Final Report: CISNet: Coral Bleaching, UV Effects, and Multiple Stressors in the Florida Keys

EPA Grant Number: R826939
Title: CISNet: Coral Bleaching, UV Effects, and Multiple Stressors in the Florida Keys
Investigators: Anderson, Susan L. , Cherr, Gary N. , Brown, Heather , Machula, Jana , Hansen, Lara , Oliver, Leah , Zepp, Richard , Jackson, Susan
Institution: University of California - Davis , Bodega Marine Laboratory , U. S. Environmental Protection Agency
EPA Project Officer: Packard, Benjamin H
Project Period: October 1, 1998 through September 30, 2001
Project Amount: $407,567
RFA: Ecological Effects of Environmental Stressors Using Coastal Intensive Sites (1998) RFA Text |  Recipients Lists
Research Category: Environmental Statistics , Ecosystems , Ecological Indicators/Assessment/Restoration


The overall objective of this research project was to evaluate the role that climate change may play in altering the interactions of ultraviolet (UV) radiation with coral reefs and potentially contributing to coral bleaching. This entails examination of both UV-specific stresses on corals and factors that influence UV penetration over the reefs. The specific objectives of this research project were to: (1) develop immunofluorescence techniques to examine UV-specific lesions in DNA (thymine dimers) of coral and zooxanthellae; (2) determine whether UV-induced DNA damage and indices of coral bleaching are correlated, including laboratory and field assessments of photoprotective pigment concentrations; and (3) understand and quantify factors that affect UV exposure of coral reefs and how climate changes are affecting UV exposure of the reefs.

Conceptual Model and Study Site. To describe the complex processes under investigation in this study, a conceptual model was developed (see Figure 1) that describes the relationships we postulated between climate change and increases in UV-B over coral reefs. In addition, possible mechanisms of UV-induced stress in corals and their relationship to detrimental effects associated with increasing sea surface temperatures were expressed. We posit that increasing sea surface temperatures can result in thermal stratification at many locales. This could result in increased photobleaching of chromophoric dissolved organic matter (CDOM). As the CDOM photobleaches, the UV-B penetration should increase. If UV-B irradiance goes up, then the potential for UV-B induced DNA damage in both coral and symbiotic zooxanthellae increases. Factors that contribute to protecting the coral from such damages include pigment protection, induction of mycosporine amino acids, and the induction of DNA repair enzymes. If a significant increase in DNA damage is incurred, then it may be an additive stressor on the corals. Recent research indicates that warming sea surface temperatures can reduce the ability of photosynthetic systems in zooxanthellae to withstand the damaging effects of increased light exposure on the proteins and DNA. In this study, we applied a state-of-the-art combination of field and laboratory studies to determine whether UV may increase over reefs under stratified conditions and to examine what UV irradiance conditions contribute to increases in DNA damage. Other relationships in the conceptual model also were examined. For example, we explored the sources of CDOM to the reefs. We also explored the topic of pigment protection and critically evaluated current indicators of coral bleaching.

Figure 1. Conceptual Model of Our Research Project Investigating the Potential Role of UV Exposure in Coral Bleaching

All investigators in this project worked together at a common study site in the Florida Keys (see Figure 2). Coral biology investigations were conducted at Maryland Shoals reef in the Middle Keys, and studies to quantify factors that affect UV exposure of coral reefs were conducted at Maryland Shoals, as well as at additional sites in the Middle and Lower Florida Keys.

Figure 2. Map Illustrating Locations of Sites Used in This Study

Summary/Accomplishments (Outputs/Outcomes):

Thymine Dimer Technique Development. We developed a powerful technique for evaluating thymine dimers in any purified DNA sample, including coral and zooxanthellae. The technique is a 96-well ELISA-type assay that utilizes a plate reader capable of detecting UV fluorescence and luminescence. By adhering known amounts of DNA to microplates, we can perform an immunoassay with chemiluminescent substrate that is directly quantified by the plate reader. Not only does this release us from problems of repeatability with the dot blot procedure, but it eliminates a time-consuming image analysis step. The plate reader quantifies each sample in relative luminescence units. The assay we have developed utilizes a commercially available thymine dimer antibody; and hence, is a more universally applicable technique than other available techniques. This methodology is an exciting new development, and we believe it can be utilized by researchers working on UV-B effects in other species, such as amphibians, fish, invertebrates, and even human skin tissues.

