Articles

African and Asian Dust: From Desert Soils to Coral Reefs

VIRGINIA H. GARRISON, EUGENE A. SHINN, WILLIAM T. FOREMAN, DALE W. GRIFFIN, CHARLES W. HOLMES, CHRISTINA A. KELLOGG, MICHAEL S. MAJEWSKI, LAURIE L. RICHARDSON, KIM B. RITCHIE, AND GARRIET W. SMITH

Many hypotheses have been proposed to explain the decline of coral reefs throughout the world, but none adequately accounts for the lack of recovery of reefs or the wide geographical distribution of coral diseases. The processes driving the decline remain elusive. Hundreds of millions of tons of dust transported annually from Africa and Asia to the Americas may be adversely affecting coral reefs and other downwind ecosystems. Viable microorganisms, macro- and micronutrients, trace metals, and an array of organic contaminants carried in the dust air masses and deposited in the oceans and on land may play important roles in the complex changes occurring on coral reefs worldwide. Keywords: aeromicrobiology, African dust, Asian dust, contaminants, coral reefs

oral reefs throughout the world are in decline. Increased incidence of coral disease, decreases in reefbuilding corals, increases in macroalgae, and lack of recovery on damaged reefs have been widely reported. The causes and processes driving the declines remain elusive. In the past 25 years, mass mortalities of a major reef-building coral (90% to 95% of Acropora palmata) and of a key herbivore (97% of Diadema antillarum) have significantly altered coral reefs in the Caribbean region. Microbial pathogens are suspected in both mortality events. Twenty years later, the abundance of both species remains low. Eleven years after a cruise ship anchor severely damaged a coral reef in the Virgin Islands, macroalgae dominate the damaged area, and stony coral abundance remains low. Although natural disturbances and human activities are known to have caused much of the damage occurring on coral reefs, they do not explain why increasing numbers of reefs are not recovering from damage but are remaining in alternate

states that appear to be stable. Nor do they adequately explain the apparent increases in disease incidence and mortality from disease or the widespread geographical distribution of coral diseases on reefs remote from as well as near to human influence. For example, sea fan disease, caused by a soilassociated fungus (Smith et al. 1996), occurs throughout the Caribbean region, including reefs off remote carbonate islands devoid of soil. Many hypotheses have been proposed, but they do not adequately explain the declines observed on reefs worldwide (e.g., global warming, pathogen introduction via ballast water or oceanic currents, chronic low levels of nutrients, long-term intense fishing pressure, and multiple low-level stressors). Here we put forward a hypothesis that addresses the widespread distribution of coral diseases and lack of recovery on coral reefs. We will (a) present an overview of the atmospheric transport of African and Asian dust; (b) review relevant background information and current research on airborne microorganisms, coral diseases, and atmospheric

Virginia H. Garrison (ginger_garrison@usgs.gov) is a marine ecologist, Eugene A. Shinn and Charles W. Holmes are geologists, and Dale W. Griffin and Christina A. Kellogg are microbiologists at the US Geological Survey (USGS), 600 Fourth Street South, St. Petersburg, FL 33701. Garrison is a coral reef ecologist; her research focuses on the effects of global-scale processes on coral reef ecosystems in the Caribbean and in the Pacific Ocean. Shinn has published more than 100 scientific articles concerning geology, limestone diagenesis, and the geologic history and growth of coral reefs in Florida, the Bahamas, the Persian Gulf, the Marshall Islands, and Puerto Rico. Griffin studies transoceanic transport of microorganisms in the atmosphere. Holmes specializes in the biogeochemical cycles of short-lived radionuclides and reconstruction of the paleoenvironment from coral skeletons. Kellogg works on characterizing the microbial communities in African dust and polluted sediments. William T. Foreman is a research chemist at USGS National Water Quality Laboratory, Box 25046, Denver, CO 80225; he develops and applies methods to characterize the transport and fate of organic chemical contaminants in air, water, sediment, and biota. Michael S. Majewski is a research chemist at USGS, 6000 J Street, Sacramento, CA 95819; his research interests include the environmental behavior and fate of organic contaminants, their long-range atmospheric transport and deposition, and the effects of air pollution on water quality. Laurie L. Richardson is an associate professor in the Department of Biology, Florida International University, Miami, FL 33199; her area of expertise is microbial physiological ecology, with a focus on the role of microorganisms in the etiology of coral diseases. Kim B. Ritchie is a scientist at MicroGenomics Inc., 5935 Darwin Court, Carlsbad, CA 92008; her research interests include molecular biology, marine microbial communities, and marine pathogens in the Atlantic and Pacific Oceans. Garriet W. Smith is an associate professor at the University of South CarolinaAiken, 471 University Parkway, Aiken, SC 29801; he is a microbial ecologist interested in the ecology of marine microorganisms, particularly disease pathogens. 2003 American Institute of Biological Sciences.

May 2003 / Vol. 53 No. 5 BioScience 469

Articles
transport of chemical contaminants; and (c) suggest causal mechanisms and strategic avenues of investigation. chemists and physicists have studied the mineralogy, elemental composition, geologic processes, transport, deposition, optical properties, atmospheric chemistry, and particle size distribution of globally transported dust and the linkages between dust and global climate. Hundreds of millions of tons of dust derived from mineral soil are transported annually from the African Sahara and Sahel to the Mediterranean and Europe, and across the tropical Atlantic to the Americas (figure 2; Moulin et al. 1997). A similar global system transports dust from the Gobi and Takla Makan deserts (western China) across Korea, Japan, and the northern Pacific (Duce et al. 1980, Husar et al. 2001) to the Hawaiian Islands (figure 2).The Asian air masses periodically reach western North America and infrequently continue eastward to the Atlantic Ocean. The primary source of African dust is thought to shift from the Bilma-Faya Largeau area of Lake Chad (October to April) to the southern Sahel, such as the Niger Delta in Mali (May to September; McTainsh et al. 1997). The annual variation in exported dust mass correlates with the North Atlantic Oscillation (NAO), which affects precipitation over North Africa and atmospheric circulation in the Northern Hemisphere (Moulin et al. 1997) and is coupled with the Arctic and Pacific Oscillations of the Northern Hemisphere and with the El Nio Southern Oscillation (ENSO). Flux of dust is greater during positive phases of NAO. The NAO Index and average annual surface concentration of African dust at Barbados have gradually increased since the 1960s (Prospero and Nees 1986), with both showing maxima during the strong ENSO of 19831984 (Moulin et al. 1997). The mass of dust transported may be an indicator of desertification of the Sahel (e.g., Prospero and Nees 1977, Swap et al. 1996) and may be linked with both changing land-use practices and a drier climate (e.g., Tegen et al. 1996). Climate is thought to be the major forcing factor in dust transport (Prospero and Nees 1986, Moulin et al. 1997). Since the early 1970s, the frequency of intense convective disturbances and associated rainfall in the Sahara has decreased, while the number of weaker disturbances that mobilize large quantities of dust has increased. Rosenfeld and colleagues (2001) describe a feedback loop wherein high concentrations of small dust particles suppress precipitation over the desert. The combination of (a) more frequent weak disturbances mobilizing large quantities of dust and (b) scant rainfall scavenging dust from the atmosphere results in greater quantities of dust available for atmospheric transport. Extending thousands of kilometers (km) downwind and lasting up to 10 days per event (Perry et al. 1997), maximum dust transport to the Americas centers in winter at approximately 5 N (Guyana, South America) and moves north to 20 N (northern Caribbean) in summer in association with the shift of the Intertropical Convergence Zone (Moulin et al. 1997). The concentration of atmospheric dust in the Niger Delta (Mali) can exceed 13,000 micrograms (g) per cubic meter (m3), with mean background concentrations of 575 g per m3 (Gillies and Nickling 1996). Aerosol concentrations during heavy dust events range from 200 to 600 g per m3 over

The hypothesis
We propose that the hundreds of millions of tons of dust transported annually from Africa and Asia to the Americas may be a significant factor in coral reef decline and may be adversely affecting other downstream ecosystems as well (Shinn et al. 2000). Why would these changes occur now, after millennia of African and Asian dust transport? We suggest that the quantities of dust transported have increased and that the composition of the dust has changed. Synthetic organic chemicals and anthropogenic pollutants, viable microorganisms, macroand micronutrients, and trace metals are likely to be carried in the dust air masses and deposited in the oceans and on land. Singly or in combination, they may play important roles in the complex changes occurring on coral reefs worldwide.

