Paul R Epstein, Timothy E Ford, Rita R Colwell *
Waterborne infection, from marine or freshwater sources, is the leading cause of illness worldwide, and fish provide more animal protein for human consumption than poultry or meat. Today the health of large marine ecosystems--their diversity, productivity, and resilience--is threatened by a "global epidemic" of coastal algal blooms. This overgrowth of algae has profound implications for water and food safety because vibrios adhere to phytoplankton (algae) and zooplankton, and "red tides" bear biotoxins responsible for fish and shellfish poisoning. Furthermore nutrient-rich effluents stimulate algal growth and warmer sea surface temperatures shift marine ecosystems towards more toxic species.
Cholera becomes endemic when water and sanitation systems are not kept apart, but other environmental factors affect both the inoculum and persistence of this ancient pathogen. In 1991 cholera struck Peru from Chancay (Lima's port city) to the port of Chimbote 400 km north, the next day. Cholera soon surfaced all along the 2000 km Peruvian coast. It then spread rapidly to ports in Ecuador, Colombia, and Chile, and then to Brazil, Venezuela, and Bolivia, following rivers and streams. Over 15 months, more than half a million people fell ill and almost 5000 died in nineteen Latin American nations.
The Peruvian coast is prolific in phytoplankton and their chief consumers, anchovies. Human activities are enhancing the blooms. From March to August, 1991, Vibrio cholerae O1, biotype El Tor, serotype Inaba was isolated from marine plankton near Lima; V cholerae O1 has been recovered from the bilge of Latin American vessels docked in US Caribbean ports, and seems to have crossed the Pacific as a "stowaway" from Bangladesh.
Since 1960 researchers in Bangladesh have related the seasonality of cholera to coastal algal blooms, but the reservoir remained a mystery. Applying fluorescent antibody and polymerase chain reaction techniques, Colwell and colleagues have identified a viable, but non-culturable "quiescent" form of V cholerae, associated with a wide range of surface marine life. Under adverse conditions they contract 15-300 fold and reduce their metabolic rates, "hibernating" to tolerate shifts in pH, temperature, salinity, and nutrients. Under favourable conditions of nitrogen, phosphorus, and warming (also conducive to algal blooms), V cholerae reverts to a culturable and infectious state. Chitinases and mucinases facilitate attachment to aquatic organisms, while algal surface films and slimes enhance growth by creating turbulence-free havens. Prolonged survival of vibrios has been associated with cyano (blue-green) bacteria, silicated diatoms, drifting dinoflagellates, seaweed, large algae, water hyacinths, and duckweed but the key commensal and/or symbiotic association is with zooplankton. Up to a million bacteria have been detected on copepod (zooplankton) egg sacs.[6,7] V cholerae is also found in molluscs and in fish skin and intestines.
In biblical times "red tides" were seen as the blood of sparring whales or as menstrua. Today toxic phytoplankton blooms are associated with paralytic, diarrhoeal, and amnesic shellfish poisoning and with histamine (Scromboid), pufferfish, and ciguatera fish poisoning. The 1972 New England outbreak of red tide paralytic shellfish poisoning, a 1978 outbreak of ciguatera poisoning in Florida (there are now 200 000 cases annually worldwide), and a rash of reports in 1985 on diarrhoeal shellfish poisoning aroused concern, and in 1987 the International Oceanographic Commission (IOC) and UN Food and Agricultural Organization combined to set up an intergovernmental panel on harmful algae blooms.