The final protocol for the experimental data includes seven basic steps: (1) DNA extraction; (2) DNA denaturation; (3) incubation with a binding solution; (4) adhering DNA to the 96-well plate; (5) blocking and incubation with the antibody; (6) developing, detecting, and imaging chemiluminescent signal; and (7) data analysis. For all analyses, tissue was airbrushed from the coral skeleton with Ca2+-free SW and frozen at -70ºC in separate aliquots for thymine dimer analysis, zooxanthellae counts, protein assay, and pigment analysis. Further details of the methodology are available from the authors.

We also finalized methods for immunolocalization of thymine dimer and coral tissue. This aspect of our work is essential for estimating the proportion of the thymine dimers that are in the coral as opposed to the zooxanthellae tissue; hence, this analysis will help us to interpret our data because the number of zooxanthellae per coral polyp changes during bleaching events. To correct for high levels of autofluorescence observed in Porites porites tissue and zooxanthellae at all wavelengths, we have worked out a method of image subtraction leaving only fluorescence given from thymine dimer labeling. At the wavelength of 350 nm, we observed the least amount of autofluorescence that leads us to use an Alexa 350 nm conjugate. Images then were collected at the wavelengths of 488 and 350 nm. The 488 nm image was used as background and subtracted from the 350 nm image, leaving only the labeled thymine dimer fluorescence. These images then were overlayed on the total 350 nm image, allowing us to analyze the number of fluorescent zooxanthellae and to quantitate the relative fluorescent values per zooxanthellae.

To evaluate the distribution of thymine dimers in the coral tissue versus the zooxanthellae, we irradiated P. porites in an Atlas Suntest CPS Solar Simulator and examined dimers in mixed (coral and zooxanthellae) tissue obtained by airbrushing the coral after irradiation. We also quantified the proportion of zooxanthellae that exhibited dimers at 0, 2, 4, and 6 hours after irradiation. The simulated light spectrum for the solar simulator was similar, although not identical to, solar radiation at Eastern Sambo Reef. We determined that intact coral tissue exhibited few thymine dimers, yet immunolocalization of thymine dimers in zooxanthellae was relatively intense. We also observed that control zooxanthellae contained dimers in less than 20 percent of the cells, while the 2-hour exposure samples show dimers in 65 percent of the cells, with decreased levels seen in the 4- and 6-hour exposures at 30 and 40 percent, respectively.

In 2001, laboratory exposures of P. porites in the Atlas Suntest CPS Solar Simulator were repeated to assess the variation in thymine dimers over time using our newly developed plate-based assay. Coral were collected at Maryland Shoals and maintained overnight in an outdoor aquarium covered with 60 percent shade cloth. The next day, coral nubbins were placed in a water-jacketed beaker at 29.5°C with five replicates per time point. Times of irradiation were 0, 1, 2, and 4 hours. Following irradiation, nubbins were immediately frozen in liquid nitrogen and later processed as described above for the enumeration of thymine dimers. Thymine dimers were significantly elevated (One-Way ANOVA, p=0.08, followed by Tukey a posteriori test p<0.05) only in the 1-hour exposures.

The solar simulator studies, using both immunolocalization and plate-based immunoassay, indicate that significant differences in thymine dimers can be induced and detected, but that effects are not proportional to time of exposure. Further evaluation using shorter exposures at increasing doses or to the more damaging shorter wavelengths would enhance our characterization of this response.

