Overview
First, we establish websites with excessive potential of offshore wind and wave power alongside the coast of the Western Interconnection. We subsequent mannequin candidate era tasks at these websites, (i) filtering out websites which might be in marine protected areas (MPAs) with strict classifications45, navy hazard zones and restricted navy exercise areas46, and (ii) calculating the hourly capability components for every candidate venture for one 12 months of information. Lastly, we use SWITCH, an influence system capability enlargement mannequin, to review the position and impacts of those offshore wind and wave power candidate tasks underneath 25 eventualities with completely different price targets.
A extra detailed overview of the methodology for this research is summarized in Fig. 9.
Overview of methodology used for this research.
Information acquisition and processing
The websites of trade curiosity characterize high-potential wave farm websites alongside the U.S. West Coast. They’re calculated as the results of a scoring framework developed by CalWave. Every website thought-about by CalWave receives a rating between 0 and 100, based mostly on a weighted sum of the next six quantitative parameters: wave power useful resource density, distance to shore, water depth, wind useful resource, bathymetry, and native inhabitants density47. The parameters are weighted based mostly on CalWave’s evaluation of their relative significance to the event of utility-scale wave power infrastructure. CalWave makes use of NREL’s report on Marine Hydrokinetic Power Website Identification and Rating48 as a suggestion for their very own rating framework. The parameters that CalWave considers which coincide with NREL’s report are wave useful resource density and water depth. Some variations between the parameters thought-about by CalWave and NREL are as follows:
Whereas NREL considers market dimension and distance to transmission connection, CalWave considers native inhabitants density and distance to shore
NREL considers power worth and transport price, however CalWave doesn’t
CalWave considers bathymetry and wind useful resource, whereas NREL doesn’t
NREL assigns equal rating to every of the parameters thought-about, whereas CalWave assigns completely different weights to every parameter empirically based mostly on their expertise as a wave power developer and enter from wave power trade and educational ocean power specialists. The parameters so as of assigned weight from highest weight to lowest weight is as follows:
-
1.
Wave useful resource
-
2.
Distance to shore
-
3.
Water depth
-
4.
Wind useful resource
-
5.
Bathymetry
-
6.
Native inhabitants density
-
1.
You will need to point out that the CalWave scoring framework doesn’t use prices and present infrastructure as in comparison with the report from NREL48 as a result of it deliberately encourages the event of wave power infrastructure within the places most technically appropriate. The websites that rank as the highest 100 websites in response to CalWave’s proprietary framework are recognized because the trade websites of curiosity for the U.S. West Coast. Determine 10 reveals every website of curiosity represented by the latitudinal and longitudinal coordinates of its middle (blue factors).
The websites of trade curiosity showing on this determine have been filtered to exclude websites in MPAs and navy hazard zones. All candidate venture areas and BOEM name areas could have offshore wind power, wave power, or each applied sciences put in. Supply knowledge are supplied as a Supply Information file.
When creating candidate tasks for offshore wind and wave power, we first filter the websites of trade curiosity to make sure that no website overlaps with MPAs which have the three strictest classifications45: No Take, No Affect, No Entry. No Take zones “prohibit the extraction or important destruction of pure and cultural assets,” No Affect zones “prohibit all actions that would hurt the positioning’s assets or disrupt the ecological and cultural providers they supply,” and No Entry zones “prohibit all human entry as a way to stop potential ecological disturbance”45. Moreover, the websites are filtered to make sure that no website overlaps with navy hazard zones and restricted navy exercise areas. No United Nations Academic, Scientific and Cultural Group (UNESCO) World Heritage Marine Websites (WHMSs) overlap with any of the websites of trade curiosity49. 4 websites of trade curiosity overlap with these restricted zones, thus they’re faraway from consideration for candidate venture places.
Websites with an ocean depth 60 m or shallower are categorised as fixed-bottom offshore wind assets, and websites with an ocean depth deeper than 60 m are categorised as floating offshore wind assets40. You will need to make this distinction as a result of fixed-bottom and floating offshore wind farms have completely different price targets and technical traits. So as to give every website an space by which arrays of wind generators and WECs might be put in, rectangular polygons are drawn round every website of trade curiosity utilizing QGIS. Determine 10 reveals these candidate venture areas alongside the U.S. West Coast. Every polygon is designed such that no website areas overlap, no MPAs of restricted classification or navy exercise zones are encroached on, and every space falls solely in shallow (≤60-m depth) or deep (>60-m depth) water. The polygons are drawn such that their size is parallel to the shoreline since waves are likely to type parallel to the shoreline. Some polygons in Fig. 10 are so small that they might not seem seen, however be aware that every one trade websites of curiosity are given a corresponding candidate venture space. Some websites which might be very near the coast have restricted areas that they might embody due to close by land within the east path and deep water within the west path.
5 U.S. West Coast offshore wind Name Areas50,51 (Coos Bay, Brookings, Humboldt, Morro Bay, and Diablo Canyon) are added to the listing of candidate tasks, bringing the whole variety of candidate venture areas to 101. Name Areas are potential business offshore wind improvement areas recognized by the Bureau of Ocean Power Administration (BOEM) for public remark in the course of the Name for Info and Nomination stage50. The offshore wind Name Areas are vital to incorporate as candidate venture areas for this research in order that the potential for offshore wind, wave power, and collocated offshore wind and wave power might also be evaluated for these federally recognized websites from a grid capability enlargement planning perspective, along with the wave power websites of trade curiosity. The most important candidate venture space (pink polygons in Fig. 10) is designed to be no bigger than the biggest offshore wind Name Space.
We don’t implement a most water depth on the offshore wind and wave power candidate tasks as a result of it’s unsure what water depths shall be doable to put in marine power units within the 12 months 2050 resulting from technological developments over the approaching many years. Moreover, the BOEM Offshore Wind Name Areas are between 200 m and 1300 m deep. Lower than 5% of the candidate venture areas have any parts of their areas past the 1300 m depth contour.