Atmospheric transport of dust

As Charles Darwin traveled across the tropical North Atlantic, dust from Africa reduced visibility and coated the HMS Beagle (Darwin 1846). More than 100 years later, Delaney and colleagues (1967) and Prospero (1968) were the first to measure soil dust particles in the tropical North Atlantic atmosphere and to hypothesize that the trade winds transport dust from the Sahara to Barbados. Confirmation of the hypothesis came from satellite imagery (figure 1) and from measurements of dust flux over Caribbean islands, the Bahamas, Miami (Florida) (e.g., Prospero 1968), and several national parks in southeastern North America (Perry et al. 1997). Since the late 1960s, geologists and atmospheric

Figure 1. National Aeronautics and Space Administration (NASA) satellite image of dust cloud from Africa crossing the Atlantic Ocean. Photograph: SeaWiFS (Sea-viewing Wide Field-of-view Sensor) Project, NASA/Goddard Space Flight Center, and ORBIMAGE. (Enhanced true color.)
470 BioScience May 2003 / Vol. 53 No. 5

Articles

Figure 2. Dust is transported in two major global dust transport systems: (1) from the Sahara and Sahel of Africa to the Americas, Europe, and Near East; and (2) from the Takla Makan and Gobi deserts of China, across China, Korea, Japan, and the northern Pacific to North America, sometimes exiting over the Atlantic Ocean. Illustration: Betsy Boynton. the mid-Atlantic and Barbados (Prospero et al. 1981, Talbot et al. 1986) and exceed 100 g per m3 in the Virgin Islands (figure 3), with mean concentrations of 3 to 20 g per m3 in the Caribbean (Prospero and Nees 1986, Perry et al. 1997). Mean mass diameters of African dust particles transported decrease with increasing distance from Africa and average less than 1 micrometer (m) in the Virgin Islands (Perry et al. 1997). The finest particles are carried into the Gulf of Mexico and over the continental United States as far west as the Rocky Mountains and as far north as Maine (Perry et al. 1997). America, Europe, and Africa. Emissions from industry and possibly biomass burning were more evident in the aerosols from Europe and Africa. The relative abundance of chemical elements, particularly the metals, has been used to distinguish soil derived from volcanic dust (Muhs et al. 1990). To define the source, various elemental ratios, such as calcium to aluminum (Ca:Al) and aluminum to silicon, have been suggested. Perry and colleagues (1997), using a particle diameter of less than 2.5 m and a Ca:Al ratio greater than 3.8 to indicate Saharan dust, identified African dust in the atmosphere of the Caribbean and of the eastern and central United States. Using the metal ratios in volcanic material from the Antilles and soils from Africa, Muhs and colleagues (1990) identified Saharan dust as the source of clay soils on carbonate platforms in the western Atlantic (Barbados, Jamaica, Bermuda, and the Florida Keys). Age dating of the soils suggests the transport of African dust to the region has been occurring for millions of years. The ratios of beryllium to lead (7Be:210Pb) in aerosols can be used to differentiate continental from oceanic sources, and these ratios show promise in elucidating aerosol transport history (e.g., Arimoto et al. 1999).

Composition of dust
The chemical composition of aerosols is a product of the origin and history of the dust (e.g., Duce et al. 1976, Arimoto et al. 1995, Perry et al. 1997). African and Asian dust consists primarily of clay soil minerals such as illite, quartz, kaolinite, chlorite, microcline, plagioclase, and calcite (Prospero 1981), which may undergo chemical change during aerosol transport (Ravishankara 1997). Some elements (e.g., manganese, iron, scandium, cobalt) occur on African dust particles in concentrations similar to average crustal abundance, whereas other elements (e.g., mercury, selenium, lead) accumulate, via scavenging, at concentrations three orders of magnitude greater than mean crustal abundance (Duce et al. 1976). In the late 1990s, the AtmosphereOcean Chemistry Experiment investigated the chemical composition of Atlantic aerosols (Arimoto et al. 1995). Trace metal analysis showed large-scale differences in pollution emissions from North

Dust chronologies in coral skeletons

Corals incorporate particles (Barnard et al. 1974) and chemical components from the surrounding water into their skeletons during growth, preserving a record of the environmental conditions under which the calcium carbonate skeleton formed. For example, mineral dust and dust from the 1883 eruption of Krakatoa (Indonesia) have been isolated and
May 2003 / Vol. 53 No. 5 BioScience 471

Articles
1991). Iron, a micronutrient, can limit phytoplankton productivity in oligotrophic waters; newly deposited iron is quickly depleted by phytoplankton (e.g., Martin and Fitzwater 1988, Coale et al. 1996, Behrenfeld and Kolber 1999) and bacteria (Butler 1998). Turner and colleagues (1996) showed that iron flux to the oceans leads to the biotic production of dimethylsulfide (DMS) and its release into the atmosphere. Subsequent oxidation of DMS and formation of sulfate in turn produces sulfuric acid, which with atmospheric mixing could increase the solubility of iron (in the form Fe [III]) in the mineral aerosols (e.g., Duce and Tindale 1991). Walsh and Steidinger (2001) and Lenes and colleagues (2001) linked African dust to the development of extensive red tides in the Gulf of Mexico. Lenes and colleagues (2001) also showed that iron in African dust deposited by rain fuels blooms of Trichodesmium, an iron-limited cyanobacterium. The nitrogen-fixing Trichodesmium produces nutrients (e.g., nitrates, nitrites) that, in combination with selective Figure 3. African dust event (bottom) and nondust (top) atmospheric predation by zooplankton and meteorologconditions in the Virgin Islands. Distance to islands in background is 2 to 5 ical conditions, fuel blooms of the red tide kilometers. Photographs: Courtesy of Christina Kellogg. dinoflagellate Karenia brevis. Recently, Bishop and colleagues (2002) documented an inidentified from annual bands of the reef-building coral Moncrease in carbon biomass in the North Pacific in response to tastraea annularis from the Florida reef tract (Merman 2001). iron influx from a strong Asian dust event. The study conclusively demonstrated that dust can be inSome scientists question the importance of atmospheric corporated in the coral skeleton. Anthropogenic air pollutants dust as a source of iron in the oceans photic zone, arguing that such as lead and cadmium (Shen and Boyle 1987) and cesium the most stable and abundant form of iron (Fe [III]) in dust radioisotope 137Cs from nuclear testing have been detected in is relatively insoluble in seawater and not readily available bicoral skeletons. Current studies indicate that a record of past ologically. However, little is known of the chemistry occurenvironmental conditions, including dust deposition, may be ring on dust particles of small volume and high surface area reconstructed using laser ablation inductively coupled plasma during atmospheric transport, and biogeochemical oceanmass spectrometry (LA-ICP-MS) to extract the record preographers are just beginning to understand the oceans served in coral skeleton annual bands (Sinclair et al. 1998). complex biogeochemical pathways. Dust from Africa and Asia is transported long distances in chemically and physically Nutrient influx and biogeochemical effects extreme environments where the small particles are exposed Our understanding of the impact of dust on biogeochemical to high levels of solar radiation, multiple freezethaw cycles, cycles is limited. Aeolian dust contributes significant quanrelatively acidic conditions, and predominantly inorganic tities of water-soluble nutrients to the oligotrophic Caribbean salts (Jickells 1999). On its deposition to the ocean, the dust (Jickells 1999), the Gulf of Mexico (e.g., Talbot et al. 1986), enters a radically different environment. The thin surface and the Pacific (e.g., Young et al. 1991, Behrenfeld and Kollayer of the ocean is characterized by concentrations of phyber 1999) and provides essential nutrients to the rain forests toplankton and zooplankton and by a steep concentration of Hawaii (Chadwick et al. 1999) and the Amazon (Swap et gradient of organic compounds and inorganic salts. It has been al. 1992). The atmosphere deposits an estimated 50% of the suggested that during atmospheric transport, photoreduction phosphorus transported to the oceans (Duce et al. 1991). of Fe (III), which is stable and relatively insoluble, produces The influx of nutrients from Asian dust events is reported to Fe (II), a biologically available and soluble species (e.g., fuel phytoplankton productivity in the northern Pacific (e.g., Graedel et al. 1986, Duce and Tindale 1991). Saydam and Young et al. 1991). Senyuva (2002) propose that oxalate released by fungi in Atmospheric deposition is thought to be the dominant desert dust facilitates photoreduction of Fe (III). Complexsource of iron in the oceans photic zone (Duce and Tindale ation of Fe (II and III) with organic ligands (Butler 1998) or
472 BioScience May 2003 / Vol. 53 No. 5

