Ocean currents (excluding tidal currents) are mainly generated by wind and affected by Coriolis forces (Rossby number <1). While they can be affected by water density gradients, the dynamics of ocean circulation lead to ocean current patterns that can be largely unrelated to the local winds. In particular, there is the phenomenon of “western intensification,” whereby wind stress causes strong, narrow currents that carry warm water from the tropics toward the poles along the western boundaries of ocean basins. One example is the Gulf Stream, with an ocean current in the Florida Strait that can exceed 2 m/s (Hanson et al., 2011).
The potential of exploiting ocean currents for power generation, in much the same way as is done with wind, has long been a matter of interest. However, the amount of power put into steady ocean currents by the local winds is estimated to be no more than about 1 TW (Wunsch, 1998). As expected, this is similar to the estimated losses through bottom friction (Arbic et al., 2009; Müller et al., 2010). It thus seems unlikely that power approaching 1 TW could be extracted from ocean currents worldwide without significant back effects. Nonetheless, turbines in strong currents might be able to provide significant amounts of power in some locations.
A single submerged turbine placed in an ocean current with a swept area A in water with density p and a current of speed v can produce power up to the Lanchester-Betz limit of 0.3pAv3.1 For example, if A is 1,000 m2 and v is 2 m/s, the power calculated is 2.4 MW. If currents are to
1 The Lanchester-Betz limit applies to a turbine in an unbounded flow.
be exploited, the cubic power of v demonstrates the advantage of deploying turbines in regions of strong ocean currents. As the density of water is approximately 850 times that of air, a marine turbine in an ocean current of 1 m/s can theoretically produce as much power as a wind turbine with the same swept area A in a wind with a wind speed of 9 m/s.2
Similarly, the drag on an ocean current device due to the higher water density is likely to result in forces on the marine turbine that are greater than that on a wind turbine. This can present significant engineering challenges. As mentioned in Chapters 2 and 7, problems with corrosion may also be significant for marine turbines, though this is also a consideration for offshore wind turbines. The reduction in turbine performance from biofouling of marine turbines may be more significant than the loss of performance caused by blade pitting due to dust particles and insects.
Based on presentations given by the ocean current resource assessment group,3,4,5 the ocean current energy assessment is being conducted using two different approaches: (1) estimation of the total ocean current technical resource around U.S. coastal waters from the Pk method, which used the predicted undisturbed flow field output from an ocean model (see Chapter 2, “Estimate of Available Tidal Power,” for more discussion on Pk), and (2) estimation of the total available power within the Gulf Stream by incorporating extra dissipation to represent energy extraction into theoretical models of the Gulf Stream western boundary current. Work using the first approach has been mostly completed,6 but investigation of the second approach was not complete when this report was written.
In the first approach, the theoretical resource (kinetic power density) and the technical resource (Pk) are calculated using velocity fields generated
2 It should be noted that a wind turbine deployed in an area with 9 m/s wind speed is likely to sweep out a much larger area than 100 m2.
3 K. Haas, H.M. Fritz, Z. Defne, and X. Yang, Georgia Tech Savannah; S.P. French and X. Shi, Georgia Tech Atlanta; V.S. Neary, P. Schweizer, and B. Gunawan, Oak Ridge National Laboratory, “Assessment of energy production potential from ocean currents along the United States coastline,” Presentation to the committee on September 27, 2011.
4 K. Haas, H.M. Fritz, Z. Defne, and X. Yang, Georgia Tech Savannah; S.P. French and X. Shi, Georgia Tech Atlanta; V.S. Neary, P. Schweizer, and B. Gunawan, Oak Ridge National Laboratory, “Assessment of energy production potential from ocean currents along the United States coastline,” Presentation to the committee on December 12, 2011.
5 K. Haas, H.M. Fritz, Z. Defne, and X. Yang, Georgia Tech Savannah; S.P. French and X. Shi, Georgia Tech Atlanta; V.S. Neary, P. Schweizer, and B. Gunawan, Oak Ridge National Laboratory, “Assessment of energy production potential from ocean currents along the United States coastline,” Presentation to the committee on April 9, 2012.
6 See preceding footnote.
from regional or global ocean models. To select the best suited models, the group initially assessed the configurations and performance of a set of models, including HYCOM-GLOBAL (Hybrid Coordinate Ocean Model), HYCOM-GOM (Gulf of Mexico), JPL-ROMS (NASA Jet Propulsion Laboratory Regional Ocean Model System), and NCOM (Navy Coastal Ocean Model).7 HYCOM-GLOBAL is a real-time 1/12-degree global nowcast/ forecast ocean circulation model maintained by the Naval Research Laboratory8 and sponsored by the National Ocean Partnership Program, while HYCOM-GOM is a regional model with grid resolution of 1/25 degree nested within the HYCOM-GLOBAL model. HYCOM employs a hybrid coordinate system that combines (1) an isopycnal coordinate in the stratified ocean, (2) a terrain-following coordinate in shallow coastal regions, and (3) a z-level coordinate in the mixed layer and/or unstratified seas. This unique approach in vertical coordinates gives HYCOM the ability to simulate different physical processes in the ocean at different scales using a single model (Halliwell et al., 2000). Power density distributions and variability and confidence intervals of the probability distributions are calculated based on the daily HYCOM-GLOBAL and HYCOM-GOM results from 2004 to the present. Model selection for the ocean current energy assessment was based on a number of criteria, including model error statistics, model grid resolution, length of simulation period, and model output intervals. HYCOM-GLOBAL was selected for the U.S. West Coast, Alaska, and Hawaii coastal regions; a hybrid NCOM-HYCOM was selected for the East Coast; and HYCOM-GOM was selected for the Gulf of Mexico and the Florida Current. The assessment group created an ocean current database for coastal waters, up to 200 nautical miles offshore, using the selected models.