Check with part 1 of the Supplementary Info for extra particulars associated to the methodology for candidate venture design. There we embody the names and coordinates of the websites faraway from consideration (Supplementary Desk 1) and particulars relating to what traits have been thought-about in the course of the websites of curiosity filtering course of.
Wave power availability might be measured utilizing the numerous wave peak (Hs) and power interval (Te) of a wave. These metrics function enter knowledge for figuring out how a lot energy a WEC can generate. We use all 699,903 coordinates obtainable alongside the U.S. West Coast from the U.S. Division of Power (DOE) Water Energy Know-how Workplace’s (WPTO) U.S. Wave dataset52. This dataset is the very best spatial decision publicly obtainable long-term (1979-2010) wave hindcast dataset52. It has an unstructured grid spatial decision that ranges from 200 meters (in shallow water) to 10 kilometers (in deep water)52. The 699,903 obtainable knowledge factors are generated from the SWAN and WaveWatch III fashions, which have been validated utilizing publicly obtainable spectral knowledge from buoys53.
We overlay these coordinates with the candidate venture areas (Fig. 10) in QGIS to establish 89,650 overlapping coordinates. We use 3-hour time decision time collection of wave traits for the 12 months 2006 corresponding to each ten (to cut back obtain time) of the 89,650 coordinates from52. A complete of 8811 coordinates are downloaded, and every coordinate has a time collection that features timestamps, important wave peak values in meters, power interval values in seconds, and latitude/longitude coordinates related to the places for which knowledge is extracted. We linearly interpolate to transform the time decision of the dataset from 3-hour to 1-hour decision. As a result of excessive spatial decision of the 3-hour dataset, the linearly interpolated knowledge is used to develop the wave attribute time collection used on this research. We assign a time collection to every wave power candidate venture by taking the common time collection of the WPTO coordinates inside every venture space.
The capability issue, CF, is outlined because the ratio between the obtainable producing energy, Pg, and the rated energy capability, Pr, as proven in Eq. (1).
$$CF=frac{{P}_{g}}{{P}_{r}}$$
(1)
Because the capability issue of a WEC is topic to the provision of the first useful resource (e.g., wave power), the capability issue modifications in response to the wave traits on the location the place the WEC is put in at a given time.
On this research, we select the Reference Mannequin 6 (RM6) Oscillating WEC because the consultant WEC54. Its rated energy capability is 350.5 kW and its energy matrix might be downloaded from NREL’s Marine Power Atlas55. The ability matrix reveals the obtainable producing energy of the WEC as a perform of the numerous wave peak (meters) and the power interval (seconds).
We use the wave peak and power interval knowledge from the linearly interpolated 1-hour time decision time collection, the RM6 energy matrix, and Eq. (1) to calculate hourly capability components equivalent to the WPTO coordinates. We calculate a median hourly time collection for the 12 months 2006 corresponding to every candidate venture space by averaging the time collection of the entire WPTO coordinates that fall inside every space. We don’t think about the wake results of WECs as a result of there’s restricted info on this matter, and wake results can differ largely from one WEC design to a different.
So as to decide the utmost doable put in wave power capability at every website, we assume the packing density of the WECs to be 1.0515 MW/km2. To derive this worth, we think about the array format design supplied by the RM6 report54. We calculate the packing density as follows (2):
$$frac{3,{{{rm{WECs}}}}}{1,{{{{rm{km}}}}}^{2}}instances frac{350.5,{{{rm{kW}}}}}{1{{{rm{WEC}}}}}instances frac{1,{{{rm{MW}}}}}{1000,{{{rm{kW}}}}}=1.0515,frac{{{{rm{MW}}}}}{{{{{rm{km}}}}}^{2}}$$
(2)
Check with part 1 of the Supplementary Info to see the small print of two error analyses associated to the wave power capability issue time collection used on this research:
-
1.
To justify taking each 10 of the overlapping factors
-
2.
To confirm that linear interpolation from 3-hour to 1-hour decision for the wave power capability issue time collection doesn’t introduce substantial error
We select the 2020 ATB Reference 15 Wind Turbine and its corresponding energy curve because the consultant wind turbine and energy curve for this research56. This is similar turbine utilized by NREL to develop the moderate-cost goal for offshore wind within the 2022 ATB40. It has a rated energy of 15 MW, a peak of 150 meters, and a rotor diameter of 240 meters56. Equally to the wave power knowledge, we extract the coordinates of the NREL Offshore NW Pacific Dataset57 for 160-meter peak, and we overlay the coordinates with the candidate venture areas proven in Fig. 10 to find out which coordinates overlap. We obtain hourly time collection knowledge of wind traits for all coordinates that lie inside the areas. The NREL Offshore NW Pacific Dataset is a 21-year wind useful resource dataset with a 5-minute time decision created utilizing the Climate Analysis and Forecasting numerical climate prediction mannequin57.
We design 101 offshore wind candidate tasks to occupy the identical areas because the wave power candidate tasks to permit the potential for collocation of those applied sciences. We create an interpolation perform utilizing the facility curve of the turbine whereas contemplating the turbine’s working limits to find out the facility generated by the turbine at any given wind pace in m/s. We assign a time collection to every offshore wind candidate venture by taking the common time collection of the NREL coordinates inside every venture space. A complete of 9207 coordinates lie inside the venture areas. We separate them based mostly on which venture space they fall inside and use them to calculate a median time collection for every space. We compute the hourly offshore wind power capability components because the ratio between the obtainable producing energy and the rated energy capability of the turbine (Eq. (1)).
So as to decide the utmost doable put in offshore wind power capability at every website, we assume the packing density of the offshore wind generators to be 4.3 MW/km2. This worth relies on the common theoretical capability density of the Morro Bay Wind Power Space58, which is a present offshore wind leasing space on the U.S. West Coast. There isn’t a commonplace for offshore wind turbine spacing as a result of packing density can differ based mostly on site-specific circumstances or farm designs. Thus, for simplicity, we assume the identical packing density for all fixed-bottom and floating generators. Moreover, we don’t think about wake results of offshore wind generators on condition that this can be a variable dependant on particular farm array design that may be minimized by builders by strategic design.
