Dossier Océan et énergie - Énergie Thermique des Mers

Sommaire IOA News Letters

Evaluation of a High Potential Ocean Thermal Energy Conversion (OTEC) site in Puerto Rico

By
P.FRIEND, E. BYFORD, S.CAMP, M.L. KERNICK, A. LUBBEN,
J. MILLS, B. POTTER, M. REYNOLDS, C.STEDMON.

Dept. of Oceanography, University of Southampton, Southampton Oceanography Centre,
European Way, Southampton SO14 3 ZH, U.K.

INTRODUCTION

Despite sporadic but continuing interest in OTEC for over a century (e.g. Carnot 1824), there is only one experimental plant in existence in Hawaii.  The reasons lie in the continuing low price of oil; unwillingness to invest in what, to many, is an unproved energy source; and design considerations regarding the Cold Water Pipe (CWP), Heat Exchanger (HX), and turbine(s).  With a doubling of the world population expected by about 2020, and a permanent decline in global oil reserves expected at about the same time (U.S. Geological Survey), as well as the recent acceptance of climate change by the I.P.C.C. (Houghton et al. 1995), there is a need to prove the viability and applicability of OTEC systems larger than the experimental plant in Hawaii.

By identifying the criteria necessary to minimise the effects of OTEC on the environment, and by considering the short distance to the cold water resource and the financial benefits made possible through the use of Deep Ocean Water Applications (DOWA), as well as predicted advances in engineering and technical considerations, we have evaluated the Punta Tuna region of south-east Puerto Rico (Fig. 1) as a potential OTEC site.  An early resource assessment (Atwood et al. 1977) concluded that the region would be suitable for the siting of nearshore OTEC plants, but did not mention the possibility of an onshore plant.  In the light of our research, we conclude that a closed-cycle, land-based plant, with a generating capacity of up to 10 MW, would have a negligible environmental physical impact and a minimal. site-specific biological impact.   The higher costs of generating electricity from OTEC compared to generation from fossil fuels would be offset by the inclusion of DOWA in the form of freshwater production, air-conditioning, agriculture and aquaculture, and tourist attractions.   Throughout the preparation of this report, we have referred to the work of a number of authors listed in the references. (please see the references we enclosed with the Newsletter)

GEOGRAPHY AND GEOLOGY

Puerto Rico, situated in the Antilles island-arc system of the Caribbean, is a self-governing member of the commonwealth of the U.S.A.  It consists of three islands, Vieques, Culebra and Puerto Rico, covering an area of 9,104 km².  In 1990 the population was estimated at about 3.5 million.  The island is urbanized, with employment based in labor intensive industries and tourism.  There are no commercial fossil-fuel deposits, requiring the importation of all petroleum necessary for the production of 9/10 of the island's electricity, the remainder being generated by hydroelectric plants.  Current total generating capacity is 4.23 MW.  A toll road, completed from San Juan to Ponce in 1982, opened up Puerto Rico's south coast,   previously isolated by the Cordillera mountains.  Yabucoa, just inland from Punta Tuna, marks the beginning of a newly developed industrial belt continuing south westwards.  The industries situated on the south east include oil refineries, pharmaceuticals, textile manufacturers and industrial chemical plants.

The headland of Punta Tuna comprises alternating limestone and mudstone beds, dipping at 12-20¢X south, and forms one end of a bay into which the river Cano de Guayanes flows.   The river, dry for most of the year, is often subject to instant variations in volume during heavy rain when the sediment load increases rapidly.  Sediments on the shelf are dominated by carbonate sands.  The continental slope comprises unconsolidated material which may be subject to slumping following a seismic event, or the deposition of sediment further offshore during times of river flood. In 1995, 644 earthquakes were recorded with a Richter magnitude of over 4.6.  Most of these, however, were only felt in the northern part of the island, with Punta Tuna last experiencing a tremor in 1970.

