Land-Based Wind

2025 Annual Technology Baseline (ATB) data for land-based wind are shown above. The predicted scenarios correspond to future potential innovations that overcome transportation challenges, advance wind turbine controls, and apply science-based modeling for next-generation wind turbines. These technology advancements enable economies of scale, balance-of-plant efficiencies, and more efficient energy extraction for various turbine configurations in different wind resource regimes.

Turbine characteristics in the Base Year (2023) correspond to a 2023 market average turbine across all wind classes (1–10) and scenarios (Advanced, Moderate, and Conservative). Future projections assume distinct wind turbine technologies will become market standards by 2035 and be deployed within specific wind turbine class(es). Details on the Base Year market average turbine and future turbine characteristics are presented in the Representative Technology section of this page. 

Costs within any given year and for any given project can vary based on site-specific or larger market factors. The chart below shows the 2025 ATB research and development (R&D) capital expenditures (CAPEX) trajectories for land-based wind in the context of historical ranges.

The following chart shows the levelized cost of energy (LCOE) scenario results presented for comparison with normalized literature projections. 

Levelized cost and performance metrics are emphasized in this analysis, reflecting priorities of the global wind industry to remain competitive with other energy technologies.

Representative Technology

The four future wind turbine technologies in the 2025 ATB allow more appropriate technology configurations in each wind resource class (details on wind resource class are provided in the Resource Categorization section of this page). We selected these four wind turbine technologies in consultation with industry to represent the range of technology we expect to be available in 2035, including a higher-specific-power machine that would be more suitable for land-constrained sites. The representative near-future (2035) wind turbine characteristics presented in the 2025 ATB are derived from industry expert interviews and literature expectations. Turbine ratings range from 3.3 to 8.3 megawatts (MW), with rotor diameters of 148–196 meters (m), hub heights of 100–140 m, and specific power ratings of 192–275 watts per square meter (W/m²). Notably, turbines nearly of this scale are commercially available and are expected to be installed at select U.S. sites in the 2020s. 

See the following table for details on each technology configuration (T1–T4) and how they compare to the 2023 market average turbine (Wiser et al., 2024).

Representative Wind Turbine and Plant Characteristics

We assume the characteristics for each technology configuration will remain constant from 2035 to 2060, and we calculate cost and performance trajectories for each configuration using Advanced, Moderate, and Conservative cost and performance assumptions (see the Scenario Descriptions section of this page for details). 

The turbine technology configuration that results in the minimum LCOE for a given range of annual average wind speeds (specified using a resource class; see the Resource Categorization section of this page) is the technology configuration reported in the ATB for that wind resource class. For example, at strong wind resource sites, T1 results in the lowest LCOE of any technology; at a low-resource site, T4 yields lower LCOE values because of the taller hub height and lower specific power rating; see the Resource Categorization section of this page for details). The wind turbine technology assumptions are intended to allow for future turbine configurations that are more appropriate for each wind resource class. 

Scenario Descriptions

The 2025 ATB scenarios for land-based wind assume wind turbine technology characterizations and projected innovations to overcome transportation challenges, advance wind turbine controls, and increase the adoption of science-based modeling. 

Advanced, Moderate, and Conservative Scenarios are described in the table below and denote three distinct cost and performance pathways. We can project cost and performance with varied assumptions about future innovation because multiple scenarios are considered. “Scenario” does not refer to other aspects of this analysis such as variable capital recovery periods, multiple wind classes, or comparison of the market and R&D cases. 🗨

Cost and Performance Assumptions by Scenario

In summary, the three scenarios are defined by combining bottom-up, engineering-based modeling with literature data, learning rates, and market average turbine characteristics. Assumed learning rates are paired with global deployment projections (DNV, 2024) to understand percent reduction of 🗨CAPEX, operating expenditures (OPEX), and LCOE (see the Scenario Assumptions section of this page for more detail).

All three scenarios start with the same cost and performance characteristics in the Base Year, and the projections are generated by linearly interpolating between the Base Year and updated cost and performance assumptions in 2035, 2050, and 2060. 

Base Year Market Average Turbine Cost and Performance Explanation and Justification

The 2023 market average turbine is a theoretical turbine derived from capacity-weighted average characteristics of land-based wind turbines installed in 2023 in the contiguous United States (Wiser et al., 2024). This is a modified approach from the 2024 ATB, which assumed the four future technology configurations remain the same for all years. The 2025 ATB attempts to improve the near-term modeling strategy by grounding the Base Year in a market average turbine for all classes, then transitioning to class-specific future technology.