SWITCH mannequin
SWITCH36 is a linear programming electrical energy capability enlargement mannequin that finds the least-cost era portfolio and transmission infrastructure topic to electrical energy demand and operational constraints. SWITCH is ready to mannequin a number of funding durations (durations of a number of years the place funding selections are made), e.g., units of many years, and a number of time collection (chronological sequences of grouped timepoints the place operational selections are made) with completely different time decision for every funding interval.
The target perform minimized corresponds to the whole energy system price, i.e., funding and operational prices of era and transmission. The choice variables of the optimization drawback might be summarized within the following units: capability funding selections for every potential new era venture in every interval, capability funding selections for every potential new or present transmission line between any load areas in every interval, hourly dispatch selections for every present and new generator put in for every interval, and selections on hourly transmitted power by the prevailing and new transmission traces.
The principle constraints within the optimization drawback are: energy steadiness in every zone the place energy mills, storage applied sciences, demand and transmission traces are related, electrical energy dispatch of the era applied sciences restricted by their corresponding energy capacities, power flows throughout the transmission traces restricted by their corresponding energy capacities, electrical energy dispatch of renewable power mills additionally restricted by geolocated hourly capability issue time collection, era from every hydropower plant restricted by historic month-to-month availability (minimal, common and most era), biomass and geothermal deployment restricted by the useful resource availability, respect yearly upkeep time for every era know-how, coverage constraints as carbon cap, carbon tax, Renewable Portfolio Requirements, amongst others. For its detailed mathematical description, check with part 6 of the Supplementary Info.
Many analysis teams have additional developed completely different variations of the SWITCH mannequin to research decarbonization pathways in numerous areas1,37,38,39,59,60,61,62,63,64. We use the SWITCH WECC65 mannequin which represents the Western Interconnection by dividing it into 50 geographical zones. The time decision can differ from hourly to sampled hours that characterize typical days in the course of the years being optimized. These modeling virtues of the SWITCH WECC mannequin enable a extra lifelike research of the enlargement and operation of huge regional electrical grids with the presence of renewable intermittent assets.
As talked about beforehand, funding selections are made in durations 2020, 2030, 2040, and 2050 which end in a zero-carbon grid by 2050. Our evaluation within the Outcomes part focuses on ends in 2050. As a reminder, we characterize every interval as ten-year durations by sampling each month in 2020, 2030, 2040, and 2050, two days monthly (median and peak load days) and each 4 hours per day (12 months × 2 days/month × 6 hour/day = 144 hours). Peak days have a weight of 1 and median days of n−1 the place n is the variety of days of that month, and this represents a full month.
Using a four-hour interval as a substitute of the everyday hourly dispatch is a part of the explanation excessive geographic decision may very well be achieved. Moreover, the decreased complexity from utilizing a four-hour time interval permits us to spend extra computational effort on having a excessive geographical decision for potential websites and having 2030, 2040, and 2050 funding selections to higher perceive the transition. A quicker run time from sampling hours additionally allowed us to create many eventualities to guage the relative deployment of offshore wind and wave power.
We mannequin the transmission system of the Western Interconnection utilizing Ventyx geolocated aggregated transmission line knowledge66 and the thermal limits from the Federal Power Regulatory Fee67. In complete, we think about 105 present transmission traces connecting load zones of the Western Interconnection. SWITCH can determine to construct extra transmission traces or increase the capability of present ones whether it is optimum. The mannequin considers transmission line derating and losses.
The electrical energy demand profiles come from historic hourly hundreds from 200668,69 (and ITRON consulting group). These profiles are projected for future years. The mannequin consists of geolocated hourly capability issue time collection for over 7000 potential new places for photo voltaic and land-based wind energy, in addition to potential new places for different renewable power applied sciences (geothermal and biomass). New energy vegetation for nuclear power, hydropower, and geothermal power are additionally included as candidate tasks, in addition to battery power storage and pumped hydro storage. We calculate hourly present and potential new land-based wind farm energy output from the 3TIER Western Wind and Photo voltaic Integration Examine wind pace dataset70,71 utilizing idealized turbine energy output curves on interpolated wind pace values. For present and potential new solar energy vegetation, we simulate the hourly capability components of every venture over the course of the 12 months 2006 utilizing the System Advisor Mannequin from NREL72. The optimization can then select from over 7000 potential new geolocated mills within the Western Interconnection. Gas worth projections for every load space are from the U.S. Power Info Administration73. Capital prices and O&M prices are from NREL ATB 202074. The historic pool of exiting energy vegetation within the Western Interconnection is from the U.S. Power Info Administration (EIA-860, EIA-923, 2020 launch75).
Eventualities description
We search to guage the position that offshore wind and wave power could play in decarbonizing the Western Interconnection by the 12 months 2050. As a result of the target perform in SWITCH minimizes system price, we anticipate deployment of offshore wind and wave power to differ with price. Due to this fact, we design twenty-five eventualities with completely different offshore wind and wave power 2050 price targets. All prices are reported in 2018 U.S. {dollars} (USD), which is the bottom 12 months we use in SWITCH WECC. The 2020 wave power in a single day and O&M prices for all eventualities are $3465/kW and $105.4/kW, respectively. We compute these values by dividing the estimated RM6 WEC in a single day and O&M prices for 10-unit deployment by 1054. This division by 10 is justified by the economies of scale of the candidate tasks designed for this research: the reported prices assume a 10-unit deployment whereas the designed wave power candidate tasks could have a number of a whole bunch of WECs deployed in every website space (based mostly on the packing density assumed and the dimensions of the candidate venture areas). The RM6 report54 demonstrates how decrease prices are related to larger-scale WEC farms. These assumed 2020 wave power prices align with the lower-end of a variety supplied by main wave power builders as an approximation of the present capital expenditure and working expenditure prices of wave power76.