Articles
with clay minerals in the aerosol acts to stabilize the iron in a bioavailable form. On their deposition to the ocean surface layer, insoluble Fe (III) and some of the more soluble first-row transition metals form stable complexes with siderophores, low-molecular-weight organic ligands produced by some species of oceanic bacteria, facilitating uptake by microorganisms and phytoplankton (Butler 1998 and references therein). Young and colleagues (1991) suggested that the smaller the dust particle, the longer the residence time in the photic zone and therefore the greater the amount of iron that could be released and made available for phytoplankton and microbes. In a more direct mechanism, Fe (III) deposited in reducing environments (e.g., carbonate muds in Florida Bay) can be reduced to Fe (II). Iron limitation is known to keep many microbial pathogens at low concentrations that directly counteract the expression of pathogenicity (Weinberg 1996). Hayes and colleagues (2001) proposed that iron in dust may play a similar role in promoting microbial diseases on coral reefs. Biomagnification and bioaccumulation processes add another layer of complexity in marine systems. Small dust particles (0.5 to 50 m) deposited in the ocean surface layer are selectively ingested by zooplankton (e.g., Syvitski and Lewis 1980), incorporated into fecal pellets, and excreted in a more concentrated form into the water column. Zooplankton fecal pellets, singly or aggregated, could efficiently carry the bioconcentrated micronutrients and chemical contaminants to benthic organisms, larger zooplankton, or planktivores (Syvitski and Lewis 1980). Doses of synthetic organic chemicals and micro- or macronutrients could affect reproduction or immune function of benthic organisms and contribute to maintaining the shift to seemingly alternative stable states (e.g., macroalgae dominated) observed on disturbed reefs worldwide. Threshold effects may be involved. Exploration of the roles of atmospheric transport of dust in the biogeochemical cycles of marine and terrestrial ecosystems is in the earliest stages. systemic fungal diseases, and viruses such as hantavirus, are typically transmitted in dust. The World Health Organization has identified drought and dust storm activity in the subSaharan region of Africa as causing regional outbreaks of meningococcal meningitis. Neisseria meningitidis, the infectious agent, causes approximately 500,000 cases and 50,000 deaths every year (see www.who.int/vaccines/en/meningococcus. shtml). Other disease outbreaks are associated with exposure to desert dust clouds in the Americas. One example is coccidioidomycosis in humans, caused by the fungus Coccidioides immitis. It is generally believed that most airborne pathogens are only transmitted over short distances. Although the transmission of coccidioidomycosis or meningococcal meningitis has been documented through desert dust cloud exposure, it has only been shown to occur within the confines of a continent. Can desert dust clouds move viable pathogens around the planet, and is there a limit to the type of airborne pathogen that can be transported in this way? There are a surprising number of reports of long-range transport of plant pathogens. Most of these infectious invaders are fungi, whose dispersal spores provide protection from ultraviolet (UV) light and other harsh environmental conditions. Culturing of air samples and species-specific detection are relatively new developments, so in most studies the fungal diseases were tracked based on the geography of the outbreaks, prevailing winds, and timing. Examples include the potato blight fungus (Phytophthora infestans), potato blackleg bacteria (Erwinia carotovora and E. chrysanthemi), lentil anthracnose (Colletotrichum truncatum), fungal wheat pathogen (Puccinia graminis), sugarcane rust (Puccinia melanocephala), coffee rust (Hemileia vastatrix), and banana leaf spot (Mycospherella musicola; Brown and Hovmller 2002). Less information is available concerning the aerosol transmission of animal pathogens. Several studies of dust collected from surfaces on poultry, swine, and dairy farms have shown that the dust contained fungi such as Aspergillus and Cladosporium (Fiser et al. 1994) as well as the bacterium Salmonella (e.g., Letellier et al. 1999). Windblown desert dust that carried fungi caused an outbreak of aspergillosis in desert locusts (Venkatesh et al. 1975). Meteorological data and molecular techniques were employed to determine the source of the pseudorabies virus (cause of Aujeszkys disease in swine) after outbreaks occurred in Denmark in December 1988. The evidence suggested the infections were a result of airborne transport of the viral pathogen from Germany (Christensen et al. 1993). Microbes and African dust. We are investigating the possibility that viable microbes are being transported across oceans in dust events. Research to date has shown that the number of bacteria and fungi that can be cultured from air samples from the Virgin Islands is two- to threefold greater when African dust is in the atmosphere than when it is not (Griffin et al. 2001). Comparisons of data from these cultures
May 2003 / Vol. 53 No. 5 BioScience 473

Living components of globally transported dust

Dust air masses transport more than mineral particles and nutrients. Sediment core studies conducted in the Atlantic Ocean off the West African coast reported the presence of freshwater diatoms and phytoliths in sediment cores and attributed their presence to African dust deposition (e.g., Maynard 1976). African desert locusts (Schistocerca gregaria, up to 8 centimeters [cm] long) have been carried in Saharan dust air masses across the tropical Atlantic and arrived alive on eastern Caribbean islands (Ritchie and Pedgley 1989). If organisms as large as desert locusts have survived the multiday trip, it is likely that viable smaller organisms may be riding with the dust as well. Airborne transport of pathogens. Many well-known pathogenic bacteria, fungi, and viruses are transmitted through airborne transport (e.g., the organisms causing plague, anthrax, tuberculosis, influenza, and aspergillosis). Most of the

