Assessment of Energy Production Potential from Tidal Streams in the United States


Dr. Kevin A. Haas, Dr. Hermann M. Fritz, Dr. Zafer Defne, Dr. Lide Jiang

Georgia Tech Savannah, School of Civil and Environmental Engineering


Dr. Steven P. French, Dr. Xuan Shi

Georgia Tech Atlanta, Center for Geographic Information Systems


Dr. Brennan T. Smith, Dr. Vincent Neary, Kevin Stewart

Oak Ridge National Laboratory, Environmental Sciences Division




Tidal streams are high velocity sea currents created by periodic horizontal movement of the tides, often magnified by local topographical features such as headlands, inlets to inland lagoons, and straits. As tides ebb and flow, currents are often generated in coastal waters. In many places the shape of the seabed forces water to flow through narrow channels, or around headlands. Tidal stream energy extraction is derived from the kinetic energy of the moving flow; analogous to the way a wind turbine operates in air, and as such differs from tidal barrages, which relies on providing a head of water for energy extraction. A tidal stream energy converter extracts and converts the mechanical energy in the current into a transmittable energy form. A variety of conversion devices are currently being proposed or are under active development, from a water turbine similar to a scaled wind turbine, driving a generator via a gearbox, to an oscillating hydrofoil which drives a hydraulic motor.


Tidal energy is one of the fastest growing emerging technologies in the renewable sector and is set to make a major contribution to carbon free energy generation. Tidal energy can be harnessed wherever there is moving water in significant volumes. The key advantage of tidal streams is the deterministic and precise energy production forecast governed by astronomy. In addition, the predictable slack water facilitates deployment and maintenance. In 2005, EPRI was first to study representative sites (Knik Arm, AK; Tacoma Narrows, WA; Golden Gate, CA; Muskeget Channel, MA; Western Passage, ME) without mapping the resources [1]. Additional favorable sites exist in Puget Sound, New York, Connecticut, Cook Inlet, Southeast Alaska, and the Aleutian Islands among others. Besides large scale power production, tidal streams may serve as local and reliable energy source for remote and dispersed coastal communities and islands. The extractable resource is not completely known; assuming 15% level of extraction, EPRI has documented 16 TWh/yr in Alaska, 0.6 TWh/yr in Puget Sound, and 0.4 TWh/yr in CA, MA, and ME [2-6]. The selection of location for a tidal stream energy converter farm is made upon assessment of a number of criteria:

o   Tidal current velocity: the speed and volume of water passing through the site in space and time.

o   Site characteristics:  bathymetry, water depth, geology of the seabed and environmental impacts will determine the deployment method needed and the cost of installation.

o   Electrical grid connection: the seafloor cable distance from the proposed site to a grid access point will help determine the viability of an installation.


Following the guidelines in the EPRI report for estimating tidal current energy resources [7], preliminary investigations of the tidal currents can be conducted based on the tidal current predictions provided by NOAA tidal current stations [8]. There are over 2700 of these stations which are sparsely distributed in inlets, rivers, channels and bays. The gauge stations are concentrated along navigation channels, harbors and rivers but widely lacking elsewhere along the coast. As an example, the maximum powers at some of these locations around the Savannah River on the coast of Georgia are shown in Figure 1. The tidal power per unit area, power density, given in this figure were calculated using the equation

where r is the density of water and V is the magnitude of the depth averaged maximum velocity.


Figure 1. Maximum available power per unit area based on NOAA tidal current predictions in the vicinity of the Savannah River.


As seen in Figure 1, these tidal currents can have significant spatial variability; therefore, measurements (or predictions) of currents at one location are generally a poor indicator of conditions at another location, even nearby.  It is clear that the majority of the data is available along the navigation channel in the Savannah River, with sparse data within the rest of the tidal area. EPRI suggest a methodology using continuity and the Bernoulli equation for determining the flow in different sections of a channel [7].  This is a reasonable approach for flow along a geometrically simple channel, but is not applicable for the flow in the complex network of rivers and creeks along much of the US coastline.  Thus we have applied a state-of-the-art numerical model for simulating the tidal flows along the coast of the entire United States including Alaska, Hawaii and Puerto Rico. 



1.      Utilize an advanced ocean circulation numerical model (ROMS) to predict tidal currents.

2.      Validate the velocities and water levels predicted by the model with available data.

3.      Compute the tidal harmonic constituents for the velocities and water levels.

4.      Build a GIS database of the tidal constituents.

5.      Develop GIS tools for dissemination of the data

a.      A filter based on depth requirements.

b.      Compute current velocity histograms based on the tidal constituents.

c.       Compute the available power density (W/m2) based on the velocity histograms.

d.      Compute the total available power within arrays based on turbine performance parameters.

6.      Develop a web based interface for accessing the GIS database and using the GIS tools.




Task 1.0  Application of ROMS for simulation of tidal currents

The Regional Ocean Modeling System (ROMS) has been configured to simulate the tidal flows along the coast of the United States.  More details about the model setup and application can be found under the model documentation in the website help menu.


Task 2.0 Compute harmonic constituents

The model output time series at 1-hr intervals, from which the T_TIDE harmonic analysis toolbox [9] for MATLAB was used to extract the harmonic constituents. The program was run for each grid point, and the constituents extracted included Q1, O1, K1, S2, M2, N2, K2, M4 and M6. K2 constituent was not extracted on the West coast and Alaska domains since it was not included in the tidal forcing. Nodal correction was also used by providing the start time of the simulation and the latitude of the location.