As a reminder, there are 5 completely different price targets for wave power, with probably the most conservative price goal equivalent to a 50% price discount by the 12 months 2050 and probably the most optimistic price goal equivalent to parity between the 2050 in a single day and O&M wave power prices and the NREL 2022 ATB40 utility-scale PV power projected 2050 prices (Fig. 1). As talked about beforehand, we assume a linear projection between the wave power 2020 and 2050 prices. Though we may use studying coefficients to mannequin the decline in wave power prices between 2020 and 2050, formulating correct price projections (or studying/expertise curves) isn’t inside the scope of this work, however it might be thought-about in future work. Moreover, the research in77 makes use of a two-stage Monte Carlo simulation to forecast the levelized price of electrical energy (LCOE) for wave power and finds that the price reductions are practically linear. Thus, we assume a linear development for its simplicity.
Equally, we design 5 offshore wind in a single day and O&M price targets based mostly on the NREL 2022 ATB price projections for fastened and floating offshore wind power (Fig. 1). We use Wind Useful resource Class 3 for fixed-bottom generators, and Wind Useful resource Class 12 for fixed-bottom generators. In line with NREL, Class 3 and Class 12 are probably the most consultant of near-term U.S. fixed-bottom and mid-term U.S. floating offshore wind tasks, respectively40. We design an extra very conservative offshore wind price goal such that $488.39/kW is added to the in a single day prices and $15.90/kW-yr is added to the O&M prices of the NREL 2022 ATB conservative projection for fixed-bottom offshore wind generators (after changing to 2018 {dollars}). For floating offshore wind generators, we add $720.55/kW to the in a single day prices and $14.98/kW-yr to the O&M prices (after changing to 2018 {dollars}) to generate a really conservative situation. Equally, we design an extra very superior offshore wind price goal such that $488.39/kW is subtracted from the in a single day prices and $15.90/kW-yr is subtracted from the O&M prices of the NREL 2022 ATB superior fastened offshore wind projection (after changing to 2018 {dollars}). For floating offshore wind generators, we subtract $720.55/kW from the in a single day prices and $14.98/kW-yr from the O&M prices of the NREL 2022 ATB superior floating offshore wind projection (after changing to 2018 {dollars}). The offsets of $488.39/kW, $15.90/kW-yr, $720.55/kW, and $14.98/kW-yr are chosen as a result of they equal the common distinction between the NREL 2022 ATB reasonable and superior situation in a single day and O&M prices and the reasonable and conservative situation in a single day and O&M prices for fastened and floating offshore wind generators, respectively.
As talked about beforehand, the 5 wave power price targets and 5 offshore wind price targets are mixed into 25 (5 × 5) eventualities, as proven in Fig. 1. Check with Figs. 4–5 of the Supplementary Info for alternate visualizations of the price goal eventualities.
All offshore wind and wave power candidate tasks assume an interconnection price of $487,000/MW of capability put in. That is the common interconnection price for an offshore wind venture with a business operation date of 2023 in 2018 {dollars} from ref. 78.
Limitations
Because the optimization mannequin chooses to collocate extra offshore wind and wave power tasks as their prices lower, we infer that pairing the applied sciences may very well be helpful in a zero-emissions grid as a result of shared land-based infrastructure price financial savings which might be achieved when the applied sciences are collocated. One limitation of this research is that it doesn’t seize the price advantages related to shared underwater transmission infrastructure in collocated offshore wind and wave power farms. Future work is deliberate to moreover seize this advantage of collocation within the mannequin, in addition to to differentiate connection prices throughout offshore power websites in response to every website’s bathymetry.
One other limitation is that present underwater pipelines and cables, present transport routes, and archeological websites apart from these included in UNESCO’s WHMSs usually are not thought-about within the strategy of filtering websites of trade curiosity. We additionally don’t think about the proximity of every website to residential areas on land. We consider that the potential impression from not together with these parameters when filtering the trade websites of is restricted however of native relevance. The principle distinction within the research if these parameters have been capable of be included would seemingly be the shapes of the person candidate venture areas (if they’re modified to keep away from extra ocean zones). Though this will barely alter the capability issue time collection of sure candidate venture areas, we consider it’s unlikely that it will considerably change the system-wide tendencies we observe when integrating numerous quantities of wave and offshore wind power into the Western Interconnection. One facet that would have a bigger impression is that if extra areas are categorised as not appropriate for financial exercise resulting from ecological issues. In that case, our research can present how you can prioritize deployment if much less offshore power put in capability might be deployed.
Moreover, our research doesn’t implement a most water depth on the websites of trade curiosity, as a result of it’s unclear what the restrict of water depth for floating offshore wind generators shall be within the 12 months 2050. Though solely 5 of the 101 candidate venture areas have any parts of their areas past 1500 meters of depth, these websites could also be difficult to develop resulting from their excessive bathymetric circumstances and important distances from shore. Therefore, the inclusion of those 5 websites with out including a value multiplier to account for his or her innate deployment challenges barely diminishes the realisticness of the mannequin. Nevertheless, since lower than 5% of the websites exhibit very deep water, we consider the impression is minimal.
Lastly, the temporal decision of our research (2 consultant days monthly, each 4-hours) could not absolutely seize the distinctive energy output qualities of offshore wind and/or wave power within the mannequin. Though the simplified temporal decision permits us to run a bigger set of eventualities with very excessive spatial decision, it diminishes the quantity of knowledge that the mannequin attracts from the capability issue time collection for every marine generator. We consider {that a} greater temporal decision would result in the identical tendencies noticed on this research, with slight variations within the numerical outcomes. In our earlier research that makes use of SWITCH to guage the impression of utilizing numerous time sampling resolutions on the utilization of long-duration storage (LDS)79, we discover that though the utilization of LDS is affected by the point sampling decision used, the general put in capability combine doesn’t differ largely between the completely different time sampling decision eventualities. Our near-term future work consists of an evaluation of the interplay between offshore wind and wave power and LDS, and we intend to run eventualities with numerous temporal resolutions, together with a situation with hourly decision general three hundred and sixty five days.
Reporting abstract
Additional info on analysis design is offered within the Nature Portfolio Reporting Abstract linked to this text.
