescu, 2022: Wind Farms and Humidity. Energies, 15, 2603, https://doi.org/10.3390/en15072603.
Archer, C. L., S. Wu, Y. Ma, and P. A. Jiménez, 2020: Two Corrections for Turbulent Kinetic Energy Generated by Wind Farms in the WRF Model. Mon. Weather Rev., 148, 4823–4835, https://doi.org/10.1175/MWR-D-20-0097.1.
Armstrong, A., S. Waldron, J. Whitaker, and N. J. Ostle, 2014: Wind farm and solar park effects on plant–soil carbon cycling: uncertain impacts of changes in ground-level microclimate. Glob. Change Biol., 20, 1699–1706, https://doi.org/10.1111/gcb.12437.
——, R. R. Burton, S. E. Lee, S. Mobbs, N. Ostle, V. Smith, S. Waldron, and J. Whitaker, 2016: Ground-level climate at a peatland wind farm in Scotland is affected by wind turbine operation. Environ. Res. Lett., 11, 044024, https://doi.org/10.1088/1748-9326/11/4/044024.
Baidya Roy, S., S. W. Pacala, and R. L. Walko, 2004: Can large wind farms affect local meteorology? J. Geophys. Res., 109, https://doi.org/10.1029/2004JD004763.
Fischereit, J., R. Brown, X. G. Larsén, J. Badger, and G. Hawkes, 2022: Review of Mesoscale Wind-Farm Parametrizations and Their Applications. Bound.-Layer Meteorol., 182, 175–224, https://doi.org/10.1007/s10546-021-00652-y.
——, H. Vedel, X. G. Larsén, N. E. Theeuwes, G. Giebel, and E. Kaas, 2024: Modelling wind farm effects in HARMONIE–AROME (cycle 43.2.2) – Part 1: Implementation and evaluation. Geosci. Model Dev., 17, 2855–2875, https://doi.org/10.5194/gmd-17-2855-2024.
——, B. T. E. Olsen, M. Imberger, H. Vedel, K. H. Møller, A. N. Hahmann, and X. G. Larsén, 2026: Modelling wind farm effects in HARMONIE-AROME (cycle 43.2.2) – Part 2: wind turbine database and application to Europe. Geosci. Model Dev., 19, 8115–8147, https://doi.org/10.5194/gmd-19-8115-2026.
Gao, L., Q. Wu, J. Qiu, Y. Mei, Y. Yao, L. Meng, and P. Liu, 2023: The impact of wind energy on plant biomass production in China. Sci. Rep., 13, 22366, https://doi.org/10.1038/s41598-023-49650-9.
Henschen, M., and Coauthors, 2011: Do Wind Turbines Affect Weather Conditions: A Case Study in Indiana. J. Purdue Undergrad. Res., 1, 22–29, https://doi.org/10.5703/jpur.01.1.4.
Keith, D. W., J. F. DeCarolis, D. C. Denkenberger, D. H. Lenschow, S. L. Malyshev, S. Pacala, and P. J. Rasch, 2004: The influence of large-scale wind power on global climate. Proc. Natl. Acad. Sci. U. S. A., 101, 16115–16120.
Larsén, X. G., J. Fischereit, S. Hamzeloo, K. Bärfuss, and A. Lampert, 2024: Investigation of wind farm impacts on surface waves using coupled numerical simulations. Renew. Energy, 237, 121671, https://doi.org/10.1016/j.renene.2024.121671.
Marjanovic, N., J. D. Mirocha, B. Kosović, J. K. Lundquist, and F. K. Chow, 2017: Implementation of a generalized actuator line model for wind turbine parameterization in the Weather Research and Forecasting model. J. Renew. Sustain. Energy, 9, 063308, https://doi.org/10.1063/1.4989443.
Mirocha, J. D., B. Kosovic, M. L. Aitken, and J. K. Lundquist, 2014: Implementation of a generalized actuator disk wind turbine model into the weather research and forecasting model for large-eddy simulation applications. J. Renew. Sustain. Energy, 6, 013104, https://doi.org/10.1063/1.4861061.
Porté-Agel, F., M. Bastankhah, and S. Shamsoddin, 2020: Wind-Turbine and Wind-Farm Flows: A Review. Bound.-Layer Meteorol., 174, 1–59, https://doi.org/10.1007/s10546-019-00473-0.
Quint, D., J. K. Lundquist, N. Bodini, and D. Rosencrans, 2025: Simulated meteorological impacts of offshore wind turbines and sensitivity to the amount of added turbulence kinetic energy. Wind Energy Sci., 10, 1269–1301, https://doi.org/10.5194/wes-10-1269-2025.
Rajewski, D. A., and Coauthors, 2013: Crop Wind Energy Experiment (CWEX): Observations of Surface-Layer, Boundary Layer, and Mesoscale Interactions with a Wind Farm. Bull. Am. Meteorol. Soc., 94, 655–672, https://doi.org/10.1175/BAMS-D-11-00240.1.
——, E. S. Takle, J. K. Lundquist, J. H. Prueger, R. L. Pfeiffer, J. L. Hatfield, K. K. Spoth, and R. K. Doorenbos, 2014: Changes in fluxes of heat, H2O, and CO2 caused by a large wind farm. Agric. For. Meteorol., 194, 175–187, https://doi.org/10.1016/j.agrformet.2014.03.023.
Rosencrans, D., J. K. Lundquist, M. Optis, A. Rybchuk, N. Bodini, and M. Rossol, 2024: Seasonal variability of wake impacts on US mid-Atlantic offshore wind plant power production. Wind Energy Sci., 9, 555–583, https://doi.org/10.5194/wes-9-555-2024.
——, ——, ——, and N. Bodini, 2025: The effects of wind farm wakes on freezing sea spray in the mid-Atlantic offshore wind energy areas. Wind Energy Sci., 10, 59–81, https://doi.org/10.5194/wes-10-59-2025.
Smith, C. M., R. J. Barthelmie, and S. C. Pryor, 2013: In situ observations of the influence of a large onshore wind farm on near-surface temperature, turbulence intensity and wind speed profiles. Environ. Res. Lett., 8, 034006, https://doi.org/10.1088/1748-9326/8/3/034006.
Wu, Y.-T., and F. Porté-Agel, 2011: Large-Eddy Simulation of Wind-Turbine Wakes: Evaluation of Turbine Parametrisations. Bound.-Layer Meteorol., 138, 345–366, https://doi.org/10.1007/s10546-010-9569-x.
Xie, S., and C. L. Archer, 2017: A Numerical Study of Wind-Turbine Wakes for Three Atmospheric Stability Conditions. Bound.-Layer Meteorol., 165, 87–112, https://doi.org/10.1007/s10546-017-0259-9.
Zhou, L., S. Baidya Roy, and G. Xia, 2022: 9.12 - Weather, Climatic and Ecological Impacts of Onshore Wind Farms. Comprehensive Renewable Energy (Second Edition), T.M. Letcher, Ed., Elsevier, 165–188, https://doi.org/10.1016/B978-0-12-819727-1.00018-2.
Citation: https://doi.org/10.5194/wes-2026-118-RC2

