the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 26 Nov 2024
A parcel-level evaluation of distributed wind opportunity in the contiguous United States
Abstract. This study examines the potential for distributed wind (DW) energy across the contiguous United States, leveraging advancements in the National Renewable Energy Laboratory's distributed wind model, dWind. The novel modeling approach described here utilizes a high-resolution dataset and analyzes over 150 million parcels, a significant improvement from prior methods that extrapolated results from a smaller random sample. This achievement is enabled through key model performance improvements, such as transitioning to parallel processing, which reduces runtime by 97 %. This optimized, high-resolution approach allows the inspection of technology deployment potential and impact on a variety of scales tailored to individual properties and regions. The results here align with prior work showing substantial opportunity for energy generation using DW technologies. Key findings reveal a substantial increase from prior results in estimated technical and economic potential for DW. Metrics tuned to highlight economic potential also show increased incentives supporting rural adoption. Results are spatially aggregated for usability and published via the U.S. Department of Energy Wind Data Portal and a custom scenario visualization platform, aiding policymakers, industry, and property owners in assessing DW viability across various scenarios and spatial scales.
- Preprint
(1317 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on wes-2024-147', Anonymous Referee #1, 17 Dec 2024
The manuscript makes a strong contribution with respect to modeling and analysis capabilities for future distributed wind deployment. The authors articulate the advances from the previous model to the current one and the impressive gains in computational efficiency. The paper is clearly structured and well written. However, the Introduction and Discussion sections are underdeveloped, and more references are needed throughout. The Introduction would benefit from an explanation of why the new approach provides a contribution to the users of this information (i.e., how does someone who is looking at potential deployment for distributed wind benefit from the model using parcel-level data relative to the previous sample approach?) and a clear objective for the paper. The Discussion should address that same idea within the context of interpreting the results, and some narrative that describes the limitations should be added. The authors should also establish the broader value of the work to the international community given the journal’s scope.
Specific Comments:
- The commentary on RAISE is misleading. RAISE did not directly make additional funds available to end users considering DW installation. RAISE is an initiative, not an incentive, so a customer cannot be eligible or ineligible for the RAISE initiative. I think you may be referring to USDA’s REAP eligibility. Please clarify.
- Line 21: Distributed wind can encompass more than turbines up to 60m in height. Please clarify if you are choosing to define DW this way because of your model structure or provide a resource that backs this. References you cite show there is more range than this.
- Line 83: Potential missing word in “account for is and revenue streams”
- Line 95: This sentence is long, and its intention unclear. With the workflow critical to understanding the paper’s contribution, I recommend rephrasing and better aligning the language with Figure 1 to maximize the figure’s usefulness.
- Table 1: In addition to descriptions of the datasets that are used, this section needs a discussion around why these specific sources were selected over others and which ones were used in the original model compared to the new version that this paper covers.
- Line 116: Define CPU
- Line 117: Define I/O
- Line 159: Why is only one turbine sited? What are the limitations of this approach?
- Figure 4: This figure needs a legend defining what the magnitude of the circles represents. I also recommend using a higher resolution figure if it is available.
- Line 190: Which of the cities are respectively urban, suburban, and rural? What makes these cities representative of those environments?
- Figure 7: Does this show the scenarios where the largest turbine is sited or sized to load? Please specify.
- Table 5 and Table 6: These tables are dense and include many N/A values. I recommend restructuring or parsing to make them to make the intention of the tables clear.
- Line 402: You don’t make any direct comparisons between your model outputs (or assumptions) and existing deployment to illustrate that the deployment potential is untapped.
Citation: https://doi.org/10.5194/wes-2024-147-RC1 -
RC2: 'Comment on wes-2024-147', Anonymous Referee #2, 30 Dec 2024
1 Overall assessment and major comments
1.1 Overall
The manuscript examines the potential for distributed wind energy in the United States, using an updated version NRELs dWind model. The improvements with respect to the prior version of the model are incremental. The novelty of the manuscript is therefore somewhat limited, but the results are relevant and have changed compared to the predictions of the prior model. However, there a number of revisions and clarifications that should be made before the manuscript is suitable for publication.
