Reef Restoration Using Seawater Electrolysis in Jamaica

Thomas J. Goreau and Wolf Hilbertz

NOTE: This paper, the first to describe the results of eight years of work on reef restoration using mineral accretion, was presented at the 8th International Coral Reef Symposium in Panama in July 1996.


Coral reefs have been restored on a pilot scale by transplanting coral fragments onto artificial reefs made of limestone and magnesium minerals precipitated onto iron and steel structures by seawater electrolysis. Corals have grown at rates comparable to the highest values reported in the field, even in areas of poor water quality. A complete assemblage of coral reef invertebrates and vertebrates have settled onto or migrated to these reefs. Because this novel technology creates the only growing, self-repairing artificial reefs that promote enhanced growth of corals and sand-producing calcareous algae while suppressing weedy fleshy algae and maintaining a complex coral reef community, it is uniquely suited for restoring physically damaged or eutrophication-affected coral reefs in order to protect shorelines, maintain biodiversity, and adapt to rising sea level. 


Jamaican coral reefs are undergoing accelerating deterioration wherever human activity physically disturbs reefs, degrades water quality, or over harvests key species (Goreau 1992). Many reefs no longer function as vital ecosystems: the coral-dominated wave-resistant upward-growing structures are turning into benthic ecosystems with a minor component of isolated corals. These ecosystems are coral communities rather than coral reefs because biodiversity is severely degraded and the reef structure, being bio-eroded faster than it grows, is less able to protect shorelines, keep up with rising sea-level, or provide shelter and food for the many other organisms which live between corals. Degraded reefs have fleshy algae dominant over calcareous algae, and can no longer provide beach sand to replenish that lost to erosion after damaged reef crests allow increased wave energy to reach the shore. 

Several types of "artificial reefs" have been built as wave-resistant barriers and hiding places for fish (Goodwin and Cambers 1983). They have a poor record because structures built from steel. poured concrete, stone, concrete blocks, gabions (wire baskets containing rocks), sand bags. sunken ships, wrecked airplanes, or old automobiles, unavoidably rust, corrode, and are broken by waves. Their fate is ultimate destruction by storms, requiring expensive and inevitably futile replacement. They turn into dangerous projectiles in hurricanes. After hurricane Andrew hit southern Florida, a survey of 'artificial reefs' found that all had moved. Some had one or several fragments found, but many vanished entirely. Although fish will hide behind any underwater obstacle, hard corals will not colonize them for a very long time, if ever, and they are mainly settled by soft corals and sponges rather than reef building corals. 'Artificial reefs' made of automobiles (a popular excuse for creating marine junkyards) rust and break apart before corals will settle on them (Goodwin and Cambers 1983). The failure of exotic materials to instigate natural hard coral reefs is caused by unsuitable surface chemistry and leaching of toxic hydrocarbons and metals from engines, paint, plastic fillers, concrete, and steel. 

A novel technology, developed by architect W. Hilbertz in the 1970s, uses electrolysis of seawater to precipitate calcium and magnesium minerals to 'grow' a crystalline coating over artificial structures to make construction materials (Hilbertz 1975). The mineral accretion, largely aragonite (CaCO3) 3nd brucite (Mg(OH)2), is very similar in chemistry and physical properties to reef limestone (Hilbertz 1992), which are primarily the remains of the aragonite skeletons of corals and green calcareous algae. This paper describes the results of work done in Jamaica since 1988 building and growing mineral accretion artificial reefs for enhanced coral growth and reef restoration. 


Electrolysis of seawater results in mineral deposition at the cathode. The physical properties of the material depend on mineralogy and crystal size, functions of deposition rate and electrical current parameters. Higher current densities result in faster growth but weaker material dominated by brucite, while lower current densities produces slower deposition dominated by harder aragonite (Hilbertz, 1992). Mineral accretion materials have a mechanical strength comparable to, and often greater than, concrete (Hilbertz, 1979). 