Status: closed

RC1: 'Comment on wes-2026-118', Anonymous Referee #1, 14 Sep 2026

The comment was uploaded in the form of a supplement: https://wes.copernicus.org/preprints/wes-2026-118/wes-2026-118-RC1-supplement.pdf

Citation: https://doi.org/10.5194/wes-2026-118-RC1
RC2:
'Comment on wes-2026-118', Anonymous Referee #2, 18 Oct 2026
Review of wes-2026-118: “Review on Near-Surface Meteorological Variations Induced by Wind Turbines” by Boumendil et al.

General:
This manuscript addresses an important topic of recurring popular interest, bringing together a useful collection of field observations, satellite analyses, and numerical studies of wind-turbine-induced changes near the surface. A careful synthesis in this area could help explain why reported responses differ among the many sites, stability regimes, and modeling approaches. The distinction between surface temperature and air temperature, and the discussion of observational attribution problems, are useful starting points.
However, the manuscript is not yet a sufficiently reliable or critical review of this literature. It devotes considerable space to describing instruments and numerical methods, but provides less assessment of what actual insight the available evidence provides. Findings from individual turbines, extensive wind farms, land and ocean environments, and very different measurement heights are frequently combined into general conclusions. Several explanations are internally inconsistent, some studies are mischaracterized, and the quantitative summaries lack the information needed to judge whether their results are comparable. The authors state that “Several studies included in this review do not explicitly focus on near-surface variations, and the reported variations are therefore derived from our interpretation of the figures presented by the authors,” but do not provide details about how these inferences are made or from which figures. The strongest claims, particularly those concerning evaporation and vegetation, exceed the evidence presented.
Although the topic is appropriate for Wind Energy Science, I cannot recommend publication of the manuscript in its present form. The authors have assembled a substantial body of literature on a challenging topic, but the breadth of the review makes it difficult to provide the careful, thorough, critical assessment that this subject requires. My concerns are based on the accuracy and traceability of the synthesis, and addressing all of these concerns would require substantial reassessment of the source studies, reconstruction of the quantitative summaries, and revision of the physical interpretation and conclusions. I therefore recommend rejection, while encouraging the authors to narrow the focus to a more manageable review. For example, a review of near-surface air-temperature responses at onshore wind farms could critically compare observational evidence and numerical results, distinguish air temperature from land-surface temperature, and assess the mechanisms underlying differences among studies. Such a narrower scope would offer a manageable path toward a rigorous and useful contribution.
My detailed comments about the present manuscript are below, organized into major issues and minor corrections.
Major:
Scope and organization: What exactly is meant by “near-surface” (title. lines 56, most of sections 204, and lines 703-706)? Please be precise and quantitative. For boundary-layer meteorologists, this label could mean the land or sea skin, the lowest 2–10 m, the surface layer, or all heights below the rotor. These heights are not interchangeable. The long methodological introduction summarizing modeling setups delays the physical synthesis until page 16, and the variable-by-variable treatment separates processes that should be interpreted together. I suggest a shorter methods section followed by a physical framework linking momentum extraction, pressure adjustment, turbulent transport, mean vertical advection, and surface exchange. Then, organize the evidence by response, with explicit distinctions between individual turbines and wind farms and between land and ocean. (A more limited scope would focus on either onshore or offshore, as recommended above.) End with a synthesis identifying well-supported findings, conditional findings, and unresolved questions. Retain detailed model descriptions only where these model descriptions explain differences in reported near-surface responses.

Literature selection and strength of evidence: The manuscript does not describe how studies were identified or selected. Please state the search sources, search terms, date cutoff, inclusion criteria, and treatment of conference papers, preprints, and other sources. (A nice example of this is in Fischereit et al. (2022).) The selection must be transparent. The review should evaluate whether each study controls for wind direction, operational status, terrain, land cover, season, and background variability. In particular, the Henschen et al. (2011) case study is repeatedly used to support relatively large temperature, humidity, and evaporation responses. However, Rajewski et al. (2014) explicitly discuss the lack of undisturbed measurements outside that wind farm and insufficient characterization of field variability that undermine the conclusions of Henschen et al. (2011). Whenever those results are discussed, those limitations need to accompany the reported values; a spatial difference is not automatically a turbine-induced effect.