Articles
with data from direct counts of aerial microbes indicate that less than 1% of the total microbial population present was recovered on the nutrient agar used for analysis. This finding is in agreement with other microbial ecology studies that have shown a culturable rate of 1% or less obtained from natural samples (Torsvik et al. 1990, Eilers et al. 2000). It is important to note that microbes that do not grow on one nutrient agar may grow on a different nutrient agar or source (e.g., lung tissue). In short, the bacteria and fungi that we have cultured from atmospheric samples most likely represent only a small percentage of what is actually viable and capable of growth. Furthermore, microbial ecology studies have shown that in environmental samples the total viral population present is usually 10 to 100 times the total bacterial population (Borsheim et al. 1990). Nonculturable microbes can be identified using whole genomic DNA extraction, followed by polymerase chain reaction (PCR) amplification, cloning, and sequencing of ribosomal RNA genes. This type of analysis is invaluable for total microbial population comparisons of dust and marine microbes and provides genetic access to the large number of microbes (> 99%) that cannot be cultured. The fungus known to cause sea fan disease in the tropical western Atlantic (Aspergillus sydowii; Smith et al. 1996) was isolated in its active, pathogenic form from air samples taken during dust events in the Virgin Islands in 1997 (Weir et al. 2000). Since that time, the pathogenic strain of A. sydowii has been found only in diseased sea fans and sea plumes and in air samples collected during dust events in the Virgin Islands. The pathogenicity of A. sydowii isolates from air samples collected in Mali is currently being tested. Recently, Griffin and colleagues (2001) isolated numerous species of viable microorganismsbacteria and fungifrom air samples taken in the Virgin Islands during dust events. Using ribosomal sequencing via PCR to identify the isolates, they found that at least 25% of those identified are known plant pathogens and 10% are opportunistic human pathogens (Griffin et al. 2001). The fungal isolate Cladosporium cladosporioides, which has been identified as a plant and human pathogen, is one of the most common species of fungi recovered from air samples. Several of the bacteria isolated have previously been identified in marine environments (Pseudomonas alcalophila, Paracoccus spp., and Kocuria erythromyxa). Microorganisms can move from the marine environment to the atmosphere via physical surface activity that results in the formation of sea spray and sea foam (e.g., Blanchard and Syzdek 1970). It is possible that dust clouds traversing marine environments serve as a vehicle for long-range transport of aerosolized marine microorganisms. Further sampling, midocean and in the Virgin Islands and Mali, is necessary before we can conclusively demonstrate what viable microorganisms are transported from Africa and in what concentrations (see sidebar). In a current study of a recent disease outbreak in Caribbean sea urchins (Meoma ventricosa), two of the authors (K. B. R. and G. W. S.) found that bacteria associated with the spines
474 BioScience May 2003 / Vol. 53 No. 5

Local contamination or long-distance transport?

To distinguish between contaminants transported long distances and those generated locally, background samples are collected to control for sources other than African or Asian dust. Ambient or background samples are used to identify and quantify the microorganisms and chemical contaminants in the atmosphere when dust air masses are not present. Sources of background contaminants include local winds advecting soil, ship traffic across oceans upwind of the affected areas, complex interactions between air layers in the atmosphere, and pollutants in the troposphere. In addition, it is essential to sample the source region, midocean, and downwind sites during dust events and nondust (background) conditions to conclusively demonstrate longdistance transport of contaminants. A microorganism or chemical contaminant is considered to have been transported long distances in a dust event to downstream sites when criteria 1 (a or b), 2, and 3 are met: 1. The microorganism or chemical (a) is identified only in dust event samples and not found in background samples or (b) occurs in significantly higher concentrations in dust event samples than in background samples. 2. The microorganism or chemical occurs in higher concentrations in the source region than in the downstream and midocean dust event samples. 3. The microorganism or chemical occurs in source region, midocean, and downstream site dust event samples. of burrowing and nonburrowing sea urchins and present in the water column are genetically identical to bacteria purified from air samples collected in the Virgin Islands during dust events. One isolate was a novel bacterium that showed 98% identity to Bacillus mojavensis, which was initially isolated from the Mojave Desert (Southern California) and associated with arid soils. The sea urchinassociated bacterium proved to be genetically and morphologically identical to a bacterium isolated from a Virgin Islands dust event sample. Moreover, another bacterium isolated from a Sargassum species in San Salvador (Bahamas) was found to be genetically and morphologically identical to isolates from both sea urchins and Virgin Islands air samples during dust events. Thus, it appears that some bacteria found in desert soils also occur in a viable state in the atmosphere over the Virgin Islands during African dust events and in the marine environment in the Western Atlantic. Whether these bacteria are actually transported long

Articles
distances in dust events or can reproduce in the marine environment is unknown. Analysis of other marine and dust isolates reveals that other bacterial types remain viable in the atmosphere and marine environments as well. Bacteria isolated from M. ventricosa, from the sea urchin Echinometra variegates, from the water column, and from a Virgin Islands dust event air sample were 100% identical to Bacillus pumilus. Similarly, bacteria isolated from the long-spined sea urchin, Diadema antillarum, and from the nonburrowing urchin Lytochinus variegates, are identical to a bacterium isolated from an African dust event air sample from the Virgin Islands, with 99% identity to Pseudomonas aeruginosa, a ubiquitous species. It had been thought that exposure to UV light during long-range transport would inactivate any microbes. Dust in the upper altitudes of dust clouds attenuates UV light and could act as a UV screen for microbes at lower altitudes (Herman et al. 1999). Also, microorganisms may attach to and survive within cracks and crevices of inorganic dust particles, which would shield them from UV light. Preliminary findings indicate that UV resistance in microorganisms may play an important role in their survival during atmospheric transport. ventalina (figure 4, left) and G. flabellum. As discussed above, it is caused by the common terrestrial fungus A. sydowii (Geiser et al. 1998), which has a worldwide distribution in soils. Spores are generally less than 2 m in diameter, a size easily carried by winds and dispersed over great distances. To date, pathogenic and nonpathogenic fungi from several different genera (Aspergillus, Penicillium, and Blastomyces) have been isolated and identified from samples collected during African dust events in the Virgin Islands. Inoculations confirmed that some dust event isolates were pathogenic to healthy G. ventalina, because inoculations produced characteristic signs of aspergillosis, such as purpling and loss of tissue. In each case where disease was observed, the original pathogen was reisolated and morphologically identified, fulfilling Kochs postulates. During the reisolations, several types of fungi in addition to A. sydowii were isolated from experimentally inoculated sea fans, indicating that (a) pathogenic fungi can be reisolated from inoculated and diseased sea fans, and (b) nonpathogenic opportunistic marine fungi may secondarily colonize diseased sea fans. The absence of Aspergillus species in control samples (not inoculated) revealed that Aspergillus is not typically found on healthy sea fans. The presence of at least one nonpathogenic fungus in such controls indicated that it is possible for some marine fungi to colonize sea fans without causing signs of disease. The data collected from these experiments suggest that African dust storms may be a source of Caribbean sea fan pathogens. In the last year, aspergillosis occurred on at least two other species of soft corals, Pseudopterogorgia americana in Bermuda and P. acerosa in St. Croix (Virgin Islands). Both of these species displayed similar signs of disease, which included gall formation and loss of tissue along the axial skeleton. Aspergillus species were morphologically identified from diseased specimens only. This may indicate an extended host range of pathogenic strains of Aspergillus. Interestingly, when hyphae of pure A. sydowii strains are extracted and separated using high-performance liquid chromatography, unique compounds are found associated with the pathogenic strains but not with the nonpathogenic strains. We do not know the identities of these compounds, but we suspect that some may play a role in the disease process. Black band disease. Black band disease was the first coral disease reported (Antonius 1973) and is one of the most well characterized. A cyanobacterium, sulfide-oxidizing and sulfate-reducing bacteria, and many heterotrophic bacteria together form a complex pathogenic microbial consortium that closely resembles microbial mats such as those found in illuminated, sulfide-rich aquatic environments (Richardson 1998). Black band disease can affect both soft corals (such as sea fans) and stony corals, but it appears to be a greater threat to the reef-building corals, in particular Montastraea annularis and Colpophyllia natans. Compared with other coral diseases, the disease is slow in terms of both degradation of coral tissue (normally 3 to 4 millimeters per day) and the time it takes to kill a coral colony (months to years) (Rtzler and
May 2003 / Vol. 53 No. 5 BioScience 475