Deposition of minerals results from alkaline conditions created at the cathode by the reduction reaction: 

2H2O + 2e- = H2 + 2OH- 

which precipitates calcium and magnesium minerals from seawater: 

OH- + HCO3- + Ca++ = CaCO3 + H2O 

2OH- + Mg++ = Mg(OH)2 

In contrast the anode becomes acidic due to: 

2H2O = 4H+ + O2 + 4e- 

and highly oxidizing conditions result in: 

2Cl- = Cl2 + 2e- 

The sum of the net reactions at both electrodes should be neutral with regard to hydrogen ion production, and hence with regard to CO2 generation through acid-base equilibriums and carbonic acid hydrolysis: 

2HCO3- = CO3-- + CO2 + H2O 

Cathodes and anodes can be made in any size and shape, with current flow dependent on their spacing and surface area. Typically the cathode is built out of expanded steel mesh constructed as simple geometric forms such as cylinders, sheets, triangular prisms, or pyramids, but we have also molded complex forms using square mesh or chicken-wire mesh. An experimental reverse catenary was even built supported by floating spheres, i.e. a buoyant metal chain structure fixed to a cathodic base plate. This structure was initially flexible, and became rigid with progressing mineral accretion. Other new applications include molding shapes out of powdered sand or other materials, containing a cathode to enhance cementation by mineral accretion. 

Pilot artificial reefs have ranged up to 3 metres high and 10 metres across, but there is no theoretical limit on their size, providing sufficient current density is applied. Anodes are typically much smaller than cathodes, and shaped as sheets, rods, or mesh, depending on the materials used. Cathode materials are entirely protected from rusting by reducing conditions, whereas anodes are subject to rapid oxidation unless resistant material is used. We have used a wide variety of anode materials, including lead, graphite, and steel, but had best results with specially coated titanium. 

Although any direct current source will work, our preference is to use solar and wind generated power rather than alternating current generated from renewable fossil fuels which pollute the atmosphere with CO2 and acid rain. Current is applied across the terminals from a variety of power sources. We have empirically found it best to use lower voltages and higher currents. We have used transformers and battery chargers at both 12 volts and 6 volts, photovoltaic panels in a direct charge mode at a range of voltages between 3.8 and 17 volts, and have plans to use windmill generated current as well. Electricity consumption of each structure is equivalent to a single light bulb. These current levels are entirely safe to swimmers and divers, and it is possible to feel only a slight tingle when one directly short circuits the current by touching both anode and cathode simultaneously with bare hands. Current is transmitted by insulated copper cables, mono- or multi-strand types. Anode cable connectors are protected by clear silicone to detect the green color formed if salt water corrodes the electrical contact. 

Small pieces of corals were transplanted onto the structures, and attached with plastic ties, iron wire, or monofilament line, or simply allowed to sit on them. These corals largely consisted of fragments which had been naturally broken by storms, damaged by anchors, divers, or spear-fishermen, or corals whose bases were so bio-eroded that they would be broken by storms, as well as small pieces of branching corals from nearby "control" colonies. Most species of Caribbean corals have been tried. Only corals were attached, but these included some epifaunal sponges, calcareous algae, and other organisms on their undersides. 

Artificial reef structures have been built in depths ranging from 0.5 metres to 7 metres, in locations ranging from extremely protected back reef sites, open sites on the leeward western end of the island, to open exposed shores fully exposed to the direct impact of winter northers. One structure, located in a depth of 1.5 metres, continues to work despite being exposed to breaking waves which can reach up to 7 metres high. They have been built on sea-grass beds, limestone hard ground, white sand, and mud bottoms. We also built control structures receiving different current levels or no current at all, and structures which were allowed to accrete for a period of time and were then turned off. In addition we have connected corals growing in-situ directly to current sources via wires leading to artificial reefs. 


Crystal growth and hydrogen gas bubbling began as soon as rust on the steel had been reduced to iron. The surface changes from red to black to grey, and then white as minerals grow on it. Minerals have accreted to a thickness of up to 20 centimetres over three years. Iron and steel remain bright and shiny as long as sufficient electrical current flows to maintain cathodic protection. They are protected from corrosion by overlying mineral layers after current is turned off, unless this coating is broken. Structures on limestone hard ground become solidly cemented onto it, while those on sand and mud remain loosely attached and are vulnerable to being toppled in severe storms. 