Quantitative synthesis and comparability: Lines 377–387 and Figs. 8, 9, 14, 16, 18, and 19. The summary tables are central to the review, but they combine means, extrema, selected cases, and different spatial and temporal averages, as well as measurements with modeled diagnostics, “temperature” with LST, and relative humidity with water-vapor content. (Because relative humidity relies on temperature, distinguishing any moistening or drying effect is complicated by the accompanying temperature changes unless results are reported in terms of water vapor content.) These differences should be acknowledged within the tables rather than relying on a general caveat located elsewhere in the manuscript that can be easily overlooked. Include measurement height, downstream distance, reference condition, averaging period, stability condition, and uncertainty or significance. The authors state that “Several studies included in this review do not explicitly focus on near-surface variations, and the reported variations are therefore derived from our interpretation of the figures presented by the authors,” but do not provide details about how these inferences are made or from which figures. Instead, please identify values digitized or inferred from figures and give the original figure or table number and extraction method. Use a consistent difference convention, such as turbine-influenced minus reference. Separate neutral conditions from missing information; crossed cells and blank cells must have unambiguous meanings (i.e. Fig. 18 Rajewski et al. unstable?). (Consider replacing the cells marked with a cross with “not considered”.) Add a table to supplementary information describing which “reported variations are therefore derived from our interpretation of the figures” and which figures from which papers were considered so that synthesis is traceable/reproducable/justified.

Mischaracterization of modeling approaches. Lines 31–35, 190–220, 249–265, and Fig. 1. The historical and methodological account needs substantial and thorough corrections.
Baidya Roy et al. (2004) used elevated energy extraction and a TKE source, and should not be grouped with surface-roughness representations like those of Keith et al. (2004) (lines 31-35). The manuscript itself describes this correctly at lines 205–209, creating an inconsistency with the discussion at lines 190–196. Explain the distinct scales and turbine representations used by Baidya Roy et al. (2004) and Keith et al. (2004). Fischereit et al. (2022) again provides a good example.

Figure 1 incorrectly labels Mirocha et al. (2014) as AL; that paper implemented a GAD in WRF-LES. The AL implementation in WRF by Marjanovic et al. (2017) should appear. Wu and Porté-Agel (2011), cited at lines 263–265 as AL, actually compared nonrotating and rotating actuator-disk models. The authors must carefully review and correct the model classifications throughout the text and tables. Also distinguish the Archer et al. (2020) bug fix from their empirical adjustment of the TKE source: the latter is not a universally established correction factor (Larsén et al. 2024; Quint et al. 2025). Assess older studies by their actual configuration rather than issuing a blanket warning based only on WRF version.

Temperature mechanisms require a consistent budget(lines 484–523). The contrast between “enhanced vertical mixing” and “vertical convergence and divergence of turbulent heat fluxes” is presented as a change in explanation, although turbulent-flux divergence is precisely how turbulent transport changes mean temperature. Similarly, a reduction in local TKE does not by itself establish a reduction in turbulent heat transport: the heat flux depends on a covariance, not TKE alone. Introduce a mean potential-temperature budget and distinguish mean advection, turbulent-flux divergence, and diabatic terms. Explain which terms have actually been quantified in each study (recognizing that not all terms are represented in all studies). The statement (lines 510–513) that coherent wake motions are represented as subgrid TKE in large-scale parameterizations is far too strong. Parameterized TKE does not reproduce the directional structure or scalar transport of rotor-scale coherent motions. Differences in resolution may contribute to differing interpretations, but should not be asserted to resolve them without budget evidence.

Sensible heat flux signs and the surface energy balance (Lines 619–640,719–722, Fig. 18). This discussion is internally inconsistent throughout the manuscript.
With the stated upward-positive convention, enhanced downward heat transport makes a negative flux more negative and increases its magnitude. It cannot explain a reduction in flux magnitude, as stated at lines 637–639.

The stable-condition Quint et al. entry in Fig. 18 reports a more negative flux inside the farm, while lines 634–636 describe a general reduction in magnitude within farms. Recheck the original results and report H, the perturbation ΔH, and changes in |H| separately.

Distinguish the surface flux from a model from a flux measured at 10m or further aloft and from its vertical divergence.

Identify whether simulations prescribe surface temperature, prescribe heat flux, or allow the surface to respond interactively. Those choices constrain which temperature and flux responses can occur and are absolutely critical to comparing studies.

Humidity and evaporation are not interchangeable (lines 525–560,642–659).
The discussion correctly acknowledges that relative humidity can decrease while atmospheric moisture content increases, but subsequently calls these changes “drying signals” and then uses reduced RH to support enhanced evaporation. The discussion must distinguish a thermodynamic reduction in RH caused by warming from an actual decrease in specific humidity or mixing ratio.