In summary, we have developed a more sensitive and widely applicable technique to measure thymine dimers than has been previously available in the scientific literature. We have developed immunolocalization procedures that overcome numerous technical hurdles, and these indicate that thymine dimers are located in the zooxanthellae. Our dose response studies in the laboratory indicate that thymine dimers can be induced by UV exposure but that further study is required to characterize the kinetics of DNA damage and repair, as well as effects at various wavelengths.

Part 1: Field Variation in Thymine Dimers and Pigments. Our goal was to determine whether diurnal variation in thymine dimers occurs under normal irradiance conditions in coral tissue. Although there has been considerable interest in the effects of UV irradiation on corals, the induction of DNA photoproducts in corals in the field has not been previously studied. For fish, data indicate that in some species and lifestages, DNA damage is proportional to exposure but not to cumulative dose. One study has been conducted using microbial communities living on the surface of corals, and diurnal variation in induction of DNA photoproducts that varied among species of corals sampled, time of day, and date of sampling was observed.

Using the sensitive immunoassay described above, we sampled the coral P. porites at Maryland Shoals Reef in the Middle Florida Keys to assess diurnal variation in thymine dimers under normal summer irradiance conditions. Field sampling was conducted at three colonies over multiple timepoints to discern factors contributing to variation in thymine dimers. This site was selected because P. porites was abundant at this site, and colonies were located at nearly identical depths, making UV-B exposure history as similar as possible. The three colonies we sampled were located within 200 meters of one another and were between 4- and 5-m depth. The depths, locations, and dimensions of each colony were carefully described.

Five replicate nubbins were randomly sampled from each colony at each timepoint. Coral nubbins (3- to 4-cm length) were removed using bone cutters and immediately placed in Whirlpak bags that were placed within opaque black nylon bags. These were returned to the boat within approximately 10 minutes, maintained in the dark, and frozen in liquid nitrogen within approximately 15 minutes. Sampling timepoints for the largest colony (colony 2) were 0800, 1030, 1300, 1430, and 1800 hours. For the two smaller colonies, sampling times were 0800, 1300, and 1800 hours. Other data such as enumeration of zooxanthellae, tissue protein, chlorophyll, and carotenoid pigments were obtained to characterize the condition of each colony and nubbin, as well as to describe factors that may covary with dimer levels.

Irradiance measurements were obtained using a Satlantic OCI-200 sensor that was horizontally mounted in close proximity to the colonies. The seven-channel optical sensor was cosine corrected and measured downwelling irradiance at 305.3, 324.9, 339.7, 379.6, 442.8, and 554.6 nm. At the same time, on board the boat above-water irradiance was measured using a seven-channel reference sensor that was mounted in an open area. The sky was cloud-free during the experiments and the standard deviation for the UV irradiance data (305-380 nm) about the mean was less than 2.5 percent in all cases.

A comparison of UV and visible exposures for the three colonies indicates that colony 2 received higher light exposure than did colonies 3 and 4, especially in the UV-B region (305 nm). For colony 2, a significant difference in the frequency of thymine dimers was detected, but effects did not directly vary with UV doses at the timepoint of sampling. Rather, dimers only were significantly elevated (One-Way ANOVA p<0.001, followed by Tukey a posteriori test p<0.05) in the morning, then declined midday. For colonies 3 and 4, no significant differences (One-Way ANOVA colony 3, p=0.898; colony 4, p=0.702) in thymine dimers were observed. The fact that significant differences were observed in colony 2 but not in colonies 3 and 4 may be attributable to the greater UV-B exposure this colony experienced.

Our data indicate that levels of induction of thymine dimers are very low in corals, under conditions of natural solar irradiance. In addition, a significant increase was only observed at the early morning timepoint. Although relatively few aquatic species have been studied under controlled conditions in the field, these data are unusual but not unprecedented. In the previously mentioned investigation using coral surface microbes, significant elevations of thymine dimers at 0600 hours were observed, and a protective role of the coral surface microbes in protecting coral against UVR was hypothesized. Variation in the pattern of induction of thymine dimers on multiple days would have added to the scope of this investigation. However, because of stringent limitations on coral collections, we could not further sample these colonies to determine how thymine dimers varied on multiple days. Results of this study indicate that coral DNA is relatively well protected against the effects of UVR. We have not determined whether the protection is associated with DNA repair or presence of protective pigments or both.