Overview
First, we establish websites with excessive potential of offshore wind and wave power alongside the coast of the Western Interconnection. We subsequent mannequin candidate era tasks at these websites, (i) filtering out websites which might be in marine protected areas (MPAs) with strict classifications45, navy hazard zones and restricted navy exercise areas46, and (ii) calculating the hourly capability components for every candidate venture for one 12 months of information. Lastly, we use SWITCH, an influence system capability enlargement mannequin, to review the position and impacts of those offshore wind and wave power candidate tasks underneath 25 eventualities with completely different price targets.
A extra detailed overview of the methodology for this research is summarized in Fig. 9.
Overview of methodology used for this research.
Information acquisition and processing
The websites of trade curiosity characterize high-potential wave farm websites alongside the U.S. West Coast. They’re calculated as the results of a scoring framework developed by CalWave. Every website thought-about by CalWave receives a rating between 0 and 100, based mostly on a weighted sum of the next six quantitative parameters: wave power useful resource density, distance to shore, water depth, wind useful resource, bathymetry, and native inhabitants density47. The parameters are weighted based mostly on CalWave’s evaluation of their relative significance to the event of utility-scale wave power infrastructure. CalWave makes use of NREL’s report on Marine Hydrokinetic Power Website Identification and Rating48 as a suggestion for their very own rating framework. The parameters that CalWave considers which coincide with NREL’s report are wave useful resource density and water depth. Some variations between the parameters thought-about by CalWave and NREL are as follows:
Whereas NREL considers market dimension and distance to transmission connection, CalWave considers native inhabitants density and distance to shore
NREL considers power worth and transport price, however CalWave doesn’t
CalWave considers bathymetry and wind useful resource, whereas NREL doesn’t
NREL assigns equal rating to every of the parameters thought-about, whereas CalWave assigns completely different weights to every parameter empirically based mostly on their expertise as a wave power developer and enter from wave power trade and educational ocean power specialists. The parameters so as of assigned weight from highest weight to lowest weight is as follows:
-
1.
Wave useful resource
-
2.
Distance to shore
-
3.
Water depth
-
4.
Wind useful resource
-
5.
Bathymetry
-
6.
Native inhabitants density
-
1.
You will need to point out that the CalWave scoring framework doesn’t use prices and present infrastructure as in comparison with the report from NREL48 as a result of it deliberately encourages the event of wave power infrastructure within the places most technically appropriate. The websites that rank as the highest 100 websites in response to CalWave’s proprietary framework are recognized because the trade websites of curiosity for the U.S. West Coast. Determine 10 reveals every website of curiosity represented by the latitudinal and longitudinal coordinates of its middle (blue factors).
The websites of trade curiosity showing on this determine have been filtered to exclude websites in MPAs and navy hazard zones. All candidate venture areas and BOEM name areas could have offshore wind power, wave power, or each applied sciences put in. Supply knowledge are supplied as a Supply Information file.
When creating candidate tasks for offshore wind and wave power, we first filter the websites of trade curiosity to make sure that no website overlaps with MPAs which have the three strictest classifications45: No Take, No Affect, No Entry. No Take zones “prohibit the extraction or important destruction of pure and cultural assets,” No Affect zones “prohibit all actions that would hurt the positioning’s assets or disrupt the ecological and cultural providers they supply,” and No Entry zones “prohibit all human entry as a way to stop potential ecological disturbance”45. Moreover, the websites are filtered to make sure that no website overlaps with navy hazard zones and restricted navy exercise areas. No United Nations Academic, Scientific and Cultural Group (UNESCO) World Heritage Marine Websites (WHMSs) overlap with any of the websites of trade curiosity49. 4 websites of trade curiosity overlap with these restricted zones, thus they’re faraway from consideration for candidate venture places.
Websites with an ocean depth 60 m or shallower are categorised as fixed-bottom offshore wind assets, and websites with an ocean depth deeper than 60 m are categorised as floating offshore wind assets40. You will need to make this distinction as a result of fixed-bottom and floating offshore wind farms have completely different price targets and technical traits. So as to give every website an space by which arrays of wind generators and WECs might be put in, rectangular polygons are drawn round every website of trade curiosity utilizing QGIS. Determine 10 reveals these candidate venture areas alongside the U.S. West Coast. Every polygon is designed such that no website areas overlap, no MPAs of restricted classification or navy exercise zones are encroached on, and every space falls solely in shallow (≤60-m depth) or deep (>60-m depth) water. The polygons are drawn such that their size is parallel to the shoreline since waves are likely to type parallel to the shoreline. Some polygons in Fig. 10 are so small that they might not seem seen, however be aware that every one trade websites of curiosity are given a corresponding candidate venture space. Some websites which might be very near the coast have restricted areas that they might embody due to close by land within the east path and deep water within the west path.
5 U.S. West Coast offshore wind Name Areas50,51 (Coos Bay, Brookings, Humboldt, Morro Bay, and Diablo Canyon) are added to the listing of candidate tasks, bringing the whole variety of candidate venture areas to 101. Name Areas are potential business offshore wind improvement areas recognized by the Bureau of Ocean Power Administration (BOEM) for public remark in the course of the Name for Info and Nomination stage50. The offshore wind Name Areas are vital to incorporate as candidate venture areas for this research in order that the potential for offshore wind, wave power, and collocated offshore wind and wave power might also be evaluated for these federally recognized websites from a grid capability enlargement planning perspective, along with the wave power websites of trade curiosity. The most important candidate venture space (pink polygons in Fig. 10) is designed to be no bigger than the biggest offshore wind Name Space.
We don’t implement a most water depth on the offshore wind and wave power candidate tasks as a result of it’s unsure what water depths shall be doable to put in marine power units within the 12 months 2050 resulting from technological developments over the approaching many years. Moreover, the BOEM Offshore Wind Name Areas are between 200 m and 1300 m deep. Lower than 5% of the candidate venture areas have any parts of their areas past the 1300 m depth contour.