Coral diseases and microbial pathogens of corals

Coral diseases were first reported in the 1970s (e.g., Antonius 1973) on reefs in the Caribbean and in the Florida Keys. Since that time, they have been observed on reefs worldwide (Richardson and Aronson 2002). Presumably, diseases have always occurred on coral reefs, but they remained undetected until the 1970s because of the low incidence of disease on reefs and the state of knowledge at the time. The current levels of coral mortality are unprecedented. It is now accepted that coral diseases are an active and important factor in the continued degradation of coral reefs. Although up to 29 coral diseases are currently being monitored in various programs worldwide, the pathogens of only five diseases (black band, aspergillosis, bacterial bleaching, white plague, and white pox) have been identified. Four of these have been proved to be pathogens, following the procedures of fulfillment of Kochs postulates: aspergillosis (Smith et al. 1996), white plague type II (Richardson et al. 1998), bacterial bleaching (Kushmaro et al. 2001), and white pox (Patterson et al. 2002). The fifth disease, black band, is caused by a pathogenic bacterial consortium, not a single pathogen. Reservoirs are known for only two coral diseases, black band disease and aspergillosis (Richardson 1998, Weir et al. 2000). Of the 10 coral diseases or disease-like states being monitored throughout the Caribbean and the Florida Keys (Richardson 1998), four are of particular interest in terms of defining a connection with African dust events: aspergillosis, black band disease, white pox, and white plague. Aspergillosis. One of the few coral diseases to be characterized fully, aspergillosis is also one of the most common. This disease primarily affects two types of sea fans, Gorgonia

Articles

Figure 4. On left, diseased sea fan (Gorgonia ventalina, approximately 60 centimeters high) showing characteristic lesions of aspergillosis (Aspergillus sydowii). On right, close-up of stony coral (Montastraea annularis) with white plague type II. White areas are bare coral skeleton where tissue has been recently killed by disease. Photographs: Sea fan, courtesy of Garriet Smith; stony coral, courtesy of US Geological Survey. Santavy 1983). The incidence is generally low: less than 1% of coral colonies on reefs in the Virgin Islands (Edmunds 1991) and in the Florida Keys, but up to 6% on reefs in Jamaica (reviewed in Richardson and Aronson 2002). The low incidence of black band disease and the low rate of black-band-induced coral mortality have led investigators to suggest that the slow death of corals caused by this disease may be balanced by the beneficial effect of opening up reef substrate for recruitment of stony corals, thereby potentially increasing genetic diversity (Edmunds 1991). However, such recruitment has not been found in two studies that documented recruitment of corals (both stony and soft) onto black-band-exposed substrate in the Virgin Islands (Edmunds 1991, 2000) and the northern Florida Keys (Richardson and Aronson 2002). The potential connection of black band disease with African dust may relate not only to potential pathogen transport but also to the iron present within dust. Iron may act as a trigger that activates black band disease reservoirs to become pathogenic. Richardson (1997) found that the reservoir of the black band microbial consortium resides in biofilms on sediment patches in depressions on healthy coral colonies of species susceptible to the disease. These biofilms are present as thin layers of filaments of the black band cyanobacterium. When samples are viewed under a microscope, the presence of the sulfide-oxidizing black band member Beggiatoa is seen. Because this genus only lives in environments with sulfide/ oxygen interfaces, the presence of sulfate reducers (the third major component of the black band consortium) is inferred.
476 BioScience May 2003 / Vol. 53 No. 5

It has been proposed that the black band reservoir biofilms are transformed into the pathogenic microbial mat community by an accumulation of biomass. As biofilm community members increase in biomass, oxygen demand increases, which leads to increased areas of anoxia within the microbial community. As anoxic microzones develop, the sulfatereducing population can increase and a zone of permanent anoxia can develop. When this occurs, sulfide concentrations may build up within the anoxic microzone to the toxic level that kills coral tissue (Richardson 1998). The trigger that may induce the buildup in biomass may very well be iron, known to be a limiting micronutrient in tropical and subtropical marine waters. White plague. To date, three forms of white plague have been reported on reefs of the northern Florida Keys: white plague type I; white plague type II (figure 4, right), a more virulent form that emerged on the same reefs in 1995; and white plague type III, first seen in 1999 (Richardson and Aronson 2002). These three forms exhibited identical disease signs but had very different short-term effects on the reef. White plague type I affected six species of stony corals and was described as having minimal impact on the reefs. White plague type II was much more destructive, affecting 17 species of stony corals along 400 km of Floridas reef tract and causing the death of entire colonies in as little as 2 to 3 days (Richardson et al. 1998). White plague type III appeared on the same reefs in 1999 (Richardson and Aronson 2002) but caused the most

Articles
rapid tissue destruction of the three white plagues, primarily affecting the largest colonies of M. annularis and C. natans. Entire colonies (2 to 3 meters in diameter) were observed to have been killed within days. The pathogen of white plague type II was isolated and identified as a new genus of gram-negative bacteria, Aurantimonas coralicida (Denner et al. 2002). The white plague pathogen may well be one of the dust-associated bacteria, and dust may serve as an inoculum that triggers white plague outbreaks. White plague types II and III have spread throughout the Caribbean in the last 2 years, even to pristine reefs that are far from human populations. Although this has been a mystery in the past, the dust hypothesis may explain regionwide disease distribution patterns. White pox. White pox, a coral disease that forms circular lesions on colonies of elkhorn coral (Acropora palmata), was recently shown to be caused by the gram-negative enterobacterium Serratia marcescens (Patterson et al. 2002). This microorganism is ubiquitously distributed and could be introduced onto coral reefs via a number of pathways, including soil runoff, river discharge, arthropod or mammal feces, and atmospheric transport of soil. Microorganisms and marine ecology. Microbiota play a significant role in marine ecology. Marine microorganisms have been shown to be significant with respect to the survivability and competitiveness of key coral reef species, drastically altering the composition and possibly the viability of some reefs. The composition and function of the microbial communities associated with healthy marine organisms are just beginning to be explored. The first priority is to identify the pathogens or agents causing the 20 to 30 disease-like states being followed on reefs today, some of which may not be diseases (Richardson 1998). Once pathogens have been identified, the sources and mechanisms of transmission and distribution need to be addressed. It is equally important to identify the environmental conditions needed to activate a disease, to induce a microorganism to become pathogenic, or to initiate a chain of events that will lead to an outbreak of disease. Many questions remain to be answered, including the source of identified coral disease pathogens. African dust may be one mechanism through which pathogens of coral reef organisms are distributed and deposited in the Atlantic and Pacific. It is still unclear whether pathogens responsible for coral disease outbreaks are newly introduced microorganisms, more virulent forms of normally present pathogens, or normal bacterial populations that become virulent due to decreased host resistance to infection. We believe that atmospheric deposition of dust to coral reefs is involved. and after their manufacture and use, many of these compounds become airborne through direct emission, combustion, volatilization, or wind erosion of soil particles to which they are sorbed (Majewski and Capel 1995). Once airborne, semivolatile organic compounds move through the atmosphere in the gas phase or sorbed to dust particles and are deposited by wet and dry depositional processes (Majewski and Capel 1995, Bidleman 1999). Persistent organic pollutants (e.g., DDT, other organochlorine insecticides, and polychlorinated biphenyls [PCBs]) can be transported great distances, continuously cycle between Earths surface and the atmosphere, become globally distributed, and eventually accumulate in polar regions and the deep seas (Bidleman et al. 1989, Looser et al. 2000). Dust air masses transport chemical contaminants from the dust source regions, and from populated areas over which the air masses move (Uematsu et al. 1992), thousands of kilometers across continents and oceans. Long-range atmospheric transport and deposition of some persistent organochlorine pollutants (especially polychlorinated dibenzo-p-dioxin and furan [PCDD/Fs], PCBs, and some chlorinated insecticides like DDT, chlordane, and toxaphene), and of anthropogenic and biogenic aliphatic hydrocarbons (e.g., n-alkanes) and polycyclic aromatic hydrocarbons (PAHs), are well documented (e.g., van Pul et al. 1999). Some studies have focused specifically on transport to the oceans (e.g., Simoneit et al. 1988, Atlas and Schauffler 1990, Duce et al. 1991, Bidleman 1999, MacDonald et al. 2000). Other persistent halogenated organic pollutants have recently been identified as important atmospherically transported persistent contaminants (see Lipnick et al. 2001), including polychlorinated n-alkanes and naphthalenes, polybrominated diphenyl ethers (used as flame retardants), and perfluorinated chemicals (e.g., perfluorooctane sulfonate; Giesy and Kannan 2001). Limited information is available on the atmospheric transport and deposition for some current-use pesticides (Majewski and Capel 1995, van Dijk and Guicherit 1999). Other pesticides, antibiotics, pharmaceuticals, and many other organic chemical contaminants and their degradates have not been studied. Potential synergistic toxicity or activity of transported chemical contaminants is essentially unknown (Kolpin et al. 2002). The contaminants may adversely affect downwind marine and terrestrial ecosystems through a number of pathways. We suggest that a combination of regional geomorphology, synthetic organic chemical use, and socioeconomic and cultural factors may be altering the quality of transported dust. In the Sahara and Sahel of West Africa, all forms of garbage are burned for fuel and for the ash that fertilizes agricultural plots. The continual burning severely degrades air quality, and respiratory complaints are common. Before the late 1980s, garbage consisted primarily of animal and plant waste. Since then, the composition of garbage has fundamentally changed. Today, plastic products (primarily plastic bags), tires, and other manufactured goods are routinely burned. Combustion of synthetic materials is known to release hazardous
May 2003 / Vol. 53 No. 5 BioScience 477