State whether percentage changes in RH are percentage points or relative percentages.

Evaporation depends on moisture availability, surface resistance, exchange coefficients, and the surface-to-air vapor gradient: reduced RH alone does not demonstrate an increase in evaporation. The 1.4 and 3.2 mm totals at lines 649–651 do not yield a 52% increase using 1.4 mm as the baseline: the increase would be about 129% (and please see my comments above regarding the limitations of Henschen et al. (2011).

Verify the original denominator, periods, and locations rather than simply changing the percentage.

Clarify if the 1.0–2.0 mm/day of Rajewski et al. (2014) is total evaporation or its perturbation and how that integration relates to the cited 60 W/m² (is this over 24 hours or just over daytime hours?).

Vegetation and agricultural responses: Lines 671–677, 690–701, and 721–725.
Separate measured scalar exchange, remotely sensed vegetation indices, modeled biophysical quantities, and harvested crop yield. Changes in these quantities cannot be equated without additional evidence. (I do realize this is difficult to do with some of these studies, but if the studies are limited, that should be reflected in the review.)

The conclusion that the meteorological alterations “have been shown to inhibit local vegetation growth and productivity” is far too strong, especially given the opposing findings discussed immediately afterward.

Evaluate land clearing, roads and turbine pads, agricultural management, irrigation, terrain, and background climate as alternative or interacting causes.

The explanation at lines 675–677 attributes differences during nonoperation to turbine structures without adequately distinguishing that factor from spatial heterogeneity and other confounding factors.

The CO2 percentages also need an explicit citation, sign convention, baseline, and conditioning; a large relative flux change need not represent an equally large change in plant carbon uptake or yield.

Fundamentally, please acknowledge that reported ecological responses remain site dependent and incompletely attributed unless stronger causal evidence is provided.

Generalization about wake and turbulence responses. Lines 393–450, 566–608, and 709–730. Several conditional findings are incorrectly presented as general rules.
Faster inflow does not necessarily produce a larger normalized velocity deficit because turbine thrust coefficients and control regimes vary with speed; the review must distinguish absolute and relative deficits.

The label of “Venturi effects” should be replaced by a careful explanation of pressure-driven flow redistribution around a permeable rotor, with evidence for the proposed mechanism.

“Wake touchdown” distances are case dependent and for interpretation, the rotor clearance, stability, ambient turbulence, and definition of touchdown should be incorporated.

Similarly, cumulative wakes, wake overlap, and vertical transport of TKE are not interchangeable explanations.

Do not infer a transport mechanism solely from a downstream TKE increase.

Define TKE and TI separately, since TI can increase when mean speed decreases even if velocity variance does not increase.

Coverage and accuracy of references.
The manuscript should situate its contribution relative to existing reviews (Fischereit et al. 2022; Porté-Agel et al. 2020; Adkins and Sescu 2022; Zhou et al. 2022) and explain clearly what additional insights this paper provides.

Lines 51–53 should recognize existing implementation and evaluation in NWP systems, including Fischereit et al. (2024, 2026) instead of implying representation in operational models has not been addressed.

Lines 27–28: Please review the references regarding sound, visual and light pollution, bats/birds and include references that are not more than nine years old.

Ensure that claims are correctly related to their citations: Gao et al. (2023) focuses on plant biomass production and does not comment on the blade-icing power-loss claim at lines 665–667. This is just one example of incorrect citations

Minor:
Lines 31–35: Numerical studies demonstrate simulated responses; “measurable” should not imply that these two modeling papers directly measured atmospheric changes.

Line 33: quantify “large-scale”. Similarly “small-scale” in line 38.

Line 38: unclear why Xie and Archer (2017) is the representative citation here?

Line 42: Calling Armstrong et al. (2014) an opinion article is not itself objectionable, but the review should complete the discussion by mentioning Armstrong et al. (2016) and the concurrent studies (Rajewski et al. 2013, 2014; Smith et al. 2013)). A neutral revision would describe the 2014 paper as identifying uncertainties that motivated subsequent field investigation.