Part 2: Laboratory Variation in Thymine Dimers, Zooxanthellae, and Pigments and Their Relationship to Coral Bleaching. The role of ultraviolet radiation in inducing coral bleaching is not well understood, despite the fact that several related field studies have been performed. To date, these studies have demonstrated that increasing UV dose, achieved by moving coral within the water column, can result in increased bleaching. Additionally, studies in the field have demonstrated that corals contain compounds such as mycosporine-like amino acids (MAA), which are known to protect organisms from UV damage. MAA concentrations in corals have been positively correlated with natural levels of UV so that organisms may regulate concentrations of UV-absorbing compounds in proportion to the amount of UV radiation they experience in the environment. In addition, some carotenoid pigments act as antioxidants or otherwise block or ameliorate the effects of UV radiation, but this effect has not been investigated specifically in corals. Finally, DNA damage associated with UVR has been characterized in selected aquatic species, although never before in coral. To date, however, no laboratory studies have been conducted to elucidate the relationship between all of these parameters, UVR, and coral bleaching. Furthermore, no studies have attempted to take these results from the field and further explore the changes that occur in corals as bleaching occurs.

This portion of the study was a comparison of responses between coral species from various geographical locations (Florida, Virgin Islands, and Hawaii) and genera. All experiments were performed in a common solar simulator at the U.S. Environmental Protection Agency (EPA) Gulf Ecology Division from May 2000 to April 2001. A spectroradiometer (Model 754 Optronics, Orlando, FL) was used to measure the spectral output of the simulator at 1 nm intervals from 280 to 800 nm. In these experiments, individual coral nubbins (3-5 cm in height) were placed in 450 mL of artificial sea water (Instant Ocean at ambient salinity) in individual 500 mL plastic containers, with four containers (replicates) per treatment level. Each experiment was terminated when UV-exposed corals began to bleach. Methods of tissue preparation varied slightly among experiments, but in general, coral were frozen at -70°C and later airbrushed with tissue subsequently preserved in several aliquots for analysis.

All exposures resulted in some degree of bleaching in the high UV treatments. The degree of bleaching and responses of other parameters varied with species and population. Below, we present the results as treatment averages. Treatments A through D represent increasing UV doses. However, details regarding detailed differences in wavelength distributions and weighted doses are available from the authors. There was no significant variation among treatments for levels of tissue protein for any species tested.

Florida P. porites (FL-1). In this experiment, bleaching increased and polyp condition decreased with increasing UV dose. A significant decrease in zooxanthellae count (p=0.053) also was observed. For thymine dimers, treatment B had a lower number of dimers per Mb DNA than the other treatments, but the response was not statistically significant because of the large sample variability. When pigment data are expressed as pigment concentration per zooxanthellae, the D treatment had the highest pigment concentration and was found to be statistically different from treatment B for all pigments (Chl c p=0.046; peridinin p=0.071; diadionoxanthin p=0.063; Chl a p=0.048), as well as when pigments are totaled (p=0.043). When data are expressed as pigment per polyp, there is no difference between treatments with regard to any of the pigments analyzed.

Florida Madracis mirabilis (FL-2). Generally, bleaching increased and polyp condition decreased with increasing UV dose. For this species, there was no significant difference in zooxanthellae count despite the visual observations of bleaching. There was no significant increase in thymine dimers with increasing dose. There were no significant differences in pigment content among treatments.

Virgin Islands Porites porites (VIPP). Bleaching increased and polyp condition decreased with increasing UV dose. Treatment A has the highest zooxanthellae count, and it is nominally statistically greater than both treatments C and D (p=0.053). Again, treatment B elicited a small but not significant decrease in thymine dimers. Pigment concentration, expressed as pigment per polyp, decreases as UV dose increases. For diadinoxanthin, treatment A is different from treatment C (p=0.037). In contrast, there was no significant difference in pigment concentrations when pigments are expressed as pigment per zooxanthellae.