Task 3.0 Validate the model output

ORNL performed a verification of the tidal energy resource database and tools.  Comparison was made with approximately 25 primary NOAA tidal data stations, including stations on the east and west continental coasts, as well as Alaska and Hawaii.  Selection of tidal stations for verification was focused on stations near high-energy sites as indicated by the model results.  Multiple statistics and parameters were used to compare tidal station data to the model database, including tidal constituents (magnitude and phase of harmonics), velocity histograms, and a limited number of tidal elevation and current time series comparisons. 


Task 4.0 Build GIS database

The GIS model consists of a database containing results from the tidal model and several computational tools to facilitate the extraction of useful information by the users. The database consists of the tidal constituents for the water level, depth-averaged, and the MLLW depths at a high resolution (on the order of hundred meters spacing).  These tidal constituents can be used to derive velocity, power density and other parameters of interest as requested by the user in near real time


Task 5.0 Develop GIS tools

The GIS tools allow the user to view the full spatial distribution of the pre-calculated available power density and then to input power density and bathymetric constraints to tailor the output for particular regions.


Subtask 5.1 Depth filter

Tidal stream energy converters are currently limited in their variety and are primarily classified in vertical and horizontal axis devices with open or shrouded rotors. Independent of their design all the devices have depth requirement based on their dimensions. The first step for assisting in site selection is to determine which locations meet the minimum depth requirements. Generally these requirements are based on minimum height of the prototype above the bed (hb), the minimum clearance of the prototype below the surface (hs) and the device dimensions (dp). The minimum depth (hmin) would then be given as .


Subtask 5.2 Velocity histograms

The model runs produce time series of the velocity which are 32 days long.  The first 2 days were neglected for the computations of the tidal constituents.  For the 32 day simulations, these constituents were used to create a new time series of the velocity for an entire year.  The one year of hourly data was then be used to create a probability histogram of the velocity magnitude.  The tool computes the histograms for sections of the coast as specified by the user.  The user can also view or extract the resulting histograms or other statistics at any particular location.


Subtask 5.3 Compute available power densities

The histograms of distribution of annual tidal current velocity were used to compute a histogram of total available power density.  These histograms can be used to compute the annual average available power at all locations.  Similar to the velocity histograms, the user can view or extract the histograms and statistics of the available power density.  In addition, the spatial distribution of the average annual available power was computed and displayed as the pre-calculated available power for the webpage. This can be filtered by the depth constraint previously specified. 


Subtask 5.4 Compute total available power

Based on the feedback from a project workshop in Atlanta with outside experts, the Garrett and Cummings [10] method for calculating the tidal stream energy potential was applied to the 4-5 regions with the most potential to provide an initial estimate of the available resource with a gamma value of 1.


Task 6.0 Web based interface

Results from this study have been made available via an internet web site. An interactive, web-based GIS system has been developed to facilitate dissemination of the tidal data to interested users in a manner equally accessible and useful to both specialists and a lay audience.



[1] Bedard R, Previsic M, Polagye B, Hagerman G, Casavant A. North American Tidal In Stream Energy Conversion Technology Feasibility Study. EPRI TP-008-NA. Electric Power Research Institute. 2006.

[2] Polagye B, Bedard R. Tidal In-Stream Energy Resource Assessment for Southeast Alaska. EPRI-TP-003 AK. Electric Power Research Institute. 2006.

[3] Hagerman G, Fader G, Carlin G, Bedard R. Nova Scotia Tidal In-Stream Energy Conversion (TISEC): Survey and Characterization of Potential Project Sites. EPRI - TP- 003 NS Rev 2. Electric Power Research Institute. 2006.

[4] Hagerman G, Fader G, Bedard R. New Brunswick Tidal-In Stream Energy Conversion (TISEC): Survey and Characterization of Potential Project Sites. EPRI -TP-003 NB Rev 1. Electrical Power Research Institute. 2006.

[5] Hagerman G, Bedard R. Maine Tidal In-Stream Energy Conversion (TISEC): Survey and Charaterization of Potential Project Sites. EPRI -TP-003 ME Rev 1. Electric Power Research Institute. 2006.

[6] Hagerman G, Bedard R. Massachusetts Tidal In-Stream Energy Conversion (TISEC): Survey and Characterization of Potential Project Sites. EPRI -TP-003 MA Rev 1. Electric Power Research Institute. 2006.

[7] Hagerman G, Polagye B, Bedard R, Previsic M. Methodology for Estimating Tidal Current Energy Resources and Power Production by Tidal In-Stream Energy Conversion (TISEC) Devices. EPRI-TP-001 NA Rev 3. Electric Power Research Institute. 2006.

[8] Tides and Currents. National Oceanic and Atmospheric Administration. <>. (Accessed 2008).

[9] Pawlowicz R, Beardsley B, Lentz S. Classical tidal harmonic analysis including error estimates in MATLAB using T_TIDE. Computers & Geosciences. 2002;28:929-37.

[10] Garrett C, Cummins P. The power potential of tidal currents in channels. Proceedings of the Royal Society a-Mathematical Physical and Engineering Sciences. 2005;461:2563-72.