Line 51: be careful not to imply that this manuscript will address wind farm impacts on wildlife, rather wind farm impacts on meteorology that might affect wildlife.

Line 52: at least two EU projects are already implementing wind farms in operational forecast models (Fischereit et al. 2024, 2026).

Figure 1 has several errors and requires very careful review and revision. Correct “Baydia Roy,” “Measurments,” “Comparaison,” “Biomasse,” and “Spectroradiometerv.” Apply a consistent author–year convention. Specify whether timeline positions denote publication dates or campaign dates; if campaign dates are intended, use the actual observation periods. Correct the AL and GAD errors noted above. Perhaps the dates for measurement campaigns should consider the dates of the campaigns, not the dates of the publications describing the measurements. So Baidya Roy could be 1989, Rajewski et al. would become 2010/2011, Smith et al. would become 2012. And CWEX had three resulting publications?

Lines 128–131: Vertical gradients of humidity or mixing ratio do not by themselves define thermal stability. Distinguish these from stability metrics and use potential or virtual potential temperature to define stability where possible.

Lines 154–160: Subtracting a regional mean does not by itself isolate a causal wind-farm signal. Replace “thereby isolating” with language describing the reduction of regional background variability.

Lines 287–288: A spin-up period does not ensure physical consistency, and a typical 24-hour duration is not universal. Explain exactly what is spun up and how adequacy is evaluated. A reference to a mesoscale modeling textbook or similar would be appropriate.

Lines 314–316: Rajewski et al. (2014) is an observational flux study, not an example of an LES setup. Its mention of z0 = 0.05 m describes the Smith et al. measurement environment. Check this attribution.

Figures 7 and 15: State the order of subtraction explicitly and consistently. Retained panel letters such as a, c, e or b, d, f should be explained or relabeled.

9 contains a duplicated Smith et al. (2013) row and doesn’t explain the asterisk for Rosencrans et al. (2025).

Figure 10: Use K, not °K . Does the vertical profile represent actual temperature or potential temperature? Define the lapse-rate sign convention and ensure it is consistent between the reproduced panels and the discussion.

Line 590: The friction-velocity expression is dimensionally incorrect (exponent should be ¼, not ½). Further, some studies use a rotated coordinate system such that the crosswind stress is 0. If that applies here, state that clearly.

Lines 596–597: The text states CWEX turbulence measurement heights as 8–10 m, while Fig. 16 says 6.5 m and Rajewski et al. (2014) reports sonic anemometers at 6.5 m. Check carefully and do not confuse flux measurement heights from temperature measurement heights.

Lines 664–670: Give a source for to blade icing for the 80% claim. Also distinguish freezing sea spray, the subject of Rosencrans et al., from other turbine-icing mechanisms, and again note the difference between simulatiosn and measurements.

Lines 679–684: which citation for the CWEX dew analysis? Clearly define the sampled conditions and comparison underlying the 120-minute difference

Lines 762–766: The comparison with other anthropogenic surface modifications needs quantitative comparisons including matched variables, scales, and averaging periods. Otherwise, remove the comparison. Difficulty of detection does not quantify its magnitude relative to other perturbations.

Throughout: Correct incomplete sentences and agreement errors, including lines 93–95, 283–286, 596–597, and 719–720. Standardize near-surface terminology, units, turbine-model abbreviations, and citation formatting.