Hawaii P. compressa (HI-1). In this experiment, bleaching increased and polyp condition decreased with increasing UV dose. No significant changes in zooxanthellae count or thymine dimers were observed for this species. No significant differences in pigment content were observed.

Hawaii Montipora capitata (HI-2). Generally, bleaching increased and polyp condition decreased with increasing UV dose. Contrary to expectation, zooxanthellae count significantly increased (p=0.037) with treatment D, resulting in higher counts than treatment A. No significant differences in thymine were observed among treatments. When data are expressed as pigment per zooxanthellae, treatment B had the highest pigment concentration. For peridinin, treatments B and A both were found to be different from both C and D (p=0.0025). For chlorophyll a, treatment B is greater than D (p=0.0333). For diadinoxanthin, treatment A is greater than D (p=0.0248). When concentrations are expressed as pigment per polyp, no significant differences were observed.

The lack of clear trends in the analysis of the treatment means has led us to examine the data further. Because of the high variability of the responses between and within species, the analysis of these data are ongoing. Currently, we are exploring the data on a nubbin-by-nubbin basis to examine the response of highly bleached versus unbleached individual nubbins.

In summary, we have determined that simulated solar irradiance with increasing UVR induces bleaching in a dose-related manner. However, our findings illustrate that indicators of bleaching, such as zooxanthellae counts and pigment change, do not exhibit clearcut responses. Variation is observed between species and endpoints. In two out of five cases, zooxanthellae counts significantly declined with increasing dose; but in one experiment, the zooxanthellae counts actually increased. This effect could be because of the shallower distribution of the latter species (M. capitata). Relationships between dose, zooxanthellae count, and pigment concentration are being subject to additional analyses on a nubbin-by-nubbin basis and these will be included in the final manuscript. This manuscript will critically examine these endpoints as indicators of coral bleaching. Finally, we did not observe a dose-dependent increase in thymine dimers. We speculate that this may be related to the fact that exposures lasted for 96 hours, and coral were able to acclimate to these doses and repair damaged DNA or induce protective pigments. These data indicate that DNA damage is not likely to contribute significantly to the bleaching associated with increasing UVR unless other co-occurring environmental changes, such as increases in temperature, reduce repair efficiencies or pigment synthesis. It is worth noting that a small decrease in dimers was observed in five of six experiments for treatment B and, although the response was not significant, it may indicate that the ambient dose is the most favorable condition for DNA repair.

UV Exposure and CDOM Concentrations. Our next goal was to further examine the relationships between temperature, UV, and coral bleaching in controlled experiments and by observations in the field. This research included measurements of underwater solar irradiance and fluorescence at locations close to the coral reef tract in the Florida Keys; measurements of optical properties of water samples collected at these sites; development of algorithms that describe the penetration of UV radiation into the Florida Keys water; and initial laboratory and field research on sources and sinks of UV-absorbing substances, principally CDOM, that control UV exposure of the corals. The location of the sites used for these studies is shown in Figure 2.

Factors Affecting Coral's UV Exposure. Our research provided evidence that UV exposure of coral reefs in the Florida Keys is controlled by the chromophoric CDOM in waters overlying the reefs. Downwelling vertical profiles of UV and visible radiation generally were exponential and diffuse attenuation coefficients [Kd(lambda)] were computed from the data using exponential regressions. Diffuse attenuation coefficients were determined at sites located at the Upper, Middle, and Lower Keys and the Dry Tortugas. Water samples were collected concurrently for additional laboratory studies. At some sites, fluorescence also was measured using fluorometers that were specifically designed to detect CDOM. After filtration (0.2 µm) to remove algae, bacteria, and other small particulates, the absorption and fluorescence spectra of water samples were measured. Absorption and diffuse attenuation coefficients were highly correlated (r2 > 0.9) in the UV-B (290-315 nm) spectral region and ratios of absorption to diffuse attenuation coefficients were greater than 0.9 throughout this spectral region. This result indicated that CDOM was primarily responsible for controlling the transmission of UV-B radiation in waters over the coral reefs. We also found that CDOM fluorescence was correlated closely with diffuse attenuation coefficients in the UV region, a finding that provides further support to our hypothesis that CDOM plays an important role in controlling exposure of the coral reefs to UV exposure in the Florida Keys.