Check with part 1 of the Supplementary Info for extra particulars associated to the methodology for candidate venture design. There we embody the names and coordinates of the websites faraway from consideration (Supplementary Desk 1) and particulars relating to what traits have been thought-about in the course of the websites of curiosity filtering course of.
Wave power availability might be measured utilizing the numerous wave peak (Hs) and power interval (Te) of a wave. These metrics function enter knowledge for figuring out how a lot energy a WEC can generate. We use all 699,903 coordinates obtainable alongside the U.S. West Coast from the U.S. Division of Power (DOE) Water Energy Know-how Workplace’s (WPTO) U.S. Wave dataset52. This dataset is the very best spatial decision publicly obtainable long-term (1979-2010) wave hindcast dataset52. It has an unstructured grid spatial decision that ranges from 200 meters (in shallow water) to 10 kilometers (in deep water)52. The 699,903 obtainable knowledge factors are generated from the SWAN and WaveWatch III fashions, which have been validated utilizing publicly obtainable spectral knowledge from buoys53.
We overlay these coordinates with the candidate venture areas (Fig. 10) in QGIS to establish 89,650 overlapping coordinates. We use 3-hour time decision time collection of wave traits for the 12 months 2006 corresponding to each ten (to cut back obtain time) of the 89,650 coordinates from52. A complete of 8811 coordinates are downloaded, and every coordinate has a time collection that features timestamps, important wave peak values in meters, power interval values in seconds, and latitude/longitude coordinates related to the places for which knowledge is extracted. We linearly interpolate to transform the time decision of the dataset from 3-hour to 1-hour decision. As a result of excessive spatial decision of the 3-hour dataset, the linearly interpolated knowledge is used to develop the wave attribute time collection used on this research. We assign a time collection to every wave power candidate venture by taking the common time collection of the WPTO coordinates inside every venture space.
The capability issue, CF, is outlined because the ratio between the obtainable producing energy, Pg, and the rated energy capability, Pr, as proven in Eq. (1).
$$CF=frac{{P}_{g}}{{P}_{r}}$$
(1)
Because the capability issue of a WEC is topic to the provision of the first useful resource (e.g., wave power), the capability issue modifications in response to the wave traits on the location the place the WEC is put in at a given time.
On this research, we select the Reference Mannequin 6 (RM6) Oscillating WEC because the consultant WEC54. Its rated energy capability is 350.5 kW and its energy matrix might be downloaded from NREL’s Marine Power Atlas55. The ability matrix reveals the obtainable producing energy of the WEC as a perform of the numerous wave peak (meters) and the power interval (seconds).
We use the wave peak and power interval knowledge from the linearly interpolated 1-hour time decision time collection, the RM6 energy matrix, and Eq. (1) to calculate hourly capability components equivalent to the WPTO coordinates. We calculate a median hourly time collection for the 12 months 2006 corresponding to every candidate venture space by averaging the time collection of the entire WPTO coordinates that fall inside every space. We don’t think about the wake results of WECs as a result of there’s restricted info on this matter, and wake results can differ largely from one WEC design to a different.
So as to decide the utmost doable put in wave power capability at every website, we assume the packing density of the WECs to be 1.0515 MW/km2. To derive this worth, we think about the array format design supplied by the RM6 report54. We calculate the packing density as follows (2):
$$frac{3,{{{rm{WECs}}}}}{1,{{{{rm{km}}}}}^{2}}instances frac{350.5,{{{rm{kW}}}}}{1{{{rm{WEC}}}}}instances frac{1,{{{rm{MW}}}}}{1000,{{{rm{kW}}}}}=1.0515,frac{{{{rm{MW}}}}}{{{{{rm{km}}}}}^{2}}$$
(2)
Check with part 1 of the Supplementary Info to see the small print of two error analyses associated to the wave power capability issue time collection used on this research:
-
1.
To justify taking each 10 of the overlapping factors
-
2.
To confirm that linear interpolation from 3-hour to 1-hour decision for the wave power capability issue time collection doesn’t introduce substantial error
We select the 2020 ATB Reference 15 Wind Turbine and its corresponding energy curve because the consultant wind turbine and energy curve for this research56. This is similar turbine utilized by NREL to develop the moderate-cost goal for offshore wind within the 2022 ATB40. It has a rated energy of 15 MW, a peak of 150 meters, and a rotor diameter of 240 meters56. Equally to the wave power knowledge, we extract the coordinates of the NREL Offshore NW Pacific Dataset57 for 160-meter peak, and we overlay the coordinates with the candidate venture areas proven in Fig. 10 to find out which coordinates overlap. We obtain hourly time collection knowledge of wind traits for all coordinates that lie inside the areas. The NREL Offshore NW Pacific Dataset is a 21-year wind useful resource dataset with a 5-minute time decision created utilizing the Climate Analysis and Forecasting numerical climate prediction mannequin57.
We design 101 offshore wind candidate tasks to occupy the identical areas because the wave power candidate tasks to permit the potential for collocation of those applied sciences. We create an interpolation perform utilizing the facility curve of the turbine whereas contemplating the turbine’s working limits to find out the facility generated by the turbine at any given wind pace in m/s. We assign a time collection to every offshore wind candidate venture by taking the common time collection of the NREL coordinates inside every venture space. A complete of 9207 coordinates lie inside the venture areas. We separate them based mostly on which venture space they fall inside and use them to calculate a median time collection for every space. We compute the hourly offshore wind power capability components because the ratio between the obtainable producing energy and the rated energy capability of the turbine (Eq. (1)).
So as to decide the utmost doable put in offshore wind power capability at every website, we assume the packing density of the offshore wind generators to be 4.3 MW/km2. This worth relies on the common theoretical capability density of the Morro Bay Wind Power Space58, which is a present offshore wind leasing space on the U.S. West Coast. There isn’t a commonplace for offshore wind turbine spacing as a result of packing density can differ based mostly on site-specific circumstances or farm designs. Thus, for simplicity, we assume the identical packing density for all fixed-bottom and floating generators. Moreover, we don’t think about wake results of offshore wind generators on condition that this can be a variable dependant on particular farm array design that may be minimized by builders by strategic design.