References:
Adkins, K. A., and A. Sescu, 2022: Wind Farms and Humidity. Energies, 15, 2603, https://doi.org/10.3390/en15072603.
Archer, C. L., S. Wu, Y. Ma, and P. A. Jiménez, 2020: Two Corrections for Turbulent Kinetic Energy Generated by Wind Farms in the WRF Model. Mon. Weather Rev., 148, 4823–4835, https://doi.org/10.1175/MWR-D-20-0097.1.
Armstrong, A., S. Waldron, J. Whitaker, and N. J. Ostle, 2014: Wind farm and solar park effects on plant–soil carbon cycling: uncertain impacts of changes in ground-level microclimate. Glob. Change Biol., 20, 1699–1706, https://doi.org/10.1111/gcb.12437.
——, R. R. Burton, S. E. Lee, S. Mobbs, N. Ostle, V. Smith, S. Waldron, and J. Whitaker, 2016: Ground-level climate at a peatland wind farm in Scotland is affected by wind turbine operation. Environ. Res. Lett., 11, 044024, https://doi.org/10.1088/1748-9326/11/4/044024.
Baidya Roy, S., S. W. Pacala, and R. L. Walko, 2004: Can large wind farms affect local meteorology? J. Geophys. Res., 109, https://doi.org/10.1029/2004JD004763.
Fischereit, J., R. Brown, X. G. Larsén, J. Badger, and G. Hawkes, 2022: Review of Mesoscale Wind-Farm Parametrizations and Their Applications. Bound.-Layer Meteorol., 182, 175–224, https://doi.org/10.1007/s10546-021-00652-y.
——, H. Vedel, X. G. Larsén, N. E. Theeuwes, G. Giebel, and E. Kaas, 2024: Modelling wind farm effects in HARMONIE–AROME (cycle 43.2.2) – Part 1: Implementation and evaluation. Geosci. Model Dev., 17, 2855–2875, https://doi.org/10.5194/gmd-17-2855-2024.
——, B. T. E. Olsen, M. Imberger, H. Vedel, K. H. Møller, A. N. Hahmann, and X. G. Larsén, 2026: Modelling wind farm effects in HARMONIE-AROME (cycle 43.2.2) – Part 2: wind turbine database and application to Europe. Geosci. Model Dev., 19, 8115–8147, https://doi.org/10.5194/gmd-19-8115-2026.
Gao, L., Q. Wu, J. Qiu, Y. Mei, Y. Yao, L. Meng, and P. Liu, 2023: The impact of wind energy on plant biomass production in China. Sci. Rep., 13, 22366, https://doi.org/10.1038/s41598-023-49650-9.
Henschen, M., and Coauthors, 2011: Do Wind Turbines Affect Weather Conditions: A Case Study in Indiana. J. Purdue Undergrad. Res., 1, 22–29, https://doi.org/10.5703/jpur.01.1.4.
Keith, D. W., J. F. DeCarolis, D. C. Denkenberger, D. H. Lenschow, S. L. Malyshev, S. Pacala, and P. J. Rasch, 2004: The influence of large-scale wind power on global climate. Proc. Natl. Acad. Sci. U. S. A., 101, 16115–16120.
Larsén, X. G., J. Fischereit, S. Hamzeloo, K. Bärfuss, and A. Lampert, 2024: Investigation of wind farm impacts on surface waves using coupled numerical simulations. Renew. Energy, 237, 121671, https://doi.org/10.1016/j.renene.2024.121671.
Marjanovic, N., J. D. Mirocha, B. Kosović, J. K. Lundquist, and F. K. Chow, 2017: Implementation of a generalized actuator line model for wind turbine parameterization in the Weather Research and Forecasting model. J. Renew. Sustain. Energy, 9, 063308, https://doi.org/10.1063/1.4989443.