1.2 Insufficient technical detail
Some key elements in the manuscript are not well defined. This makes it difficult to assess the methodology and would also make it very hard for another researcher to reproduce the results.
1.2.1 Definition of distributed wind
A definition of distributed wind appears to be lacking. Although DW turbines are on average certainly smaller than utility-scale wind turbines, it is not clear where the proposed tower height range 10 - 60 m comes from. Especially since front of the meter turbines appear to be included in the (implicit) definition of DW (l. 39), larger tower heights are possible.
1.2.2 Calculation of wind resource and energy yield
How is the wind resource calculated? For instance, is a reference height used, and are the results influenced by vertical extrapolation and wind shear? How was AEP calculated? Which wind turbines were used? Some of this information can supposedly be found in the links in Table 1, but it is not very clear. E.g. reference turbine power curves can be found, according to the manuscript, in a zenodo link in table 1 but this does not seem to show any power curves or at least none that can be easily found. In a paper investigating the energy potential for distributed wind, there should be no ambiguity on how the energy yield is calculated.
1.3 UQ
Results are given without an estimate of their level of uncertainty. Uncertainties should be added, together with an informative discussion of the main contributions to this uncertainty. For instance, one may expect the uncertainty to depend on tower height. Is that the case? If not, why not?
1.4 Blade height
Eq. 1 gives a minimum blade height of zero if the canopy percentage is below a certain threshold. This sounds odd and does not fit standard practice. I am concerned that this may lead to an overestimate of the RSA of the sited turbines in the proposed evaluation. If this is not the case, could the authors explain this in the manuscript?
1.5 Agents and MCA
There is a pipeline for generating agents but little detail is given. Although there is a breakdown between residential and commercial above Table 3, it is not clear what these agents look like. Is some multicriteria analysis performed, as different agents may have a different weighting for different criteria? If the authors do not believe this is relevant, please explain
2 Minor comments
- In the abstract, the statement transitioning to parallel processing, which reduces runtime by 97 % is meaningless without mentioning the number of nodes.
- In the intro, a few references on the positive effect of front of the meter DW on network resilience would not come amiss. Some other statements also deserve to be backed up by a reference.
- "agents", l. 73: use smart quotes (as is done elsewhere in the manuscript)
- Figs. 1 and 2 have low resolution. This may not look good if printed or displayed on large screen. Please check, also for other figures.
Citation: https://doi.org/10.5194/wes-2024-147-RC2 -
RC3: 'Comment on wes-2024-147', Anonymous Referee #3, 02 Jan 2025
The paper examines an approach to evaluating the energy potential of DW in the US.
The approach is an extension of a previous work within the effort to develop the use of distributed wind energy generation in two different configurations: behind-the-meter and front-of-the-meter.
The manuscript is clear and quite well written, but there are some issues that need to be addressed for it to be considered suitable for publication:
- It is very important to include a list of symbols and abbreviations as well as a precise list of definitions for some important terms and parameters (first of all the term "parcel" and the blade height): this can greatly improve the scientific contribution and the universal value of the study.
- The "blade height setback factor" needs to be described and discussed.
- A title in each sub-figure can be useful to improve the readability of the plots.
- The technical and economic differences between BTM and FOM are not discussed in depth; a discussion on this point can be valuable.
- On the one hand, it is quite clear how useful the application of the approach can be to the specific case of the USA, but on the other hand, the universal value of the approach cannot be appreciated; there is a need to improve the discussion and conclusions in this direction.
Moreover, the results can be summarised in a more efficient way.
Anyway, the main problem of the paper is that in its current state it presents interesting results for the specific application, but the scientific contribution needs to be demonstrated with a good discussion on the universality of the approach.
Citation: https://doi.org/10.5194/wes-2024-147-RC3 -
AC1: 'Comment on wes-2024-147', Jane Lockshin, 04 Feb 2025
Thank you to all the reviewers for the helpful comments and insights!