SWITCH mannequin
SWITCH36 is a linear programming electrical energy capability enlargement mannequin that finds the least-cost era portfolio and transmission infrastructure topic to electrical energy demand and operational constraints. SWITCH is ready to mannequin a number of funding durations (durations of a number of years the place funding selections are made), e.g., units of many years, and a number of time collection (chronological sequences of grouped timepoints the place operational selections are made) with completely different time decision for every funding interval.
The target perform minimized corresponds to the whole energy system price, i.e., funding and operational prices of era and transmission. The choice variables of the optimization drawback might be summarized within the following units: capability funding selections for every potential new era venture in every interval, capability funding selections for every potential new or present transmission line between any load areas in every interval, hourly dispatch selections for every present and new generator put in for every interval, and selections on hourly transmitted power by the prevailing and new transmission traces.
The principle constraints within the optimization drawback are: energy steadiness in every zone the place energy mills, storage applied sciences, demand and transmission traces are related, electrical energy dispatch of the era applied sciences restricted by their corresponding energy capacities, power flows throughout the transmission traces restricted by their corresponding energy capacities, electrical energy dispatch of renewable power mills additionally restricted by geolocated hourly capability issue time collection, era from every hydropower plant restricted by historic month-to-month availability (minimal, common and most era), biomass and geothermal deployment restricted by the useful resource availability, respect yearly upkeep time for every era know-how, coverage constraints as carbon cap, carbon tax, Renewable Portfolio Requirements, amongst others. For its detailed mathematical description, check with part 6 of the Supplementary Info.
Many analysis teams have additional developed completely different variations of the SWITCH mannequin to research decarbonization pathways in numerous areas1,37,38,39,59,60,61,62,63,64. We use the SWITCH WECC65 mannequin which represents the Western Interconnection by dividing it into 50 geographical zones. The time decision can differ from hourly to sampled hours that characterize typical days in the course of the years being optimized. These modeling virtues of the SWITCH WECC mannequin enable a extra lifelike research of the enlargement and operation of huge regional electrical grids with the presence of renewable intermittent assets.
As talked about beforehand, funding selections are made in durations 2020, 2030, 2040, and 2050 which end in a zero-carbon grid by 2050. Our evaluation within the Outcomes part focuses on ends in 2050. As a reminder, we characterize every interval as ten-year durations by sampling each month in 2020, 2030, 2040, and 2050, two days monthly (median and peak load days) and each 4 hours per day (12 months × 2 days/month × 6 hour/day = 144 hours). Peak days have a weight of 1 and median days of n−1 the place n is the variety of days of that month, and this represents a full month.
Using a four-hour interval as a substitute of the everyday hourly dispatch is a part of the explanation excessive geographic decision may very well be achieved. Moreover, the decreased complexity from utilizing a four-hour time interval permits us to spend extra computational effort on having a excessive geographical decision for potential websites and having 2030, 2040, and 2050 funding selections to higher perceive the transition. A quicker run time from sampling hours additionally allowed us to create many eventualities to guage the relative deployment of offshore wind and wave power.
We mannequin the transmission system of the Western Interconnection utilizing Ventyx geolocated aggregated transmission line knowledge66 and the thermal limits from the Federal Power Regulatory Fee67. In complete, we think about 105 present transmission traces connecting load zones of the Western Interconnection. SWITCH can determine to construct extra transmission traces or increase the capability of present ones whether it is optimum. The mannequin considers transmission line derating and losses.
The electrical energy demand profiles come from historic hourly hundreds from 200668,69 (and ITRON consulting group). These profiles are projected for future years. The mannequin consists of geolocated hourly capability issue time collection for over 7000 potential new places for photo voltaic and land-based wind energy, in addition to potential new places for different renewable power applied sciences (geothermal and biomass). New energy vegetation for nuclear power, hydropower, and geothermal power are additionally included as candidate tasks, in addition to battery power storage and pumped hydro storage. We calculate hourly present and potential new land-based wind farm energy output from the 3TIER Western Wind and Photo voltaic Integration Examine wind pace dataset70,71 utilizing idealized turbine energy output curves on interpolated wind pace values. For present and potential new solar energy vegetation, we simulate the hourly capability components of every venture over the course of the 12 months 2006 utilizing the System Advisor Mannequin from NREL72. The optimization can then select from over 7000 potential new geolocated mills within the Western Interconnection. Gas worth projections for every load space are from the U.S. Power Info Administration73. Capital prices and O&M prices are from NREL ATB 202074. The historic pool of exiting energy vegetation within the Western Interconnection is from the U.S. Power Info Administration (EIA-860, EIA-923, 2020 launch75).
Eventualities description
We search to guage the position that offshore wind and wave power could play in decarbonizing the Western Interconnection by the 12 months 2050. As a result of the target perform in SWITCH minimizes system price, we anticipate deployment of offshore wind and wave power to differ with price. Due to this fact, we design twenty-five eventualities with completely different offshore wind and wave power 2050 price targets. All prices are reported in 2018 U.S. {dollars} (USD), which is the bottom 12 months we use in SWITCH WECC. The 2020 wave power in a single day and O&M prices for all eventualities are $3465/kW and $105.4/kW, respectively. We compute these values by dividing the estimated RM6 WEC in a single day and O&M prices for 10-unit deployment by 1054. This division by 10 is justified by the economies of scale of the candidate tasks designed for this research: the reported prices assume a 10-unit deployment whereas the designed wave power candidate tasks could have a number of a whole bunch of WECs deployed in every website space (based mostly on the packing density assumed and the dimensions of the candidate venture areas). The RM6 report54 demonstrates how decrease prices are related to larger-scale WEC farms. These assumed 2020 wave power prices align with the lower-end of a variety supplied by main wave power builders as an approximation of the present capital expenditure and working expenditure prices of wave power76.