Chemical contaminants
Although dust events have occurred for millennia, the production, use, and release of synthetic organic chemicals into the environment is a relatively recent phenomenon. During

Articles
contaminants. For example, burning of municipal or household waste containing various chlorinated materials, such as polyvinyl chloride plastics, is a major source of PCDD/F emissions to the atmosphere (e.g., Lemieux et al. 2000). These combustion practices are believed to concentrate heavy metals. The primary source of PAHs in the atmosphere is combustion of fuels ranging from petroleum to wood. Although low compared with the levels in some regions of the world, pesticide use in the Sahel region of Africa is increasing in order to control disease vectors and protect crops from plant and insect pests. Organochlorine pesticides (e.g., DDT) historically have been used, but newer-generation pesticides (e.g., synthetic pyrethroids) are increasingly being applied in dust source regions (van der Valk and Diarra 2000). Many of these pesticides are known to volatilize or sorb to small clay particles, which can be transported over long distances. Antibiotics and pharmaceuticals are widely used to treat the numerous diseases and infections that occur in West Africa. Basic sanitary facilities such as pit toilets and sewage treatment are lacking, and a large river (Niger), the primary depository of waste, floods annually in a country with less than 2 cm of precipitation a year. The result is a stew of excreted antibiotics, pharmaceuticals, microorganisms, pesticides, combustion products, other organic compounds, contaminant breakdown products, and silt deposited on the floodplain. Chemical contaminants sorbed to the small, dry soil particles could be advected into the atmosphere by strong convective storms and transported thousands of kilometers. Known or suspected compounds that are toxic, carcinogenic, or mutagenic, or that act as endocrine disruptors, may be carried sorbed on dust particles or in the gas phase in the dust air mass (Bidleman 1999). These chemical contaminants may adversely affect the marine and terrestrial ecosystems on which the contaminants are deposited (e.g., Solomon 1998, Daughton and Ternes 1999). Similar source considerations exist in Asian dust regions. In addition, Asian dust air masses move across areas, especially China, Korea, and Japan, where chemical emissions have increased substantially in recent years because of extensive construction of major chemical production facilities and increased use of industrial and agricultural chemicals. The aquatic toxicology of numerous chemical contaminants, such as PAHs, organochlorines (e.g., DDT), and metals, has been studied extensively. These compounds have been found to act as neurotoxins, narcotics, endocrine disruptors, mutagens, carcinogens, or disruptors of growth and reproductive processes (Rand 1995). Concentrations as low as a few parts per trillion of some contaminants have been shown to produce significant effects in marine organisms and humans. The effects of atmospherically transported and deposited chemical contaminants on coral reefs at the molecular, cellular, organismal, and system levels are essentially unknown. We propose four questions for investigation: (1) Are chemical contaminants transported with African or Asian dust impairing immune response or reproduction in coral reef organisms? (2) Are these contaminants interfering with the
478 BioScience May 2003 / Vol. 53 No. 5

ability of coral larvae to settle and grow, or with the survival of young corals? (3) Are effects at the cellular or molecular level affecting metabolism, reproduction, calcification, or immunocompetency? (4) What synergistic, cumulative, or threshold effects are at work? The implications of atmospheric transport and deposition of chemical contaminants on coral reefs and other ecosystems may be substantial.

Conclusions
We suggest that African and Asian dust air masses transport chemical and viable microbial contaminants to downwind ecosystems in the Americas, the Caribbean, and the northern Pacific and may be adversely affecting those ecosystems. Atmospheric flux of pathogenic microorganisms may be responsible for the widespread distribution of some diseases occurring on coral reefs and associated habitats. Chemical contaminants in dust air masses may alter the resistance of coral reef organisms to disease pathogens, affect reproduction or survival of larvae, interfere with calcification, or act as toxins, initiating a cascade of effects. Episodic pulses of micro- and macronutrients are known to initiate phytoplankton blooms and may similarly trigger pathogen reservoirs or act to sustain the shift from coral- to algae-dominated reefs. Threshold effects may be at work. Human health may also be adversely affected, primarily by inhalation of known or suspected components in dust events, including nonpathogenic and pathogenic viable microorganisms; chemical contaminants such as carcinogens, toxins, endocrine disruptors, and toxic metals; and small particles that may trigger other physiological reactions (e.g., asthma, cardiovascular events). To understand the effects of these global-scale transport systems on ecosystems and their organisms, collaborative teams of researchers from diverse disciplines will be needed to elucidate the linkages of the biotic, chemical, and physical factors at spatial scales from molecular to global.

Acknowledgments
The authors thank the US Geological Survey, the National Aeronautics and Space Administration, the US Embassy in the Republic of Mali, the Friends of Virgin Islands National Park, and the American International School of Bamako (Mali, West Africa) for providing funding and in-kind support for this project. Special thanks go to Ray Smith and Mark Ranneberger for sample collection and to Scott Carr and Pat Chavez for their innovative ideas. We also thank our colleagues for hours of stimulating discussions and the reviewers for their constructive comments.

References cited
Antonius A. 1973. New observations on coral destruction in reefs. 10th Meeting of the Association of Island Marine Laboratories of the Caribbean 10: 3. Arimoto R, Duce RA, Ray BJ, Ellis WG Jr, Cullen JD, Merrill JT. 1995. Trace elements in the atmosphere over the North Atlantic. Journal of Geophysical Research 100: 11991213. Arimoto R, Snow JA, Graustein WC, Moody JL, Ray BJ, Duce RA, Turekian KK, Maring HB. 1999. Influences of atmospheric transport pathways on