Here are the changes to document in response to the reviewer's comments:
- In response to Reviewer #1’s comments:
- Added to last sentence in Introduction: “The improved spatial accuracy of this approach ensures better estimates of technical and economic distributed wind potential by analyzing granular information and capturing local variations rather than broad approximations.”
- Added limitations in Discussion and conclusion section
- Table 1/Line 365/Line 367: “RAISE ineligibility” information to “REAP ineligibility”
- Line 21: Changed range of distributed wind definition from 10 to 60m to 10 to 80m as this reflects our modeling
- Line 83: Updated sentence as follows: “accounting for incentives and revenue streams”
- Line 95: changed “As shown in this figure, calculation of economic generation potential at a given site involves three main phases corresponding to agent generation and technical potential calculation, inclusion of the wind resource to calculate estimated annual energy production, and the consideration of local energy demand and system costs to determine the techno-economic subset” to “As shown in this figure, separate model components calculate different metrics throughout the model pipeline: the agent generation process sites and sizes systems to determine parcel-level technical potential, the reV model calculates annual energy production according to wind resource, and the dWind model assesses parcel-level economic potential according to local energy demand and system costs” for clarity.
- Table 1: The description of the datasets that were updated from the prior study are included in the section above the table.
- Line 116: CPU defined as “Central Processing Unit”
- Line 117: I/O defined as “Input/Output”
- Figure 4: Clarification that the magnitude of the circles represents the available land area for siting on parcels.
- Figure 7: Updated caption to include: “The maximum BTM turbine has been used to show maximum capacity, rather than the sized down turbine.”
- Line 402: Removed “these gains underscore the vast untapped DW potential”
- In response to Reviewer #2’s comments:
- Definition of Distributed Wind: Changed range of distributed wind definition from 10 to 60m to 10 to 80m as this reflects our modeling
- Calculation of wind resource and energy yield: Included a reference to the turbine power curves used in Table 1 and included an “Energy production” entry in Table 1 that describes that the reV model calculates energy production using the wind resource data (WIND Toolkit) and turbine power curves.
- Blade height: changed language from “minimum/maximum blade height” to “allowable minimum/maximum blade height” for clarity.
- Changed “parallel processing” to “multiprocessing” as both the prior study and this one both used parallelization, but the updated model also incorporates multiprocessing.
- Added references on the value of DERs and grid reliability and resilience in the introduction section.
- Added uncertainties as limitation in the Discussion section
- In response to Reviewer #3’s comments:
- Added the geographic constraints of the modeling as a limitation in the Discussion section.
- Other changes:
- Updated reference to Lightbox parcels dataset in Table 1
- Removed colored cells in tables
- Added the following sections: “Competing Interests,” “Disclaimer,” “Acknowledgements,” “Financial Support”
Responses Reviewer #1:
- The commentary on RAISE is misleading. RAISE did not directly make additional funds available to end users considering DW installation. RAISE is an initiative, not an incentive, so a customer cannot be eligible or ineligible for the RAISE initiative. I think you may be referring to USDA’s REAP eligibility. Please clarify.
Yes, while discussing incentives, we are referring to REAP eligibility. - Line 21: Distributed wind can encompass more than turbines up to 60m in height. Please clarify if you are choosing to define DW this way because of your model structure or provide a resource that backs this. References you cite show there is more range than this.
Yes, we also include turbines up to 80m in our modeling. - Line 83: Potential missing word in “account for is and revenue streams”
Sentence should be “accounting for incentives and revenue streams.” - Line 95: This sentence is long, and its intention unclear. With the workflow critical to understanding the paper’s contribution, I recommend rephrasing and better aligning the language with Figure 1 to maximize the figure’s usefulness.
Sentence rephrased for clarity. - Table 1: In addition to descriptions of the datasets that are used, this section needs a discussion around why these specific sources were selected over others and which ones were used in the original model compared to the new version that this paper covers.