In almost all cases transplanted corals healed quickly and were cemented solidly onto the mineral accretion within weeks. They showed bright healthy tissue pigmentation and prolific polyp feeding tentacle extension. However, some Acropora cervicornis have been killed by bristle worm (Hermodice carunculata) and gastropod (Coralliophila) attack, and some broken by severe storm waves. Transplanted corals grew skeletons at rates comparable to the highest values measured in the field (Gladfelter et al, 1978), even though all sites had sub-optimal water quality. Growth rates were determined by periodically measuring the diameter of colonies with a ruler or by measurements from sequential photographs or video images. Fragments of Porites porites grew from 5 cm to 30 cm across in two years. Acropora cervicornis branched prolifically, and grew by 5 to 8 cm in just 10 weeks. The tissue of corals attached to the structures via wires soon begins to grow over the mineral accretion. Such corals are visibly brighter than adjacent corals of similar species, but become less colourful when the current is turned for periods of up to 2 months, and then regain bright pigmentation within days when the current is restored. Young corals colonize and grow on the mineral accretion. We have found juvenile coral colonies up to 1 millimetre in diameter at densities of around 0.7 per square centimetre on three year old artificial reef substrate. One artificial reef has been colonized by around a hundred young Agaricia agaricites and Favia fragum in the last two years and these have grown to a size of several centimetres across, in a polluted lagoon where little or no natural recruitment is being observed. 

Except for transplanted corals and a few small organisms encrusting their bases, all other species found on the artifical reef have spontaneously settled on it or migrated to it. A highly diverse coral reef community (Table 1) established itself on the mineral accretion structures, including foraminifera, cyanobacteria, chlorophytes, rhodophytes, phaeophytes, porlfera, bryozoans, cerianthids, coralliomorpharia, gorgonaceans, sabellid, serpulid, and nereid polychaetes, oysters, gastropods, octopods, squids, echinoids, holothurians, ophiuroids, crinoids, cleaning shrimp, crabs, hermit crabs, and spiny lobsters. A large variety of adult and juvenile fish have been permanent or temporary residents, including morays, trumpetfish, squirrelfish, seabass, fairy basslets, cardinalfish, grunts, drums, butterflyfish, angelfish, damselfish, wrasses, parrotfish, blennies, gobies, surgeonfish, filefish, and porcupinefish. The geometry of the structure appears to strongly affect the type of species recruited. Dolphins have been observed swimming near the structures. No organism has been observed to show aversive behavior. 

The main difference between our artificial reefs and nearby natural reefs is the preponderance of fleshy algae which are overgrowing corals on nearby reefs while the artificial reefs have balanced coral and algal growth and the algae are predominantly sand-producing calcareous reds and greens with a much lower density of weedy algae than adjacent natural reefs. Large masses of calcareous Jania, Amphiroa, and Halimeda grow on the sides of the structures, generating sand. Mineral accretion structures whose power is turned off have subsequently had their calcareous algae and corals overgrown by fleshy algae. In sharp contrast to electrified structures, the control structures which received no current rusted and fell apart within months, and the crumbling fragments were not colonized by corals or other organisms. 


Rapid coral growth and recruitment even in areas of known poor water quality (Goreau 1992) show that our method is able to partly counteract coral reef eutrophication due to coastal zone nutrient fertilization, and so can contribute to restoring damaged reefs and creating new ones in even degraded areas. As the structures become stronger with age, they are also able to contribute more and more to shore protection from waves, and to keep pace with rising sea level. Unlike "artificial reefs" made of exotic materials, our reefs get constantly stronger with time. As long as current is applied, they are self-repairing, since any cracks and breaks of mineral accretion are rapidly and preferentially filled in by new material. While some structures have been damaged by storm waves or impacting objects, such damage is easily repaired

The stimulation of calcareous organisms of all types on the artificial reef, and the relative paucity of non-calcifying organisms, is probably largely due to the boost the former receive from locally alkaline conditions, which allow them to grow their skeletons at lower energetic cost because they do not have to use metabolic energy to pump protons away from calcification sites to maintain internal pH homeostasis (Goreau 1977). The bright colors of the colonies and their high degree of tentacle expansion may be due to the extra biochemical energy freed as a result. A possible alternative explanation could be due to the high density of electrons on the cathode, some of which may be trapped and used to generate ATP, but if this were the major factor, non-calcareous organisms would also be stimulated. The general stimulation of growth of marine organisms by both processes on mineral accretion substrates is covered under pending US patent #08/374993, issued to W. Hilbertz and T. Goreau (1995). 