Articles
radionuclide activities in aerosol particles from over the North Atlantic. Journal of Geophysical Research 104: 21,30121,316. Atlas EL, Schauffler S. 1990. Concentration and variation of trace organic compounds in the North Pacific atmosphere. Pages 161184 in Kurtz DA, ed. Long Range Transport of Pesticides. Chelsea (MI): Lewis Publishers. Barnard LA, MacIntyre IG, Pierce JW. 1974. Possible environmental index in tropical coral reefs. Nature 252: 219220. Behrenfeld MJ, Kolber ZS. 1999. Widespread iron limitation of phytoplankton in the South Pacific Ocean. Science 283: 840843. Bidleman TF. 1999. Atmospheric transport and air-surface exchange of pesticides. Water, Air and Soil Pollution 115: 115166. Bidleman TF, Patton GW, Walla MD, Hargrave BT,Vass WP, Erickson P, Fowler B, Scott V, Gregor DJ. 1989. Toxaphene and other organochlorines in Arctic Ocean fauna: Evidence for atmospheric delivery. Arctic 42: 307313. Bishop JKB, Davis RE, Sherman JT. 2002. Robotic observations of dust storm enhancement of carbon biomass in the North Pacific. Science 298: 817821. Blanchard DC, Syzdek L. 1970. Mechanism for the water-to-air transfer and concentration of bacteria. Science 170: 626628. Borsheim KY, Bratbak G, Heldal M. 1990. Enumeration and biomass estimation of planktonic bacteria and viruses by transmission electron microscopy. Applied and Environmental Microbiology 56: 352356. Brown JKM, Hovmller MS. 2002. Aerial dispersal of pathogens on the global and continental scales and its impact on plant disease. Science 297: 537541. Butler A. 1998. Acquisition and utilization of transition metal ions by marine organisms. Science 281: 207210. Chadwick OA, Derry LA, Vitousek PM, Huebert BJ, Hedin LO. 1999. Changing sources of nutrients during four million years of ecosystem development. Nature 397: 491497. Christensen LS, Mortensen S, Botner A, Strandbygaard BS, Ronsholt L, Henriksen CA, Anderson JB. 1993. Further evidence of long distance airborne transmission of Aujeszkys disease (pseudorabies) virus.Veterinary Record 132: 317321. Coale KH, et al. 1996. A massive phytoplankton bloom induced by an ecosystem-scale iron fertilization experiment in the equatorial Pacific Ocean. Nature 383: 495501. Darwin C. 1846. An account of the fine dust which often falls on vessels in the Atlantic Ocean. Quarterly Journal of the Geological Society of London 2: 2630. Daughton CG, Ternes TA. 1999. Pharmaceuticals and personal care products in the environment: Agents of subtle change? Environmental Health Perspectives 107: 907938. Delaney AC, Parkin DW, Griffin JJ, Goldberg ED, Reimann BEF. 1967. Airborne dust collected at Barbados. Geochimica et Cosmochimica Acta 31: 885909. Denner EBM, Smith G, Busse HJ, Schumann P, Narzt T, Polson SW, Lubitz W, Richardson LL. 2002. Aurantimonas coralicida gen. nov., sp. nov., the causative agent of white plague type II on Caribbean scleractinian corals. International Journal of Systematic and Evolutionary Microbiology. (1 April 2003; http://dx.doi.org/10.1099/ijs.0.02359-0) Duce RA, Tindale NW. 1991. Atmospheric transport of iron and its deposition in the ocean. Limnology and Oceanography 36: 17151726. Duce RA, Ray BJ, Hoffman JL, Walsh PR. 1976. Trace metal concentration as a function of particle size in marine aerosols from Bermuda. Geophysical Research Letters 23: 339342. Duce RA, Unni CK, Ray BJ, Prospero JM, Merrill JT. 1980. Long-range atmospheric transport of soil dust from Asia to the tropical North Pacific: Temporal variability. Science 209: 15221524. Duce RA, et al. 1991. The atmospheric input of trace species to the world ocean. Global Biogeochemical Cycles 5: 193259. Edmunds PJ. 1991. Extent and effect of black band disease on Caribbean reefs. Coral Reefs 10: 161165. . 2000. Recruitment of scleractinians onto the skeletons of corals killed by black band disease. Coral Reefs 19: 6974. Eilers H, Pernthaler J, Glockner FO, Amann R. 2000. Culturability and in situ abundance of pelagic bacteria from the North Sea. Applied and Environmental Microbiology 66: 30443051. Fiser A, Lanikova A, Novak P. 1994. Mold and microbial contamination of dust deposition in cowsheds for heifers and dairy cows. Veterinary MedicineCzech 39: 245253. Geiser DM, Taylor JW, Ritchie KB, Smith GW. 1998. Cause of sea fan death in the West Indies. Nature 394: 137138. Giesy JP, Kannan K. 2001. Global distribution of perfluorooctane sulfonate in wildlife. Environmental Science and Technology 35: 13391342. Gillies JA, Nickling WG. 1996. Dust concentrations and particle-size characteristics of an intense dust haze event: Inland delta region, Mali, West Africa. Atmospheric Environment 30: 10811090. Graedel TE, Weschler CJ, Mandich ML. 1986. Kinetic studies of atmospheric droplet chemistry, 2: Homogeneous transition metal chemistry in raindrops. Journal of Geophysical Research 91: 52055221. Griffin DW, Garrison VH, Herman JR, Shinn EA. 2001. African desert dust in the Caribbean atmosphere: Microbiology and public health. Aerobiologia 17: 203213. Hayes ML, Bonaventura J, Mitchell TP, Prospero JM, Shinn EA, Van Dolah F, Barber RT. 2001. How are climate and marine biological outbreaks functionally linked? Hydrobiologia 460: 213220. Herman JR, Krotkov N, Celarier E, Larko D, Labow G. 1999. The distribution of UV radiation at the Earths surface from TOMS measured UVbackscattered radiances. Journal of Geophysical Research 104: 12,05912,076. Husar R, et al. 2001. Asian dust events of 1998. Journal of Geophysical Research 106: 18,31718,330. Jickells TD. 1999. The inputs of dust derived elements to the Sargasso Sea: A synthesis. Marine Chemistry 68: 514. Kolpin DW, Furlong ET, Meyer MT, Thurman EM, Zaugg SD, Barber LB, Buxton HT. 2002. Pharmaceuticals, hormones, and other organic wastewater contaminants in U.S. streams, 19992000: A national reconnaissance. Environmental Science and Technology 36: 12021211. Kushmaro A, Banin E, Stackebrandt E, Rosenberg E. 2001. Vibrio shiloi sp. nov: The causative agent of bleaching of the coral Oculina patagonica. International Journal of Microbiology and Systematic Evolution 51: 13831388. Lemieux PM, Lutes CC, Abbott JA, Aldous KM. 2000. Emissions of polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans from the open burning of household waste in barrels. Environmental Science and Technology 34: 377385. Lenes JM, et al. 2001. Iron fertilization and the Trichodesmium response on the West Florida shelf. Limnology and Oceanography 46: 12611277. Letellier A, Messier S, Pare J, Menard J, Quessy S. 1999. Distribution of Salmonella in swine herds in Quebec. Veterinary Microbiology 67: 299306. Lipnick RL, Jansson B, Mackay D, Petreas M, eds. 2001. Persistent, Bioaccumulative, and Toxic Chemicals II: Assessment and New Chemicals. Washington (DC): American Chemical Society. Looser R, Froescheis O, Cailliet GM, Jarman WM, Ballschmiter K. 2000. The deep-sea as a final global sink of semivolatile persistent organic pollutants? Part II: organochlorine pesticides in surface and deep-sea dwelling fish of the North and South Atlantic and the Monterey Bay Canyon (California). Chemosphere 40: 661670. MacDonald RW, et al. 2000. Contaminants in the Canadian Arctic: 5 years of progress in understanding sources, occurrence and pathways. Science of the Total Environment 254: 93234. Majewski MS, Capel PD. 1995. Pesticides in the Atmosphere: Distribution, Trends, and Governing Factors. Pesticides in the Hydrologic System, Vol. 1. Chelsea (MI): Ann Arbor Press. Martin JH, Fitzwater SE. 1988. Iron deficiency limits to phytoplankton growth in the northeast Pacific subarctic. Nature 331: 341343. Maynard NG. 1976. Relationship between diatoms in surface sediments of the Atlantic Ocean and the biological and physical oceanography of overlying waters. Paleobiology 2: 99121. McTainsh GH, Nickling WG, Lynch AW. 1997. Dust deposition and particle size in Mali, West Africa. Catena 29: 307322.