Algal blooms (red, green, golden, brown, bioluminescent), covering vast expanses of marine, estuarine, and inland water have now been described from points as diverse as California, North Carolina, Guatemala, Iceland, Japan, Thailand, and the Tasman Sea. The global distribution of coastal blooms is visualised on a composite image from a satellite no longer collecting data (figure 1); the sea-viewing wide field sensor is scheduled for launching in 1994. This increase in blooms is a direct consequence of human activities, some local and some impacting on global climate systems. The following is just a sample of reports of the consequences for human health, tourism, and food production. In 1973 and 1974 blooms of brevetoxin-producing Gymnodinium breve blanketed Florida beaches; in 1976 G catenatum bloomed off the Spanish coast causing hundreds of cases of poisoning from contaminated mussels; Pyrodinium blooms off the Philippines in 1983 and 1987 caused 1127 cases and 34 deaths and also 26 deaths on the Pacific coast of Costa Rica and Guatemala in 1987. In the 1980s, diarrhoeal shellfish poisoning associated with dinoflagellates spread from Japan and the Americas to previously uncontaminated areas of Ireland, Portugal, Italy, India, Thailand.
An outbreak of waterborne disease may be thought of as an inevitable consequence of specific environmental changes amplifying plankton and associated bacterial proliferation. Sunlight, pH, currents, winds, and river runoffs govern the location and timing of plankton blooms. The major anthropogenic influences, global warming apart, are:
Excess nutrient from sewage and fertiliser effluents is a primary cause of marine eutrophication; soil erosion and acid rain add additional nitrogen and phosphorus. Freshwater lacks dissolved carbonates so an increase in atmospheric CO2 can "fertilise" algal growth on ponds and sewage lagoons. Other pollutants upset the balance of marine ecosystems. Toxins, such as polychlorinated biphenyls, heavy metals, and pesticides, accumulate in food chains, causing damage to marine organisms, altering the ecosystem's equilibrium. Oil slicks and solid plastics harm sea mammals and birds, altering predation pressures.
The over-harvesting of fish and shellfish reinforces algal growth also by reducing algivorous grazing. Nine of the world's seventeen major fisheries are in serious decline.
Loss of habitat
The wetlands ("nature's kidneys") filter nitrogen and phosphorus, store carbon, and support fish and seabirds. They are disappearing. Salt marshes, sea grasses, and mangroves are suffering from coastal urbanisation; and California has now lost most of its wetland area. Coral reefs (the "oceans' rainforests") that protect coasts and cradle marine life are being widely mined for road and housing construction. Warming directly harms reefs, causing the algal symbiont to sprout flagella and swim off, leaving bleached polyps behind.
Ship recordings ("sea truth") since 1850 demonstrate ocean warming, and Maskell and colleagues illustrated this in the first article in the series. Confirmation comes from Yan and colleagues' time-series studies in the Western Pacific in the mid-1980s (figure 2) but the pattern since then has not been consistent. Global warming probably has contributed to recent variations in sea surface temperature but it is not the only possible explanation. The climate phenomenon known as El Niño (see below) also raises sea temperatures. Figure 3 shows this for the Eastern Pacific.
Warming reduces dissolved oxygen and, within limits, stimulates photosynthesis and metabolism (figure 4), favouring cyanobacteria and dinoflagellates. The 1980s saw several "natural experiments". In 1982/83 a strong El Niño (see below) warmed the North Atlantic, altering zooplankton, fish, seal, and seabird communities throughout the decade. In the Pacific raised surface temperatures overwhelmed the cool, rich, upwelling Humboldt current, altering Peruvian sardine and anchovy yields for years.
In 1986/87 warming supported new growths of species in higher northern and southern latitudes. Since 1987 the dinoflagellate G breve, previously blooming in the Gulf of Mexico, has persisted off North Carolina, following a shoreward intrusion of the Gulf Stream onto the Continental Shelf. In 1987, local factors combined with warm eddies of the Gulf Stream that swept unusually close to Prince Edward Island, and the pennate diatom (Nitzschia pungens) produced domoic acid, killing 5 Canadian mussel consumers and causing 156 cases of amnesic shellfish poisoning. Hundreds of dolphins and whales died that year and there were reverberations throughout the fish and marine mammal populations. In September, 1991, domoic acid appeared in Monterey Bay, California, and hundreds of birds were poisoned. In 1992, massive blooms of Pseudonitzschia pseudodelicatissima occurred in Scandinavian waters; and in 1991/1992 saxotoxins appeared for the first time as far south as the Straits of Magellan.