The view that mineral accretion is the most suitable substitute substrate for coral recruitment compared to natural limestone is supported by marine archaeology. Where only iron metal is found in shipwrecks, it rusts away and is not colonized by hard corals unless first covered by encrusting calcareous red algae. Where several dissimilar metals are found in wrecks as well, such as brass, bronze, copper, magnesium, or aluminum, the differing electromotive potentials of the metals results in electrolytic current flows which cause deposition of mineral accretion over the cathodic metals until the anodes are consumed, ending the reactions. Natural electrolysis is responsible for preservation of most metal artifacts in shipwrecks dating as far back as the Bronze Age, which are found under thick concretions of limestone minerals. We have observed old iron anchors and chains completely covered with hard mineral accretion, allowing corals to settle and grow prolifically on them. This would probably not have happened without electrolytic mineral coatings and concurrent cathodic protection. 

We believe that apart from protection of living reefs, mineral accretion is the best substitute for enhancing coral growth and restoring natural coral reef ecosystems even under stressed conditions. Since the method is able to rely entirely on non-polluting and renewable energy, it is suitable for remote areas. Laboratory experiments showed that 1.07 kilograms of mineral accretion was precipitated per kilowatt-hour of electricity. At Jamaican residential customer rates for imported fossil fuel generated electricity (US$ 0.10 per kilowatt-hour), resulting materials are nearly an order of magnitude cheaper then the equivalent weight in concrete blocks. 

Typical costs for sea-walls and breakwaters using conventional techniques run around US$ 8,000 per metre, the amount it cost the Maldives to replace mined-out reefs with stacked pre-cast concrete tetrapod breakwater structures to protect the shore from erosion and the aquifers from salt water intrusion. Unlike concrete blocks, mineral accretion structures can be built in any size and shape, contain internal steel reinforcement, and get stronger with age rather than weaker. Submerged sea-walls could therefore be built which would eventually become much stronger than concrete structures, at a fraction of the cost. 

We expect that mineral accretion technology will eventually become the preferred form of reef restoration and shore protection where reefs have been degraded due to anthropogenic or natural causes, especially if sea level continues to rise more rapidly than coral reefs grow upward. Over the last two years the global average sea level rise measured by the TOPEX/Poseidon radar satellite has been 2 millimetres per year, as fast as most healthy reef structures are accumulating, but faster than degraded reefs or bleached corals can grow (Goreau and Macfarlane 1990). 


We are grateful for support from the European Union for the latest phase of artificial reef construction under a grant to the Negril Coral Reef Preservation Society for establishment of the Negril Marine Park. We thank Ursula Hilbertz-Rommerskirchen. Maya Goreau, Bill Wilson, Katy Thacker, Karen McCarthy, Martin Brinn, and Bert Bentley for assistance during artificial reef construction and deployment. This work would not have been possible without permission from Richard Murray and David Cunninghame to use Tensing Pen property to house the power supply and their support for electricity bills. Anodes were donated by Heraeus Elektrochemie G.m.b.h., Germany. 


Gladfelter E, Monahan R, Gladfelter, W (1978) Growth rates of five reef-building corals in the northeastern Caribbean. Bulletin Marine Science 28: 728-734 

Goodwin MR, Cambers G (1983) Artificial reefs: a handbook for the Eastern Caribbean. Caribbean Conservation Association. Barbados 

Goreau TJ (1977) Coral skeletal chemistry: physiological and environmental regulation of stable isotopes and trace metals in Montastrea annularis. Proceedings Royal Society London B 193: 291-315 

Goreau TJ (1992) B1eaching and reef community change in Jamaica. American

Zoologist 32: 683-695 

Goreau TJ, Macfarlane AH (1990) Reduced growth rate of Montastrea annularis following the 1987-1988 coral bleaching event. Coral Reefs 8: 211-215 

Hilbertz W (1975) Towards self-growing structures. Industrialization Forum 6: 53-56 

Hilbertz W (1973) Electrodeposition of minerals in seawater: Experiments and applications. Oceanic Engineering 4:94-113 

Hilbertz W (1992) Solar-generated building material from seawater as a sink for carbon. Ambio 21: 126-129 

Hilbertz W, Goreau TJ (1995) A method for enhancing the growth of aquatic organisms and structures created thereby. US Patent #08/374993  


These are species which were regularly found on the artificial reefs, including corals which had been transplanted, attached organisms which had spontaneously settled and grown on them, and mobile organisms which were repeatedly found on the structures and appeared to have taken up permanent residence on them. Many other organisms were also seen on the structures on once or twice, but these are not listed as they may have only have been temporarily passing by.  