ENGINEERING

This distance at Punta Tuna to the cold water resource is approvimately 2-3 km, giving an average calcualted slop angle of about 26¢X.  While offering the prospect of substantial savings in the cost of the CWP, the relatively steep gradient is an important parameter in the design and deployment of the CWP.  Although slumping is rare, its effects and those of the irregular slope morphology, need to be considered.  It is considered that a buoyant pipe anchored above the seabed by a series of fixtures mounted into the bed, or held in place via dead weights at intervals along the pipe trajectory, to which cables supported by floats are attached, is the best practicable design (Vilain, 1985).  According to Brewer, a 10 MW plant would require a CWP of Fibre Reinfored Plastic (FRP) with a diameter of 7.61 m and a wall thickness of 3.8-11 cm. Using slope angle and FRP density, it is estimated that the average shear stress acting on a 100m length of pipe would be approximately 103N.

The seaward extension of the CWP is within the limit designated by UNCLOS (1982) as "territorial seas" (12 nautical miles), and therefore permission from the coastal state is required for the emplacement of any seabed structure.  An OTEC plant is likely to have to provide the necessary equipment to deal with any pollution and must be able to satisfy the Puerto Rican government that any pollution can be dealt with quickly and efficiently.

IMPACT ASSESSMENT

Physical

In the prediction of the physical effects of OTEC operations, much work has been carried out using modeling techniques based on moored plants in the Gulf of Mexico (e.g. Martin and Roberts 1977).  These studies indicated  that, for 100 x 200 MW moored plants after 30 years operation, there would be a decrease in surface temperature of about 0.05¢XC, and a gain of
0.8¢XC.

For a moored 100 MW plant with a combined cold and warm water intake of 300-500 m³/s, Pritchard et al. (1978) predicted the formation of a radial inflow, in a pattern dependent on surfact currents.  The resulting horizontal pressure field would support a geostrophically balanced, cyclonic flow characteristic of a cold core eddy.  First order differential models for 100MW plant drawing water from a 100 m mixed layer indicate that the eddy will have a radium of 10-20km, and a maximum tangential velocity of approximately 15 cm/s at 1.5 km from the sink.  The thermocline would be raised about 15m.

The effects of decreased gas solubility, especially in 'open' systems, have been studied to establish whether a net outgassing of CO2 may occur (Ditmars and Paddock 1979).  It was estimated that an OTEC plant would release 1/3 to 1/4 of the amount of the CO2 released by a fossil-fuel plant of the same capacity.  The Martin and Roberts temperature model does not, however, allow for the addition or removal of heat by surface and deep water currents flowing into the Gulf of Mexico, while the Pritchard current model assumes the existence of an infinitely deep and motionless bottom layer.  Results based on a study of the Hawaiian OTEC plant (Greene and Guenther 1990) indicate that CO2 emission rate is 38.5g CO2 kW/hr, while the amount for equivalent fossil-fueled plants is 531 g CO2 kW/hr (natural gas) and 934g CO2 kw/hr (coal).  The results indicate that CO2 release is 1/12 and 1/24 of that from an equivalent sized gas or coal plant.

No changes in sea temperature or interaction with island coastal circulation have been observed as a result of 3 years operation at the Hawaiian plant (Dr. Tom Daniel - Pers. Comm).  In order to minimise any physical effects arising from the estimated combined intake for a 10 MW plant of approximately 120 m³/s, short flushing times are considered to be important.  As a cold water source, a plant at Punta Tuna would use the N. Atlantic deepwater of the Virgin Island basin, and as a warm water source part of the 15 Sverdrup current flowing past the Lesser Antilles.  Circulation patterns between 0-100 m depth show a general south-west to westerly drift of about 10-40 cm/s, below this the sub-surface currents are influenced by the steep bottom topography.  Some flow reversals and large scale eddies exist in the 100-300 m layer, but generally there is a south-west trend, with 100-300 m layer, but generally there is a south-west trend, with currents running parallel to the bottom isobaths, at speeds of 5-25 cm/s (Stoddard et al. 1985).  Flow reversals have also been suggested by current meters at depths of 400-600 m, although these effects have not been quantified.  Based on a cold water intake of 4 m³/s per MW, the vertical heat flux in the CWP is equivalent to the natural vertical heat flux in an area of about 1300 km², a significant figure because the Virgin Islands basin is approximately 600 km².  However, the estimated basin flushing time, based on average bottom current velocities, is between 2 to 8 days.