Absorption coefficients a in the 300 to 500 nm spectral region could be closely described by a non-linear exponential function, aç = aço exp [-S(lambda - lambdao]), where aço is the absorption coefficient at lambdao (i.e., 290 nm) and S is the spectral slope coefficient. S ranged from 0.022-0.026 nm-1 for oligotrophic seawater outside the reefs and 0.017-0.018 nm-1 for the shallow regions close to land, such as Hawk Channel and Florida Bay. Using these algorithms, it should be possible to use satellite measurements of CDOM absorption coefficients in the visible region to estimate UV-B attenuation coefficients and thus exposure of corals to UV at the regional scale. Currently available satellite measurements do not extend into the UV spectral region.

Our results indicated that light exposure in the waters around the Florida Keys strongly varies with time and location. Generally, absorption coefficients increased sharply along south-to-north transects from the deep bluewaters of the Florida Straits into Hawk Channel, the shallow coastal shelf region between the reef tract and the Keys (see Figure 2). The largest change occurred over a narrow region that represented the interface between the green-yellow waters in Hawk Channel and the blue Atlantic water. We also obtained diurnal irradiance data during the field investigations of thymine dimer content in corals located at the Maryland Shoals site in the Florida Keys. Other direct measurements at the corals surfaces indicated that the levels of thymine dimers correlated with the UV-B irradiance reaching the surface.

Our results obtained at the deep stations just south of the coral reefs also indicated that the depth dependence of both the light and temperature greatly differs between the warm summer months and the cold winter months. We found that the upper ocean water close to the coral reefs generally was much more opaque to UV and photosynthetically active radiation (PAR) during the cold winter months than during the summer. During the winter, the temperature is almost uniform in the upper ocean, and depth dependence of the downwelling irradiance is close to exponential. However, during the warm summer months, a much more complex depth dependence of temperature and light develops. The temperature profiles during the summer indicate that the water has stratified (i.e., that it has developed a poorly mixed thermocline that blocks upward transport of cooler, deep waters to the surface layer). The thermocline is the region where temperatures rapidly decrease with depth. The pronounced stratification effect of the water is accompanied by a substantial increase in UV penetration in the surface waters above the thermocline compared to that below it. This is evidenced by a change in the slope of log plots of the irradiance versus depth in the vicinity of the thermocline. We attribute this effect to combined photobleaching and microbial degradation of the CDOM in the upper water column coupled with reduced inputs of cooler, more opaque deep water. The term "photobleaching" refers to the decrease of absorption coefficients of the CDOM in the UV and visible spectral regions in irradiation. Laboratory studies indicated that the CDOM in water samples collected near the reef tract photobleached with decreases in UV absorption coefficients and fluorescence and an increase in spectral slope coefficient when exposed to simulated solar radiation. This result suggests that the extensive stratification that occurs under El Nino/Southern Oscillation (ENSO) conditions may be greatly increasing exposure of the reefs to UV. Additional research is required to confirm this possibility.