May 2003 / Vol. 53 No. 5 BioScience 479

Articles
Merman EA. 2001. Atmospheric inputs to the tropical oceanunlocking the record in annually banded corals. Masters thesis. University of South Florida, St. Petersburg. Moulin C, Lambert CE, Dulac F, Dayan U. 1997. Control of atmospheric export of dust from North Africa by the North Atlantic Oscillation. Nature 387: 691694. Muhs DR, Bush CA, Stewart KC, Rowland TR, Crittenden RC. 1990. Geochemical evidence of Saharan dust parent material for soils developed on Quaternary limestones of Caribbean and Western Atlantic islands. Quaternary Research 33: 157177. Patterson KL, Porter JW, Ritchie KB, Polson SW, Mueller E, Peters EC, Santavy DL, Smith GW. 2002. The etiology of white pox, a lethal disease of the Caribbean elkhorn coral, Acropora palmata. Proceedings of the National Academy of Sciences 99: 87258730. Perry KD, Cahill TA, Eldred RA, Dutcher DD. 1997. Long-range transport of North African dust to the eastern United States. Journal of Geophysical Research 102: 11,22511,238. Prospero JM. 1968. Atmospheric dust studies on Barbados. Bulletin of the American Meteorological Society 49: 645652. . 1981. Arid regions as sources of mineral aerosols in the marine atmosphere. Geological Society of America Special Paper 186: 7186. Prospero JM, Nees RT. 1977. Dust concentration in the atmosphere of the equatorial North Atlantic: Possible relationship to the Sahelian drought. Science 196: 11961198. . 1986. Impact of the North African drought and El Nio on mineral dust in the Barbados trade winds. Nature 320: 735738. Prospero JM, Glaccuum RA, Nees RT. 1981. Atmospheric transport of soil dust from Africa to South America. Nature 289: 570572. Rand GM, ed. 1995. Fundamentals of Aquatic Toxicology: Effects, Environmental Fate and Risk Assessment. 2nd ed. Washington (DC): Taylor and Francis. Ravishankara AR. 1997. Heterogeneous and multiphase chemistry in the troposphere. Science 276: 10581065. Richardson LL. 1997. Occurrence of the black band disease cyanobacterium on healthy corals of the Florida Keys. Bulletin of Marine Science 61: 485490. . 1998. Coral diseases: What is really known? Trends in Ecology and Evolution 13: 438443. Richardson LL, Aronson R. 2002. Infectious diseases of reef corals. Pages 12251230 in Kasim Moosa MK, Soemodihardjo S, Nontji A, Soegiarto A, Romimohtarto K, Sukarno, Suharsono, eds. Proceedings of the Ninth International Coral Reef Symposium; 2327 October, Bali, Indonesia. Bali (Indonesia): Indonesian Institute of Sciences and State Ministry of Education. Richardson L, Goldberg W, Kuta K, Aronson R, Smith G, Ritchie K, Halas J, Feingold J, Miller S. 1998. Floridas mystery coral killer explained. Nature 392: 557558. Ritchie M, Pedgley D. 1989. Desert locusts cross the Atlantic. Atenna 13: 1012. Rosenfeld D, Rudich Y, Lahav R. 2001. Desert dust suppressing precipitation: A possible desertification feedback loop. Proceedings of the National Academy of Sciences 98: 59755980. Rtzler K, Santavy D. 1983. The black band disease of Atlantic coral reefs. I. Description of the cyanophyte pathogen. Marine EcologyPubblicazione della Stazione Zoologica di Napoli II 4: 301319. Saydam AC, Senyuva HZ. 2002. Deserts: Can they be potential suppliers of bioavailable iron? Geophysical Research Letters 29: 19.119.3. Shen GT, Boyle EA. 1987. Lead in corals: Reconstruction of historical industrial fluxes to the surface ocean. Earth and Planetary Science Letters 82: 289304. Shinn EA, Smith GW, Prospero JM, Betzer P, Hayes ML, Garrison VH, Barber RT. 2000. African dust and the demise of Caribbean coral reefs. Geological Research Letters 27: 30293032. Simoneit BRT, Cox RE, Standley LJ. 1988. Organic matter of the troposphere: IV. Lipids in the Harmattan aerosols of Nigeria. Atmospheric Environment 22: 9831004. Sinclair D, Kinsley L, McCulloch M. 1998. High resolution analysis of trace elements in corals by laser ablation ICP-MS. Geochimica et Cosmochimica Acta 212: 18891901. Smith GW, Ives LD, Nagelkerken IA, Ritchie KB. 1996. Caribbean sea-fan mortalities. Nature 383: 487. Solomon KR. 1998. Endocrine-modulating substances in the environment: The wildlife connection. International Journal of Toxicology 17: 159171. Swap R, Garstang S, Greco S, Talbot R, Kallberg P. 1992. Saharan dust in the Amazon basin. Tellus 44: 133149. Swap R, Ulanski S, Cobbett M, Garstang M. 1996. Temporal and spatial characteristics of Saharan dust outbreaks. Journal of Geophysical Research 101: 42054220. Syvitski JPM, Lewis AG. 1980. Sediment ingestion by Tigriopus californicus and other zooplankton: Mineral transformation and sedimentological considerations. Journal of Sedimentary Petrology 50: 869880. Talbot RW, Harris RC, Browell EV, Gregory GL, Sebacher DI, Beck SM. 1986. Distribution and geochemistry of aerosols in the tropical north Atlantic troposphere: Relationship to Saharan dust. Journal of Geophysical Research 91: 51735182. Tegen I, Lacis AA, Fung I. 1996. The influence on climate forcing of mineral dust disturbed soils. Nature 380: 419422. Torsvik V, Goksoyr J, Daae FL. 1990. High diversity in DNA of soil bacteria. Applied and Environmental Microbiology 56: 782787. Turner SM, Nightingale PD, Spokes LJ, Liddicoat MI, Liss PS. 1996. Increased dimethyl sulphide concentrations in sea water from in situ iron enrichment. Nature 383: 513517. Uematsu M, Sugita T, Anikiev VV, Medvedev AN. 1992. Large-scale transport of pollution aerosol over the east coast of Asia. Geophysical Research Letters 19: 22192221. van der Valk H, Diarra A. 2000. Pesticide use and management in the African Sahelan overview. Sahelian Studies and Research 45: 1327. van Dijk HFG, Guicherit R. 1999. Atmospheric dispersion of current-use pesticides: A review of the evidence from monitoring studies. Water, Air and Soil Pollution 115: 2170. van Pul WAJ, Bidleman TF, Brorstrom-Lunden E, Builtjes PJH, Dutchak S, Duyzer JH, Gryning SE, Jones KC, van Dijk HFG, van Jaarsveld JA. 1999. Atmospheric transport and deposition of pesticides: An assessment of current knowledge. Water, Air and Soil Pollution 115: 245256. Venkatesh MV, Joshi KR, Harjai SC, Ramdeo IN. 1975. Aspergillosis in desert locust (Schistocerca gregaria Forsk). Mycopathologia 57: 135138. Walsh JJ, Steidinger KA. 2001. Saharan dust and Florida red tides: The cyanophyte connection. Journal of Geophysical Research 106: 11,59711,612. Weinberg ED. 1996. Acquisition of iron and other nutrients in vivo. Pages 7994 in Roth J, et al., eds. Virulence Mechanisms of Bacterial Pathogens. Washington (DC): American Society for Microbiology. Weir JR, Garrison VH, Shinn EA, Smith GW. 2000. The relationship between Gorgonian coral (Cnidaria: Gorgonacea) diseases and African dust events. Abstracts of the Ninth International Coral Reef Symposium, Bali, Indonesia, October 2327, 2000: 78. Young RW, et al. 1991. Atmospheric iron inputs and primary productivity: Phytoplankton responses in the north Pacific. Global Biochemical Cycles 5: 119134.

480 BioScience May 2003 / Vol. 53 No. 5