In the austral summer the sun beats down on the Pacific to generate an eastward-flowing warming centre known as El Niño. Long cycles in earth's tilt are now bringing the Southern Hemisphere closer to the sun so ever more heat permeates this thermal "sink". Ocean-atmospheric modelling predicts stronger and more frequent El Niños, with rising CO2.[11,12] The pace and the duration of strong El Niños, which are associated with extreme weather events worldwide, have accelerated. In 1987/88, 1990/91, and 1992/93 and well into 1993 (and persisting). There have been 2-3deg.C anomalous sea surface temperature rises associated with floods and droughts. The sea is a global thermostat, absorbing atmospheric heat for worldwide distribution through broad surface "streams" coursing north and grand submarine "rivers" returning to the equator. Might this conveyor belt have stalled or even reversed direction to produce the rapid climate swings disclosed in Greenland ice-core records?
Plankton interact with climate by taking in CO2, by absorbing and scattering solar heat, and by emitting dimethylsulphide that seeds clouds, which cool the earth's surface through precipitation and sunlight reflection. These biotic feedbacks help to modulate the earth's temperature, the salinity of its oceans, and the composition of its atmosphere.
By harnessing photosynthetically active radiation marine microflora produce twice as much carbohydrate (60 billion tonnes annually) as terrestrial plants. Phytoplankton produce 70% of atmospheric oxygen, that in turn generates protective ozone in the stratosphere. Having evolved before the ozone shield appeared, phytoplankton can produce auxiliary protective pigments, while small increases in the dose of ultraviolet may harm zooplankton (eg, krill).
Emerging diseases and novel strains
New clinical conditions (or their potential) involving algae include diarrhoea associated with Cyclospora-like bodies (CLB) and as yet unidentified toxins off the French coast.
While pollution-related stress can increase the susceptibility of sea mammals to infections such as phocine distemper in Mediterranean dolphins it is now clear that vast numbers of viruses exist in marine waters and estuaries. They could be involved in the exchange of genetic information. The role of environmental changes in increasing pathogenicity of marine viruses, or creating conditions for the survival of known pathogenic viruses introduced through sewage may become a crucial question for the health of large marine ecosystems.
In 1992 a modified vibrio known as V cholerae O139 emerged, first in coastal communities of India and now rapidly spreading in Asia; and in sewage lagoons near Lima chlorine-resistant vibrios have been isolated. These developments may not seem easily linked to global climate change but they represent altered biodiversity and stability within large marine ecosystems. Extensive monsoon flooding of Bangladesh in July, 1993, has enhanced the dissemination of the "Bangladesh strains".
Climate is controlled by the interaction of oceans, atmosphere, land systems, ice cover, and biota; and a change in any one will destabilise the entire system. The transformation in biomass and community structures of key elements in the marine food web is a result of human activities and a byproduct of the ocean's role in the global thermal budget.
The costs in human health and yields are mounting. The degradation of marine ecosystems increases the risk of diseases emerging, and an enhanced plankton reservoir may help explain the rapid invasion of cholera in the Americas. Changes along coastlines are contributing to public health hazards, and are causing hypoxia in the breeding grounds of marine animals and plants. Physicians may see the remedial actions--a reduction in inputs, protection of wetlands, and preservation of species diversity in the oceans--too remote from their clinical practice. Our contribution, and the Lancet series as a whole, is aimed at closing that gap in perception.
We acknowledge the invaluable input of Dr Patricia Tester of the National Marine Fisheries Service, NOAA, Beaufort Laboratory; Dr. James J McCarthy of the Joint Global Ocean Flux Study and director of the Harvard University Museum of Comparative Zoology; and Dr David Bor, infectious disease specialist at Cambridge Hospital, Harvard Medical School.