Biological

Depending on the screen mesh size, organisms in the intake vicinity would be entrained or impinged, while those passing through the plant would be liable to temperature and pressure changes.  The survival rate on discharge of these organisms is not known, and more work is required in this area.  Local fisheries may be affected by removal of eggs and larvae from the water column, and by the entrapment of fish when inflow velocity is greater than their avoidance capabilities.

The methods available for the limiting of biofouling in the pipes may have detrimental effects. Manual brushing  can kill impinged organisms, the use of biocides such as chlorine may lead to toxin production through reaction with bromide,and subsequent bio-concentration in higher trophic levels.  Where the sue of ultra-violet irradiation stuns rather than kills an organism, there is concern that irreparable damage to DNA may occur (e.g. Myers et al., 1986).

There is a lack of data regarding the behavior of aluminium alloys and stainless steel used in the system components.  There is concern that pitting and crevice corrosion may lead to working fluid leakage; a concentration of 14 µ g/g of NH³ is lethal to many species (Duedall et al., 1985).

The operation of the seawater pumps produces noise in the same frequency range (0.5Hz-1.0kHz) which has been shown to affect mammals and fish (Rucker and friedl, 1985). As a result of pumping deep and surface waters, a transfer of nutrients and chemical species to new locations may occur, enhancing phytoplankton productivity and leading to algal blooms and eutrophication.  An important consideration will be monitoring for the development of toxic blooms that would seriously affect fisheries and pose a threat to human health.  Coral reefs that thrive in low nutrient conditions are presently not protected in the Punta Tuna area, although a park has been proposed.

The south-east waters of Puerto Rico contain a wide variety of fish including parrotfish, grouper, snapper, mullet, tuna, mackerel, barracuda, dolphin, moray eels and sting rays.  These fish are susceptible to temperature changes, but it is felt that the short flushing times mentioned above will severely reduce the biological impact of an OTEC plant on these and other organisms.  That entrainment and impingement will inevitably lead to loss biota cannot be denied, and research would be needed to quantify the effect, with eventual replacement possible via the DOWA aquaculture unit.  More research is needed into the effects of noise and acceptable methods of reducing bioaccumulation in the pipes and heat exchanger.

FINANCE

The main parameters that require consideration for the installation and running of a commercial OTEC plant are the costs of start-up and working capital, labor and material, as well as taxation levels, current and predicted costs of electricity generation by different methods, and the rate of return on investment.  It is envisaged that the high capital costs, arising from expensive system components such as the pumps, turbines, HX, and CWP, will gradually fall with ongoing research and development.  The cost of the seawater system (CWP and pumps) is expected to fall by around 30%, and that of the heat exchanger by about 50% over the period 1990-2000 (Vega 1994).  A rough estimate of the capital costs for a 10 MW open cycle land based plant was made by Downs (1991), at about US$70 million (1989 US$).  Add to this a saving of around 10% (lower than expected) through modification due to research and development, as mentioned above, and allow for inflation (average of 3% per year) gives a total of about US$ 77 million (US$ 1996).

It is likely that funding for any proposal would have to be supplied from a number of public and private investors.  Past projects have been supported by both governments and industrial consortia.  To date, investment in Puerto Rico has been dominated by the U.S.A., but there is some evidence that the government is looking for diversification and would therefore welcome foreign investment.