Mirocha, J. D., B. Kosovic, M. L. Aitken, and J. K. Lundquist, 2014: Implementation of a generalized actuator disk wind turbine model into the weather research and forecasting model for large-eddy simulation applications. J. Renew. Sustain. Energy, 6, 013104, https://doi.org/10.1063/1.4861061.
Porté-Agel, F., M. Bastankhah, and S. Shamsoddin, 2020: Wind-Turbine and Wind-Farm Flows: A Review. Bound.-Layer Meteorol., 174, 1–59, https://doi.org/10.1007/s10546-019-00473-0.
Quint, D., J. K. Lundquist, N. Bodini, and D. Rosencrans, 2025: Simulated meteorological impacts of offshore wind turbines and sensitivity to the amount of added turbulence kinetic energy. Wind Energy Sci., 10, 1269–1301, https://doi.org/10.5194/wes-10-1269-2025.
Rajewski, D. A., and Coauthors, 2013: Crop Wind Energy Experiment (CWEX): Observations of Surface-Layer, Boundary Layer, and Mesoscale Interactions with a Wind Farm. Bull. Am. Meteorol. Soc., 94, 655–672, https://doi.org/10.1175/BAMS-D-11-00240.1.
——, E. S. Takle, J. K. Lundquist, J. H. Prueger, R. L. Pfeiffer, J. L. Hatfield, K. K. Spoth, and R. K. Doorenbos, 2014: Changes in fluxes of heat, H2O, and CO2 caused by a large wind farm. Agric. For. Meteorol., 194, 175–187, https://doi.org/10.1016/j.agrformet.2014.03.023.
Rosencrans, D., J. K. Lundquist, M. Optis, A. Rybchuk, N. Bodini, and M. Rossol, 2024: Seasonal variability of wake impacts on US mid-Atlantic offshore wind plant power production. Wind Energy Sci., 9, 555–583, https://doi.org/10.5194/wes-9-555-2024.
——, ——, ——, and N. Bodini, 2025: The effects of wind farm wakes on freezing sea spray in the mid-Atlantic offshore wind energy areas. Wind Energy Sci., 10, 59–81, https://doi.org/10.5194/wes-10-59-2025.
Smith, C. M., R. J. Barthelmie, and S. C. Pryor, 2013: In situ observations of the influence of a large onshore wind farm on near-surface temperature, turbulence intensity and wind speed profiles. Environ. Res. Lett., 8, 034006, https://doi.org/10.1088/1748-9326/8/3/034006.
Wu, Y.-T., and F. Porté-Agel, 2011: Large-Eddy Simulation of Wind-Turbine Wakes: Evaluation of Turbine Parametrisations. Bound.-Layer Meteorol., 138, 345–366, https://doi.org/10.1007/s10546-010-9569-x.
Xie, S., and C. L. Archer, 2017: A Numerical Study of Wind-Turbine Wakes for Three Atmospheric Stability Conditions. Bound.-Layer Meteorol., 165, 87–112, https://doi.org/10.1007/s10546-017-0259-9.
Zhou, L., S. Baidya Roy, and G. Xia, 2022: 9.12 - Weather, Climatic and Ecological Impacts of Onshore Wind Farms. Comprehensive Renewable Energy (Second Edition), T.M. Letcher, Ed., Elsevier, 165–188, https://doi.org/10.1016/B978-0-12-819727-1.00018-2.
Citation: https://doi.org/10.5194/wes-2026-118-RC2