             Stephanocoenia michelinii, Madracis decactis, Madracis mirabilis, Acropora palmata, Acropora cervicornis, Agaricia agaricites, Agaricia tenuifolia, Agaricia fragilis, Helioseris cucullata, Siderastrea radians, Siderastrea siderea, Porites porites, Porites furcata, Porites astreoides, Porites branneri, Favia fragum, Diploria clivosa, Diploria strigosa, Diploria labyrinthiformis, Manicina areolata, Colpophyllia natans, Montastrea annularis, Montastrea cavernosa, Meandrina meandrites, Dichocoenia stokesii, Dendrogyra cylindrus, Mussa angulosa, Isophyllia sinuosa, Isophyllastrea rigida, Mycetophyllia ferox, Mycetophyllia lamarckiana, Eusmilia fastigiata 


porifera:             Iotrochota birotulata

hydrozoans:       Millepora alcicornis, Stylaster roseus

cerianthids: Ceriantharia sp.

actinaria: Lebrunia danae

coralliomorpharia: Ricordea florida

            gorgonaceans:   Briareum abestinum

            sabellid:            Sabellastarte magnifica

            serpulid:            Spirobranchus giganteus

            nereid:  Hermodice carunculata

            terebellid:          Eupolymnia nebulosa

            oysters:            Pteria colymbus, Isognomon sp.

            gastropods:       Coralliophila sp.

            octopods:          Octopus briareus

            squid:   Sepioteuthis sepiodea

            echinoida:         Diadema antillarum, Eucidaris tribuloides

            holothuria:         Holothuria mexicana

            ophiuroida:        various species

            crinoida:            Nemaster rubiginosa

            shrimp: Stenopus hispidus

            isopods:            Anilocra sp.

            crabs:   Percnon gibbesi, Mithrax sp.

            hermit crabs:     various species

            lobster: Panulirus argus 


cyanobacteria: Schizothrix sp 

PROTOZOAforaminifera: Homotrema rubrum, Gypsina sp. 


chlorophytes: Halimeda opuntia, Caulerpa racemosa

rhodophytes: Jania rubens, Galaxaura oblongata

phaeophvtes: Lobophora variegata 


morays: Gymnothorax funebris, Gymnothorax moringa, Muraena miliaris

trumpetfish: Aulostomus maculatus

squirrelfish: Holocentruus rufus, Myriprlstes jacobus

seabass: Epinephelus cruentatus

fairy basslets: Gramma loreto

cardinalfish: Apogon maculatus

grunts: Haemulon flavolineatum

drums: Equetus punctatus

butterflyfish: Chaetodon ocellatus, Chaetodon striatus

angelfish:          Holacanthus tricolor, Pomacanthus paru

            damselfish:        Eupomacentrus partitus, Eupomacentrus dorsopunicans, Eupomacentrus leucostictus, Abudefduf saxatalis

            wrasses:           Thalassoma bifasciatum, Halichoeres maculipinna

parrotfish: various species

blennies: various species

gobies: various species

surgeonfish: Acanthurus coelerus, Acanthurus chirurgus, Acanthurus bahiana

filefish: Cantherhines macroceros

porcupinefish: Diodon holocanthus



Figure 1. Top of artificial reef after two years of mineral accretion growth. This structure reaches within 0.5 metres of the surface. Photograph by Dr. Peter D. Goreau.


Figure 2. Artificial reef seen from above. This structure is 8 feet (2.2 metres) tall. The prolifically branching Porites porites facing the camera grew from 10 cm to 30 cm in diameter in two years following transplantation onto the structure. The sides of the structure are largely covered with the red calcareous alga Jania. The base of the structure is solidly cemented to the limestone bottom by mineral accretion. The three hemi-cylindrical ballast chambers around the base were filled with rocks to stabilize the structure before mineral accretion attached it to the bottom. The chamber at the right contained an anode and mineral accretion is seen to be much greater than on the chamber at left due to the higher electrical current density. Photograph by Dr. Peter D. Goreau.


Figure 3. Close-up of Porites porites growing on the artificial reef, showing prolific polyp extension. Photograph by Dr. Peter D. Goreau.


Figure 4. Close-up of Porites astreoides growing on the artificial reef. Protograph by Dr. Peter D. Goreau.


Figure 5. Close-up of Diploria strigosa growing on the artificial reef. Photograph by Dr. Peter D. Goreau

Figure 6. Close-up of the artificial reef, showing young colonies of Acropora palmata and Porites divaricata which spontaneously settled and grew on the structure. No Acropora palmata was observed growing in the surrounding back-reef. Photograph bv Dr. Peter D. Goreau.