All data sources and vintages have remained consistent with the prior study except for U.S. Census boundary definitions (as new data from 2010 to 2020 have become available) and utility boundaries (as updated 2020 definitions have become available). - Figure 4: This figure needs a legend defining what the magnitude of the circles represents. I also recommend using a higher resolution figure if it is available.
The magnitude of the circles represents the available land area for siting on parcels. - Line 190: Which of the cities are respectively urban, suburban, and rural? What makes these cities representative of those environments?
All of these cities include some urban, suburban, and rural parcels. - Figure 7: Does this show the scenarios where the largest turbine is sited or sized to load? Please specify.
The maximum BTM turbine has been used to show maximum capacity, rather than the sized down turbine.
Responses Reviewer #2:
1.2.1 Definition of distributed wind
A definition of distributed wind appears to be lacking. Although DW turbines are on average certainly smaller than utility-scale wind turbines, it is not clear where the proposed tower height range 10 - 60 m comes from. Especially since front of the meter turbines appear to be included in the (implicit) definition of DW (l. 39), larger tower heights are possible.
Our modeling includes heights of 80m as well, so we have updated this definition.1.2.2 Calculation of wind resource and energy yield
How is the wind resource calculated? For instance, is a reference height used, and are the results influenced by vertical extrapolation and wind shear? How was AEP calculated? Which wind turbines were used? Some of this information can supposedly be found in the links in Table 1, but it is not very clear. E.g. reference turbine power curves can be found, according to the manuscript, in a zenodo link in table 1 but this does not seem to show any power curves or at least none that can be easily found. In a paper investigating the energy potential for distributed wind, there should be no ambiguity on how the energy yield is calculated.
The wind resource is derived from the WIND Toolkit (described in Table 1) and includes resource information at various heights. Energy production was calculated using the reV model as reV uses the wind resource data and reference turbine power curves to calculate annual and hourly estimates of energy production at a site and choice of turbine model. More information regarding annual energy production is described in section 2.4: Modeling annual energy production and economic potential.1.3 UQ
Results are given without an estimate of their level of uncertainty. Uncertainties should be added, together with an informative discussion of the main contributions to this uncertainty. For instance, one may expect the uncertainty to depend on tower height. Is that the case? If not, why not?
Added to limitations in the Discussion and conclusion section.
1.4 Blade height
Eq. 1 gives a minimum blade height of zero if the canopy percentage is below a certain threshold. This sounds odd and does not fit standard practice. I am concerned that this may lead to an overestimate of the RSA of the sited turbines in the proposed evaluation. If this is not the case, could the authors explain this in the manuscript?
Although the minimum blade height is set to 0 if the canopy coverage is below a certain threshold, Turbine systems are classified and sized according to the minimum and maximum allowable blade height requirements calculated for each parcel using the system configuration information in Table A3. Therefore, there are never any turbines that are sited as “0 m” in our modeling. Additionally, parcels that do not site turbines (or have a determined system size of zero) were filtered from the model pipeline.1.5 Agents and MCA
There is a pipeline for generating agents but little detail is given. Although there is a breakdown between residential and commercial above Table 3, it is not clear what these agents look like. Is some multicriteria analysis performed, as different agents may have a different weighting for different criteria? If the authors do not believe this is relevant, please explain
Agents are categorized as residential, commercial, or industrial according to their land use category (inherent in the Lightbox parcel dataset). This mapping of land use category to sector is identified in Table A1.Responses Reviewer #3:
- The "blade height setback factor" needs to be described and discussed.
The blade height setback factor is identified (and cited) in the “Modeling parcel-level technical potential” section.
Citation: https://doi.org/10.5194/wes-2024-147-AC1 - In response to Reviewer #1’s comments:
Data sets
Distributed Wind Model Results Caleb Phillips https://a2e.energy.gov/project/dw
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 247 | 63 | 11 | 321 | 12 | 10 |
- HTML: 247
- PDF: 63
- XML: 11
- Total: 321
- BibTeX: 12
- EndNote: 10
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1