CAPEX and OPEX estimates for the 2023 🗨market average turbine are derived from the 2024 Cost of Wind Energy Review (COWER), which provides cost estimates for a representative 2023 land-based wind project (Stehly et al., 2024). We generated power curves with market average turbine characteristics to calculate net capacity factors in the Base Year, in all designated wind speed classes. This was done using the National Laboratory of the Rockies (NLR) System Advisor Model (SAM) (MISSING, forthcoming). 

Future Technology Bottom-Up Cost Modeling and Performance Explanation and Justification

In 2035, wind turbine component, transport, and balance-of-system (BOS) costs are estimated using bottom-up cost models. BOS innovations include climbing cranes and reduced turbine spacing because wake steering reduces access road costs and collection costs. In the Moderate case, technology assumptions include the following: 

• Segmented blades longer than 70 m reduce transport costs but increase blade manufacturing and installation costs.
• Advanced manufacturing reduces blade mass and enables larger rotors.
• Spiral-welded towers (both factory and on-site manufactured) decrease tower manufacturing costs, and on-site manufacturing removes transportation limits to tower size—enabling greater hub heights.
• Transport costs are lower because of lighter-weight nacelles, fewer tower sections, and segmented blades.

Technology advancement assumptions in the Moderate Scenario are based on the following conditions(Wiser et al., 2024). First, various engineering firms have developed modular blade systems. For example, the Nabrajoint (see Nabrawind Technologies) uses a bolted connection between blade modules that can be transported individually and assembled on-site, eliminating the logistics barriers for blade lengths greater than 70 m; the first full-scale joint in a real blade segment has already been manufactured and tested to demonstrate strength under operative loads.

Current research such as that by the Big Adaptive Rotor project is investigating various rotor configurations, including two-bladed, downwind, and partial pitch technologies. And current research is investigating transportation options for large blades (e.g., airship blades).

The Advanced Scenario assumes several steel construction and concrete/steel hybrid tower designs from various design and manufacturing firms are available on the market that enable cost-effective towers at 120 m. One example of an Advanced steel construction designed tower is the large-diameter steel tower launched by Vestas in 2014 (Vestas Wind Systems A/S, 2014). In addition, the on-site fabrication of continuous spiral-welded towers has been demonstrated by Keystone Tower Systems, which has also designed optimal high hub-height towers up to 180 m (see Keystone Tower Systems).

Expected turbine performance in the Advanced Scenario is also justifiable. Wind industry and national laboratory R&D programs are focused on enabling Advanced high-fidelity modeling to capture rotor wake dynamics and full resolution of rotating blades, assessment of wake development properties from dynamic wind plant control strategies (e.g., yaw, thrust, and tilt), and evaluation of wind plant controls that elevate high system loads and impact system design. The industry continues to advance rotor size and configuration, atmospheric sciences and forecasting, novel sensing technologies and measurement techniques, computer and computational sciences, multiscale and multidisciplinary computational models, digitalization, big data, and information/data science (Dykes et al., 2019). 

Scenario Assumptions

The cost reduction estimates for the Conservative and Advanced Scenarios are derived from the calculated number of doublings for projected global land-based wind deployment projections (DNV, 2024), and they assume an applied learning rate with respect to the following equation: 

$$ \text{% reduction} = \text{no. of doublings} \times \text{learning rate (LR)} $$

Learning rates are calculated and/or applied for several variables in this analysis: LCOE, CAPEX, and OPEX.

The 2035 Conservative and Advanced learning rates, as well as calculated percentage reductions with respect to 2023, are summarized in the following table as an example. The Moderate Scenario estimates are produced from bottom-up, engineering-based models and literature estimates. 

Scenario Assumptions for CAPEX (2035)

Scenario Assumptions for OPEX (2035)

CAPEX and OPEX learning rates between 2023 and 2035 are grounded by the transition from a market average turbine to a class-specific, future standard technology. Therefore, learning rates follow a new pattern in the 2025 ATB compared to previous data updates. For example, Advanced and Conservative learning rates may appear steep in the transition to T1 (Classes 1–7) because of the difference between market average turbine costs in 2023 and T1 costs in 2035. The Moderate case sets a calculated learning rate based on bottom-up turbine cost models, and the Conservative and Advanced learning rates are adjusted accordingly. More accelerated learning over short periods of time has been justified in literature (Bolinger et al., 2022) and is more likely for newer technology.