An independent validation of the database was conducted by the Oak Ridge National Laboratory (ORNL, 2012). The main challenge of validating the predicted ocean currents is the paucity of observational data. The most complete datasets are those from high-frequency (HF) radar and from a cable off Florida’s east coast (both of which were quite limited in time), and an additional stationary acoustic Doppler current profiler (ADCP) dataset from Florida Atlantic University was also used. In general, the model results had a reasonably good match to the ADCP data but a poor one to the HF radar. Predicted daily flow from the model matched the distribution pattern of the submarine cable data, although the cable data showed a higher percentage of occurrence in the maximum flow range.
7 See footnote 3.
A Web-based geographic information system (GIS) for the ocean current resource assessment will be created to disseminate the study results, as was done for the tidal energy assessment study (Georgia Tech Research Corporation, 2011). The GIS database was developed using ArcView GIS software that allows for downloading model results for further analysis, such as GIS layers of the monthly and yearly means, standard deviations9 and power densities, probability distributions for current speed and direction, and the effective power using a specified number of turbines, efficiencies, and dimensions.
The project group finished compiling the database for flow field output from the models, and the validation group has a draft report of the group’s results (Kevin Haas, personal communication; ORNL, 2012). The group began analysis of the database in June 2012. Web page design is scheduled to be completed by spring 2013 and the final estimate of total theoretical resource will be completed by June 2013.
As part of the model selection process, model results were compared to about 1,250 Argo temperature and salinity profiling floats.10 Scatter plots of current speed indicated that the Argo data were more scattered than model results, which is likely due to the difference in time interval between the model results (daily) and the Argo data (every 6 hours). Similarly, the higher occurrence rate in the maximum flow range for submarine cable data used during the validation process could also be caused by a shorter (6-hour) data interval. Comparison of model results and data might be improved by creating daily averages of both the Argo and the cable data.
Although HYCOM provides the most complete model outputs that are available for the ocean current energy assessment, there is a concern that using the undisturbed velocity field for the estimates of the theoretical and technical resource without considering the back effect due to the presence of turbines will overestimate the ocean current resource.
To calculate the technical resource, the assessment group used the Pk formulation (Haas et al., 2012), which accounts for the turbine array configuration (e.g., turbine size, spacing, efficiency):
9 In an early presentation, the project group indicated that it used a 2 percent exceedence value; the committee strongly suggested using a simple standard deviation instead.
where Ef is efficiency, As is the swept area of the turbine (m2), Ac is the surface area of the computation cell (m2), and N is the number of turbines per unit surface area. As discussed in Chapter 2, Pk is meaningful only when it is assumed that back effects on the surrounding flow fields are small. This assumption fails when Pk becomes large as the turbine size and density increase, causing Pk to overestimate the technical resource.
The validation for the ocean current resource assessment was conducted for the Florida Strait region with a very limited data set for HF radar (184 days) and ADCP measurements (1 year) (Haas et al., 2012; ORNL, 2012). A 7-year record of submarine telephone cable data was also used, although it did not provide direct current measurement. The validation results indicate some discrepancies between the model and the independent data but find that the HYCOM-GOM model provides an acceptable representation of Florida Strait ocean currents. Because there are extensive observational studies of the Florida Current (e.g., Schmitz and Richardson, 1968; Larsen and Sanford, 1986; Johns and Schott, 1987; Meinen et al., 2010), the validation could have been strengthened by obtaining more existing data for comparison with the model results. This is also noted in the validation report, with an emphasis on the value of additional stationary ADCP data.
Estimate of Ocean Power Potential
Although model results are available along the entire U.S. coast, calculated mean power density indicated that ocean energy in the Gulf Stream is significantly higher than that in other U.S. coastal waters. Therefore, the main effort of the resource assessment is focused on the Gulf Stream, primarily the Florida Current in the Gulf of Mexico. To demonstrate the Pk methodology, the assessment group presented some preliminary results of the total technical resource available for the Florida Current using a set of specific values for turbine configuration (Haas et al., 2012). Based on the group’s calculation, Pk in the Florida Current is 14.1 GW, ∼62 percent of the calculated total natural kinetic energy flux (∼20 GW). The committee is concerned about such a high Pk value, as the back effect on the currents could be significant. While the potential of a single turbine, or a small number of turbines, may be estimated using the existing currents, an array of turbines will impede and reduce the current in the region. Thus, it is incorrect to estimate power potential by adding up the potential of all the currents in an arbitrary array of turbines on the assumption that the current is unaffected.