As a reminder, there are 5 completely different price targets for wave power, with probably the most conservative price goal equivalent to a 50% price discount by the 12 months 2050 and probably the most optimistic price goal equivalent to parity between the 2050 in a single day and O&M wave power prices and the NREL 2022 ATB40 utility-scale PV power projected 2050 prices (Fig. 1). As talked about beforehand, we assume a linear projection between the wave power 2020 and 2050 prices. Though we may use studying coefficients to mannequin the decline in wave power prices between 2020 and 2050, formulating correct price projections (or studying/expertise curves) isn’t inside the scope of this work, however it might be thought-about in future work. Moreover, the research in77 makes use of a two-stage Monte Carlo simulation to forecast the levelized price of electrical energy (LCOE) for wave power and finds that the price reductions are practically linear. Thus, we assume a linear development for its simplicity.
Equally, we design 5 offshore wind in a single day and O&M price targets based mostly on the NREL 2022 ATB price projections for fastened and floating offshore wind power (Fig. 1). We use Wind Useful resource Class 3 for fixed-bottom generators, and Wind Useful resource Class 12 for fixed-bottom generators. In line with NREL, Class 3 and Class 12 are probably the most consultant of near-term U.S. fixed-bottom and mid-term U.S. floating offshore wind tasks, respectively40. We design an extra very conservative offshore wind price goal such that $488.39/kW is added to the in a single day prices and $15.90/kW-yr is added to the O&M prices of the NREL 2022 ATB conservative projection for fixed-bottom offshore wind generators (after changing to 2018 {dollars}). For floating offshore wind generators, we add $720.55/kW to the in a single day prices and $14.98/kW-yr to the O&M prices (after changing to 2018 {dollars}) to generate a really conservative situation. Equally, we design an extra very superior offshore wind price goal such that $488.39/kW is subtracted from the in a single day prices and $15.90/kW-yr is subtracted from the O&M prices of the NREL 2022 ATB superior fastened offshore wind projection (after changing to 2018 {dollars}). For floating offshore wind generators, we subtract $720.55/kW from the in a single day prices and $14.98/kW-yr from the O&M prices of the NREL 2022 ATB superior floating offshore wind projection (after changing to 2018 {dollars}). The offsets of $488.39/kW, $15.90/kW-yr, $720.55/kW, and $14.98/kW-yr are chosen as a result of they equal the common distinction between the NREL 2022 ATB reasonable and superior situation in a single day and O&M prices and the reasonable and conservative situation in a single day and O&M prices for fastened and floating offshore wind generators, respectively.
As talked about beforehand, the 5 wave power price targets and 5 offshore wind price targets are mixed into 25 (5 × 5) eventualities, as proven in Fig. 1. Check with Figs. 4–5 of the Supplementary Info for alternate visualizations of the price goal eventualities.
All offshore wind and wave power candidate tasks assume an interconnection price of $487,000/MW of capability put in. That is the common interconnection price for an offshore wind venture with a business operation date of 2023 in 2018 {dollars} from ref. 78.
Limitations
Because the optimization mannequin chooses to collocate extra offshore wind and wave power tasks as their prices lower, we infer that pairing the applied sciences may very well be helpful in a zero-emissions grid as a result of shared land-based infrastructure price financial savings which might be achieved when the applied sciences are collocated. One limitation of this research is that it doesn’t seize the price advantages related to shared underwater transmission infrastructure in collocated offshore wind and wave power farms. Future work is deliberate to moreover seize this advantage of collocation within the mannequin, in addition to to differentiate connection prices throughout offshore power websites in response to every website’s bathymetry.
One other limitation is that present underwater pipelines and cables, present transport routes, and archeological websites apart from these included in UNESCO’s WHMSs usually are not thought-about within the strategy of filtering websites of trade curiosity. We additionally don’t think about the proximity of every website to residential areas on land. We consider that the potential impression from not together with these parameters when filtering the trade websites of is restricted however of native relevance. The principle distinction within the research if these parameters have been capable of be included would seemingly be the shapes of the person candidate venture areas (if they’re modified to keep away from extra ocean zones). Though this will barely alter the capability issue time collection of sure candidate venture areas, we consider it’s unlikely that it will considerably change the system-wide tendencies we observe when integrating numerous quantities of wave and offshore wind power into the Western Interconnection. One facet that would have a bigger impression is that if extra areas are categorised as not appropriate for financial exercise resulting from ecological issues. In that case, our research can present how you can prioritize deployment if much less offshore power put in capability might be deployed.
Moreover, our research doesn’t implement a most water depth on the websites of trade curiosity, as a result of it’s unclear what the restrict of water depth for floating offshore wind generators shall be within the 12 months 2050. Though solely 5 of the 101 candidate venture areas have any parts of their areas past 1500 meters of depth, these websites could also be difficult to develop resulting from their excessive bathymetric circumstances and important distances from shore. Therefore, the inclusion of those 5 websites with out including a value multiplier to account for his or her innate deployment challenges barely diminishes the realisticness of the mannequin. Nevertheless, since lower than 5% of the websites exhibit very deep water, we consider the impression is minimal.
Lastly, the temporal decision of our research (2 consultant days monthly, each 4-hours) could not absolutely seize the distinctive energy output qualities of offshore wind and/or wave power within the mannequin. Though the simplified temporal decision permits us to run a bigger set of eventualities with very excessive spatial decision, it diminishes the quantity of knowledge that the mannequin attracts from the capability issue time collection for every marine generator. We consider {that a} greater temporal decision would result in the identical tendencies noticed on this research, with slight variations within the numerical outcomes. In our earlier research that makes use of SWITCH to guage the impression of utilizing numerous time sampling resolutions on the utilization of long-duration storage (LDS)79, we discover that though the utilization of LDS is affected by the point sampling decision used, the general put in capability combine doesn’t differ largely between the completely different time sampling decision eventualities. Our near-term future work consists of an evaluation of the interplay between offshore wind and wave power and LDS, and we intend to run eventualities with numerous temporal resolutions, together with a situation with hourly decision general three hundred and sixty five days.
Reporting abstract
Additional info on analysis design is offered within the Nature Portfolio Reporting Abstract linked to this text.