Paul Boumendil, Pierre-Antoine Joulin, Quentin Rodier, and Valéry Masson

Viewed

Total article views: 841 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
459	266	116	841	134	110

HTML: 459
PDF: 266
XML: 116
Total: 841
BibTeX: 134
EndNote: 110

Views and downloads (calculated since 20 Jul 2026)

Month	HTML	PDF	XML	Total
Jul 2026	88	36	11	135
Aug 2026	89	43	5	137
Sep 2026	148	68	74	290
Oct 2026	124	112	25	261
Nov 2026	10	7	1	18

Cumulative views and downloads (calculated since 20 Jul 2026)

Month	HTML	PDF	XML	Total
Jul 2026	88	36	11	135
Aug 2026	89	43	5	137
Sep 2026	148	68	74	290
Oct 2026	124	112	25	261
Nov 2026	10	7	1	18

Viewed (geographical distribution)

Total article views: 796 (including HTML, PDF, and XML) Thereof 796 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 03 Nov 2026

Short summary

Wind turbines can modify atmospheric conditions in and around wind farms. This review brings together results from field measurements, satellite observations, and numerical simulations to examine meteorological changes occurring near the ground, where they may influence ecosystems and human activities. The reviewed studies show that wind turbines can affect wind, temperature, humidity, turbulence and surface exchanges, with effects that depend on weather conditions and site characteristics.


Total:	0
HTML:	0
PDF:	0
XML:	0