Resource Categorization

In the 2025 ATB, the cost and performance data for wind technologies are specified for various resource categories, consistent with those used to represent the full wind resource in the NLR Regional Energy Deployment System (ReEDS) model (Brown et al., 2020). These specified resource classes are no longer used in ReEDS but are reported in the ATB to illustrate resource differences. In ATB editions before 2020, these classes were referred to as techno-resource groups (TRGs) and were designed based on site-specific LCOE by considering, in combination, the wind resource quality (e.g., wind speed) and turbine configuration (e.g., specific power). The TRG methodology is described in Appendix H of the Wind Vision study (DOE, 2015). Starting with the 2020 ATB, the TRG-based classification was replaced with a simpler set of resource wind speed classes defined only by annual mean wind speed. 

For land-based wind, each of the potential wind sites represented in the ReEDS model is associated with 1 of 10 wind speed classes. Annual mean wind speeds, averaged for all years from 2007 through 2013, range from 1.72 to 12.89 meters per second (m/s). To identify the break points that define the 10 wind speed classes within this wind speed range, we specify the percentile of the total wind resource technical potential in capacity terms associated with each class. For example, the top wind speed class (Wind Speed Class 1) is defined based on the mean wind speed range of the top 1% of all potential wind capacity in the contiguous United States. We specify a narrower percentile range for the top classes, so ReEDS has higher-resolution representation for the best sites.

The following table shows the percentile ranges assumed for each resource class as well as the resulting mean wind speed ranges that define each class. We apply these percentiles to a representation of the wind resource using only the most basic exclusions, which are referred to as the open access scenario (Lopez et al., 2021) and based on analysis using the Renewable Energy Potential (reV) model (Maclaurin et al., 2021). Although the ReEDS-based analysis and other analyses can and do rely on different resource representations with various exclusion assumptions, the same mean wind speed break points are used for the 10 wind speed classes shown in the table.

The average wind speed varies by project across the United States. Wind Speed Class 4 is indicative of a Moderate-quality wind regime and is intended to be a representative wind resource for most land-based wind projects installed in the United States. Wind Speed Class 1 is wind-resource-rich and would be the most attractive for wind project development, whereas Wind Speed Class 10 corresponds to the least favorable wind resource sites.

In the 2025 ATB, the future turbine configuration (T1–T4) that yields the lowest LCOE in a given wind speed class will be assigned to that class (see the Representative Technology section of this page). This method results in technology configuration T1 being selected for Wind Speed Classes 1–7, T2 for Wind Speed Class 8, T3 for Wind Speed Class 9, and T4 for Wind Speed Class 10. Assigning specific wind turbine technologies to wind classes is expected to represent a more accurate supply curve.  

Land-Based Wind Resource and Technology Classes

Methodology

This section further details the methodology used to estimate Base Year and future CAPEX, OPEX, and net capacity factor (NCF). The Base Year and future cost and performance estimates assume a 200-MW wind plant, which is consistent with recently installed project sizes (Wiser et al., 2024). For standardized assumptions, see labor cost, regional cost variation, materials cost index, scale of industry, policies and regulations, and inflation.

Capital Expenditures (CAPEX)

Definition: Capital expenditures refer to all up-front costs associated with the development of a land-based wind farm and include the items listed in the Components of CAPEX table below.

Base Year: The Base Year CAPEX is the same for each wind speed class and corresponds to the 2023 market average turbine (see the Scenario Descriptions section of this page). The 2025 ATB includes two sets of cost data with diverging CAPEX, referred to as R&D Costs and Expanded Costs data. The former applies R&D improvements to the market and supply chain conditions of the Base Year; therefore, we model a decrease in CAPEX over time with respect to Base Year estimates. Alternatively, the Expanded Costs case accounts for changing market and supply chain conditions. To model this case, CAPEX increases directly following the Base Year because of supply chain stress, high commodity prices, and increased logistics costs. These near-term factors are accounted for by applying an average CAPEX multiplier of 10% in 2024 and 5% in 2025, relative to 2023 costs.

Future Years: CAPEX associated with the four future turbine technologies is estimated using bottom-up engineering models for hypothetical commercial-scale (e.g., 200-MW) projects installed in 2035. The site-specific design optimization process, which is often reflected in various CAPEX values across wind speed classes, is simplified. In 2035, the CAPEX changes for each scenario (i.e., Conservative, Moderate, and Advanced) and for each turbine. Conservative Scenario estimates are derived assuming a technology-specific (4%–10%) learning rate, and the Advanced Scenario assumes a 10%–27% learning rate.

CAPEX learning rates vary by technology from 2023 to 2035 for the Conservative and Advanced Scenarios and are tied to bottom-up, engineering-based costs in the Moderate Scenario. Steeper learning is expected in some cases compared to the 2024 ATB because of the change in technology over time. It is expected over the long term that wind turbine designs will be optimized for project-specific site conditions. In the ATB, CAPEX reflects typical wind plants and does not include differences in regional costs associated with labor, materials, taxes, or system requirements.