A report published in  August 1996 by Calero Consultants Inc. indicates that it is essential for Puerto Rico to privatize part of its electricity generation in order to improve its running and efficiency.  The report states that the country proposes to develop multimillion dollar projects requiring non-local investment in order to increase its attractiveness to other investors. This gives an indication of how the Purerto Rican government would react to an OTEC project.  It is perceived that funding would facilitated by Puerto Rico's close ties with the U.S.A., and that there would be ready access to New York credit markets.  In addition, the Puerto Rican government has funding available through the Economic Development Corporation (E.D.C.).

While future electricity generation in Puerto Rico could supplied by other methods, it is the economic benefits from DOWA that make the establishment of an OTEC plant in Puerto Rico an attractive proposition.  Studies indicate that in areas where air conditioning is in use for at least 70% of the year, an OTEC plant can run profitably through the sale of its air conditioning facility alone, and not by the sale of DOWA or electricity (Van Ryzin et al. 1992).

Through tourism, an interesting extension of DOWA is possible, namely the opening of the plant to the public as a tourist attraction, with the inclusion of an aquarium.   Studies on recently established aquaria show a high degree of profitability. For example, an aquarium in Newport, U.S.A., (pop.9500) made a profit of US$8 million in its first year (brown, 1994).  Although no public aquaria currently exist in combination with an OTEC system, this would be an interesting avenue for further research and investment.

Puerto Rico has one of the highest Gross National Products (GNP) in the Caribbean, of which only 3% comes from fisheries, agriculture and forestry, the main component being derived from the manufacturing, trading and financial services sectors.  Due to close trade links and proximity with the U.S., a large established market exists, offering the opportunity to further increase GNP by the export of  DOWA products.   Seven-eights of Puerto Rico's exports are currently to the U.S., the remainder are to the Caribbean.

The political stability of a country is seen as one of the criteria necessary for the attraction of foreign investment.  In 1952 Puerto Rico became a member of the U.S. Commonwealth, possessing its own constitution and represented in the U.S. Congress by an non-voting elected delegate.  There has been some terrorist activity by the F.A.L.N. (Fuerzas Armadas de Liberacion nacional) in the late 1970's and early 1980's, however the majority of the population indicated their preference for either statehood or commonwealth status in 1993.

DOWA

Fresh water

The average annual rainfulll in Puerto Rico varies from 0.9m in the south to 1.5m in the north, creating a freshwater imbalance over the island.  While annual natural freshwater production is about 5 x 1011 1, southern rivers remain dry for most of the year. There are two main areas of concern regarding the water supply.  Most of the rivers are infected with the bacteria 'chisto', which can cause severe damage to internal organs, and reservoirs are rapidly filling with sand, reducing storage population increase and recent droughts, there exists the potential for long-term water shortage.  In 1995, rationing was imposed throughout half of the island due to insufficient stocks.

The development of freshwater aquifers serving the population on the North coast is being limited by saline water intrusion.  In the Cabo Rojo area of south-west Puerto Rico, increased water consumption, resulting from increasing domestic and international tourist trade, requires the development of additional sources of freshwater.  In 1993, water consumption in Puerto Rico was 3.54 x 106 m3 (U.S. Geological Survey, 1996).   Irrigation using OTEC-produced freshwater is seen as a method of enhancing agricultural productivity, limited in Puerto Rico by the predominantly karstic landscape.   The Common heritage Corporation believes that more than 100 commercial   species may be cultivated.

Considering a 1 MW plant, fitted with a second stage fresh water production unit, Takahashi and Trenka (1996) estimate that annual freshwater prodcution would be approximately 1.7 x 106 m3, sufficient for a population of 20,000.  The cost of fresh water production by reverse osmosis and flash desalination is between $1.30 and $2.00 per cubic meter for a plant with annual output of about 1.5 x 106 m3.  At these prices, a 1 MW plant would produce about $3 million worth of desalinated water per year.  If these estimates are correct, it is likely that a 10 MW plant would produce around ten times this amount.  However, the only data point is the 210 kW plant in Hawaii which uses an add-on surface condenser producing about 104 m3/yr.  For a closed-cycle system, running the condenser output water through an external condenser, A. Johnson calculates that 1 m3 of freshwater per 200 m3 deep seawater could be produced by condensation from tropical air.  The potential exists for the production of 6.23 x 106 m3 freshwater per year for a 10 MW plant with a 40 m3/s intake.