Allowing for the back effects of an in-stream turbine array deployed in a limited region of a larger scale flow was discussed in Chapter 2. As noted, a theoretical study by Garrett and Cummins (2013) examined the maximum power that could be obtained from an array of turbines in an otherwise uniform region without lateral boundaries. The study assumes water of constant depth, with the turbines effectively assumed to occupy the whole vertical water column so that the flow is modeled as two-dimensional (horizontal variations only). The effect of the turbines is represented as a drag in addition to any natural friction. As the additional drag is increased, the power also increases at first, but the currents inside the circle decrease as the flow is diverted and, as in other situations, there is a point at which the extracted power starts to decrease. The maximum power obtainable from the turbine array depends strongly on the local flow. In most instances, the power from the array of turbines will be limited to a fraction of the kinetic energy flux impinging on the turbine array, most likely no more than approximately 0.7 times the incident flux. If, however, the array is very large (for example, hundreds of square kilometers), the theoretical power could be limited by friction and would be between 50 and 75 percent11 of the natural frictional dissipation of the undisturbed flow in the region containing the turbines (Garrett and Cummins, 2013).
In other words, more than a fraction of the incident energy flux can be obtained only for very large arrays, and then it scales with the natural dissipation in the region. The technical resource will be less than the theoretical resource, possibly considerably less, because of factors that include wake losses and drag on supporting structures. Furthermore, this estimate does not allow for a reduction of the incident flow due to the impact of a large turbine farm on the larger-scale regional flow. High-resolution numerical modeling of disturbed flow due to MHK array deployment would provide a fundamental theoretical foundation for optimizing turbine locations. An optimal layout would locate turbines outside the wake of upstream turbines, while minimizing the distance between devices to lower cable costs. Because these types of optimizations are site-specific, it would be useful to model the most favorable layout for the Florida Current to provide a better estimate of the available technical resource.
The committee further considered the global potential of ocean currents. While there is 800 GW of global dissipation by steady currents, 80 percent of that occurs in the Southern Ocean. B. Polagye and M. Kawase of the University of Washington (personal communication, 2011) apportion the rest by area, assigning only 20 GW to the North Atlantic.
11 The fraction is 0.5 if the friction is taken to be linear in the local current speed and 0.75 if it is taken to be quadratic.
A greater estimate of 70 GW is obtained by Csanady (1988). However, if only a small fraction can be extracted without noticeably disrupting natural ocean circulation, it is unlikely that more than a few gigawatts can be obtained from the ocean current resource in U.S. waters.
The most promising site for the exploitation of strong currents in the North Atlantic is in the Florida Straits. Modeling the flow as if it were in a confined channel would be inappropriate, as the flow takes the form of a jet with weak currents on either side of the strait. Placing turbines in the jet would tend to broaden it, maintaining the volume flux to a first approximation but reducing the current speed. If the reduction in speed is to be no more than 10 percent, then the committee estimates that no more than 20 percent (4 GW) of the 20 GW incident energy flux could be extracted. Allowing for wake losses, drag on supporting structures, and internal turbine and transmission losses, it is unlikely that more than 1 or 2 GW could practically be transmitted to the electricity grid. Additionally, high turbine density in the water column may substantially divert the Florida Current and force the current flow around the Bahamas. This would reduce the local volume flux, creating a practical extractable power that would be even less than 1 or 2 GW. These preliminary estimates need refinement to account for the actual current profile and for stratification effects. The ocean current assessment group should properly account for back effects by simulating total extractable energy using three-dimensional numerical models that include representation of turbine arrays.
The committee also believes that the assessment group needs to further explore and discuss the effects of meandering and seasonal variability of the Florida Current on the extractable power estimate, as the current shows strong meandering and seasonal variability at various frequencies (Johns and Schott, 1987; Lee et al., 1995).12 These aspects of spatial and temporal variability in the resource could potentially limit the placement of MHK devices to narrow regions with consistent flow and could impact the ability to bring ocean current power into the electrical grid. Furthermore, an accurate assessment of the large-scale technical ocean current resource requires consideration of the near-field wake effect near the device.
The ocean current resource assessment is valuable because it provides a rough estimate of ocean current power in U.S. coastal waters. However, less time could have been spent looking at the West Coast
in order to concentrate more fully on the Florida Current. Overall, the committee does not expect the practical resource of the Florida Current to be more than 1 or 2 GW, though further work is required. Given that observation in the region of the Florida Current is very limited, a comprehensive field observation with focus on the Florida Current is necessary in the future. Additionally, the ocean current validation was conducted with a very limited data set and could be substantially strengthened by including more existing data for comparison with model results.
Recommendation: Any follow-on work for the Florida Current should include a thorough evaluation of back effects related to placing turbine arrays in the strait by using detailed numerical simulations that include the representation of extensive turbine arrays. Such models should also be used to investigate array optimization of device location and spacing. The effects of meandering and seasonal variability within the Florida Current should also be discussed.