In writing this article we have drawn on many other journal publications and official reports, not all readily accessible to Lancet readers. These include Harmful Algae News (a supplement to UNESCO's International Marine Science 1992; 62); the programme of the Sixth International Conference on Toxic Marine Phytoplankton (Nantes, 1993); the journal Oceanus; and bulletins of the US Climate Analysis Center.
Division of Social and Community Medicine, Harvard Medical School, Cambridge Hospital, Cambridge, MA 02139, USA (P R Epstein MD); Department of Environmental Biology, Harvard University (T E Ford PhD); and University of Maryland Biotechnology Institute (R R Colwell PhD)
Correspondence to: Dr Paul R Epstein
Correspondence to: Dr Paul R Epstein
1 Sherman L, Alexander LM, Gold BD, eds. Large marine ecosystems: stress, mitigation, and sustainability. Washington, DC: AAAS Press, 1993: 301-19.
2 Smayda TJ, Shimizu Y. Toxic phytoplankton blooms in the sea. London: Elsevier, 1993.
3 Swerdlow DL, Mintz ED, Rodriguez M, et al. Waterborne transmission of epidemic cholera in Trujillo, Peru: lessons for a continent at risk. Lancet 1992; 340: 28-33.
4 Samadi AR, Chowdhury NK, Huq MI, Khan MU. Seasonality of classical and El Tor cholera in Dhaka, Bangladesh: 17 year trends. Trans R Soc Trop Med Hyg 1983; 77: 853.
5 Byrd JJ, Huai-shu XU, Colwell RR. Viable but non-culturable bacteria in drinking water. Appl Environ Microbiol 1991; 57: 875-78.
6 Islam MS, Drasar BS, Bradley DJ. Long-term persistence of toxigenic Vibrio cholerae O1 in the mucilaginous sheath of a blue-green alga, Anabaena variablis. F Trop Med Hyg 1990; 93: 133-39.
7 Huq A, West PA, Small EB, Huq MI, Colwell RR. Influence of water temperature, salinity, and pH on survival and growtb of toxigenic Vibrio cholerae serovar O1 associated with live copepods in laboratory microcosms. Appl Environ Microbiol 1984; 48: 420-24.
8 Hayes L, Goreau TJ. The tropical coral reef ecosystem as a harbinger of global warming. Wld Resource Review 1991; 3: 306-22.
9 Yan X-H, Ho C-R, Zheng Q, Klemas V. Temperature and size variabilities of the Western Pacific Warm Pool. Science 1992; 258: 1 643-45.
10 MacKenzie D. Mystery of mussel poisoning deepens in Canada as the chain of death spreads to whales. New Sci Jan 28, 1988: 30.
11 Meehl GA, Washington WM. South Asian summer monsoon variability in a model with doubled atmospheric carbon dioxide concentration. Science 1993; 260: 1101-104.
12 Manabe S, Stouffer RJ. Century-scale effects of increased atmospheric CO2 on the ocean-atmosphere system. Nature 1993; 364: 215-18.
13 Glantz MH, Kacz RW, Nicholls N, eds. Teleconnections linking worldwide climate anomalies. Cambridge: Cambridge University Press, 1991.
14 Dansgaard W, Johnson SJ, Clausen HB, et al. Evidence for general instability of past climate from a 250-kyr ice-core record. Nature 1993; 364: 218-20.
15 Bendall RP, Lucas S, Moody A, Tovey G, Chiodini PL. Diarrhoea associated with cyanobacterium-like bodies: and new coccidian enteritis of man. Lancet 1993; 341: 590-92.
16 Valiela I. Marine ecological processes. New York: Springer, 1984.
17 Houghton JT, Jenkins GL, Ephraumis JJ, eds. Climate change: the IPCC scientific assessment. Cambridge: Cambridge University Press, 1990.