Air-Conditioning and Agriculture

With an average annual temperature in Puerto Rico of 30¢XC, a year-round demand for air-conditioning exists.  Through increased tourism, this demand is expected to rise.   In 1994 there were 4 million visitors to Puerto Rico, with 7154 hotel rooms available and an average occupancy rate of 70%.  In 1996, 20 new hotels have been under construction.  By circulating the OTEC cold water supply through space heat exchangers or cooling a working fluid for use in heat exchangers, it is predicted that a 300 room hotel can be air-conditioned at less than 25% of the cost of the electricity required to operate a conventional system, and that the return period on capital investment for the installation of such a system would be four years or less (Takahashi & Trenka, 1996).  The same system could be adapted to include cold storage facilities.

Air-conditioning may be used for the control of greenhouses and growing room habitats, allowing production of non-indigenous crops such as strawberries, asparagus and artichokes.  By cooling plant root temperatures through the use of cold water pipes embedded at depth, strawberries have been grown with a sugar content approximately 5 times greater than that using conventional techniques.  By the control of air and soil surface temperatures to provide an optimum temperature difference between root and soil, it has been possible to enhance the transport of nitrates and phosphates in carrots.   The control of temperatures by manipulation of water flow rates allows asparagus to be grown through three growth cycles in less than 9 months, reducing the planting to harvesting by more than one year.

Aquaculture

The high solar radiation in OTEC regions, and the fact that deep seawater is almost pathogen free, and has a high trace metals and nutrients content, causes increased algal growth and carotene production.  In Hawaii, the Cyanotech Corporation produces more than one ton per month of micro-algae (e.g. Spirulina and Dunaliella spp.) for human consumption.  These same species are used in the pharmaceutical industry, while phycobiliproteins and fluorescent pigments are used in immunological diagnostics.   The naturally occurring pink pigment derived from microalgae, astaxanthin, is used in fish feed as a colouring agent for salmon and shrimps.  the world market is worth $125 million per year, with pigments costing around $2,500 per kg. In addition, the sue of deep seawater has produced high quality zooplankton and phytoplankton for use as fish feed.

The cultivation of kelp, which in turn is used as a food source for the Californian red abalone and sea urchins is possible using deep seawater at about 15¢XC.  The abalone is a high value luxury food on the U.S. market.  Cultivation techniques similar to those used for kelp are sed to produce edible seaweeds such as makonbu, wakame and nori, favored as delicacies in the Middle East.  The control of water temperatures by mixing the warm and cold exhausts permits the cultivation of finfish (e.g. salmon, trout, flounder) and shellfish (e.g. abalone, shrimps, clams, mussels, oysters, lobsters) for home consumption or export.  Tropical fish may be cultivated for sale to aquaria and shrimps can be provided to re-stock shrimperies destroyed by viruses.

CONCLUSIONS

We conclude taht the site of Punta Tuna, Puerto Rico satisfies the criteria necessary for the installation and operation of an OTEC plant with up to 10 MW capacity.  With the predicted fall in the cost of system components, and the advantages that DOWA would bring in increased profitability and research, we believe that proposals for funding should be actively sought.  OTEC would provide a renewable energy source sufficient to supply the increasing demand for generating capacity, and the attraction of foreign investment and the sale of DOWA products would promote Puerto Rico's economic competitiveness.

We would like to thank Prof. Steve Thorpe (Southampton Oceanography Centre) for help with the preparation of this paper, Dr. Tom Daniel (Natural Energy laboratory of Hawaii) for his valuable information, and Kate Davies for the graphics.