In 2050, CAPEX is back-calculated based on LCOE learning rates of 10%, 14%, and 20% from 2035 to 2050 for the Conservative, Moderate, and Advanced Scenarios, respectively.

Starting in 2024, Grid Connection Costs (i.e., tie line, new or upgraded substations, and interconnection costs) are included in CAPEX. Although the related NREL Standard Scenarios adjusts CAPEX regionally, the ATB methodology differs by presenting CAPEX variation with respect to wind resource class.

Use the following table to view the components of CAPEX and how they change with the scenarios.

Components of CAPEX

Cost Details by Representative Wind Turbine Technology in 2035, Compared to the Market Average Turbine in 2023

Operation and Maintenance (O&M) Costs

Definition: OPEX represents the all-in fixed and variable expenditures required to operate and maintain a wind plant, including items noted in the Components of O&M Costs table below. For land-based wind, the all-in O&M expenditures are reported as fixed operations and maintenance (FOM), assuming no variable O&M costs throughout the project lifetime. 

Base Year: The all-in OPEX costs in 2023 correspond to the market average turbine (Stehly et al., 2024) and are captured in the FOM costs for the ATB. OPEX remains the same between the R&D and Expanded Costs cases. Base Year OPEX is the same for all wind classes.

The historical OPEX values shown in the chart below are based on wind turbine technology characteristics installed in each project commission year (Wiser et al., 2024). Project OPEX is significantly reduced by 2035 in Wind Speed Classes 1–7 because technology assumptions transition to a larger 6-MW turbine.

Future Years: In 2035 and beyond, OPEX is different for each representative technology because this cost is expected to vary by wind turbine rating, with projections showing lower FOM costs as turbine rating increases. Future FOM is assumed to decline by approximately 32% from 2023 to 2035 in the Moderate Scenario, 12.1% in the Conservative Scenario (assuming a 10% learning rate), and 33.8% in the Advanced Scenario (assuming a 28% learning rate). The values provided reflect estimates for the market average turbine and T1 over time in Classes 1–7 and are different for T2–T4 in Classes 8–10 because greater OPEX reductions result from larger machine ratings. The ATB does not consider differences in regional FOM costs associated with labor, materials, or differences in O&M strategies.

Use the following table to view the components of O&M costs.

Components of O&M Costs

Cost Details by Representative Wind Turbine Technology in 2035 and Market Average Turbine in 2023

Capacity Factor

Definition: The capacity factor represents the percentage of equivalent full load hours that a turbine generates electricity at nameplate capacity in a year. Capacity factor is influenced by the wind plant's generation profile, expected downtime, and energy losses within the plant. The specific power (i.e., ratio of machine rating to rotor-swept area) and hub height are design choices that influence the capacity factor. Most installed U.S. wind plants generally align with ATB estimates for performance in Wind Speed Classes 2–7. Projects located in high wind resource sites associated with Wind Speed Class 1 and very low wind resource sites associated with Wind Speed Classes 8–10 are not as common in the historical data, but the range of observed data encompasses ATB estimates.

Base Year: The Base Year capacity factors are calculated by generating a power curve for the 2023 capacity-weighted market average turbine (Wiser et al., 2024) and using the Weibull distribution with average wind speeds in each of the appropriate wind speed classes (see the Resource Categorization section of this page) to produce an annual energy production estimate. The hub height of each representative wind turbine is considered by extrapolating the wind speed at hub height from a reference point 110 m above ground level, assuming a power law shear exponent of 0.14.

The following chart shows a range of capacity factors based on variation in the resource for wind plants in the contiguous United States and the future capacity factor estimates for the Conservative, Moderate, and Advanced Scenarios, which vary by scenario and technology.

Future Years: Net capacity factors are expected to improve over time. In 2035 and beyond, net capacity factors correspond to the four future technology configurations rather than the 2023 market average turbine. Technology innovations are assumed to increase wind plant energy capture through advanced controls and to reduce total system losses that increase capacity factor for all wind speed classes (Dykes et al., 2017). Trade-offs exist between turbine rotor diameter, specific power, and hub height to achieve a given capacity factor, depending on site conditions and costs for pursuing one approach or the other; wind plant layout and operating strategies that impact losses may also be adjusted to yield a given capacity factor. The 2025 ATB presents three of many capacity factor improvement pathways for LCOE reduction. 

Performance Details by Wind Turbine Technology in 2035, Compared to the Market Average Turbine in 2023

References

The following references are specific to this page; for all references in this ATB, see References.