CDOM Sources and Sinks. Other studies were conducted to elucidate various sources of UV-absorbing materials in the Florida Keys waters and the response of these sources to changes in climate and solar UV radiation. Both decomposing seagrasses and mangrove leaves were found to be important sources of CDOM to waters in this region. Underwater flux chambers sealed over litter derived from the common seagrass Thalassia testudinum and over T. testudinum beds were used to quantify in situ production of CDOM at sites close to the Looe Key and Maryland Shoals reefs. In laboratory studies, the temperature dependence of the spectral (UV-visible, fluorescence) and molecular mass properties of CDOM produced during the degradation of T. testudinum litter was investigated to determine its contribution to the coastal CDOM pool. A combination of ultraviolet-visible absorption spectroscopy (UV-visible region), fluorescence spectroscopy (excitation-emission matrix [EEM] technique), and ion trap mass spectroscopy was used to directly connect the molecular mass distribution with specific optical properties. Results indicate that degradation of T. testudinum litter may be a significant source of CDOM that has not been previously identified. The DOM released by the degrading seagrass litter has specific absorption coefficients, spectral slope coefficients, fluorescence EEM features, and molecular mass distribution that are similar to those observed for CDOM attributed terrestrial sources. The rate of CDOM production is temperature dependant, with rates more than doubling with a temperature increase from 21.9 ± 0.2°C to 32.3 ± 0.2°C. Other studies showed that the photobleaching of the seagrass-derived CDOM by solar radiation was caused almost completely by its UV component. By considering this production rate, the average residence time of a parcel of water in Florida Bay, and the biomass and turnover rate for T. testudinum, this source was shown to be a potentially significant source of CDOM in Florida Bay, a shallow coastal region located off the coast of South Florida, and in Hawk Channel near the coral reef tract.

Summary. In summary, our studies have shown that the UV exposure of coral reefs in the Florida Keys is highly variable and that the variability is, in turn, linked to climatic changes that are occurring over the region. Warmer temperatures induce a clarification of the water over the reefs and this factor alone causes large increases in light exposure during the summer months compared to winter. The clarification is linked in part to stratification of the ocean water near the reefs; stratification results in concurrent increases in water temperatures, and UV light exposure. This factor likely has contributed to the massive coral bleaching that has occurred during El Nino events. We also have shown that the colored CDOM is largely responsible for UV attenuation over the Florida Keys reefs. Decomposing organic matter from seagrasses and mangroves is a major source of the CDOM in this region. Human activities and climatic changes that adversely impact these biological sources of CDOM can adversely impact corals health by reducing the UV protection that they provide.

General Relationship of Findings to Coral Bleaching. Although there has been considerable interest in the effects of UV irradiation on corals, no previous investigations have considered effects of UVR on coral DNA. Because DNA is a major target of UV effects, this investigation will fill an important gap in the scientific literature. Our findings indicate that thymine dimers can be detected in coral tissue but levels of induction are low compared to other aquatic organisms and are not proportional to dose at the exposure times used. Even elevated UVR doses in the laboratory did not induce significant thymine dimers in multiple species of coral. Thymine dimer induction, therefore, is not likely to be related to coral bleaching. However, it is still possible that at high temperatures, DNA repair may be inhibited (as in certain terrestrial plants) and/or protective pigment synthesis impaired and then net dimer formation may increase. If this scenario is correct, then dimer formation with concurrent impairment of zooxanthellae function could help explain observed relationships between bleaching and higher temperature and light intensity. Therefore, because we have shown that UVR plays a role in coral bleaching, the effect may be on another target molecule. Laboratory studies also revealed that indicators of coral bleaching, such as zooxanthellae counts and pigment concentrations, require further critical analysis.

Factors that affect UV penetration over reefs were a major aspect of this study, and findings obtained support the hypotheses presented in the conceptual model. Large increases in UVR penetration over reefs can be attributed to stratification of ocean waters in warmer periods and photobleaching of the CDOM.


We have advanced the science of corals as it relates to UV effects on coral reefs in the following ways:

· We demonstrated that the UV exposure of coral reefs in the Florida Keys is highly variable and that this variability is linked to climate changes that are occurring over the region. The linkage stems from concurrent changes in physicochemical properties of the waters, such as warmer temperatures and increased water clarity.

· We showed that the colored CDOM in the water over the reefs plays a key role in controlling light exposure. Thus, changes in CDOM concentrations caused by climate change and/or land-based human activities can translate into significantly altered UV exposure of coral reefs.

· We identified what may be a major pathway for the large-scale impact of El Nino events on mass bleaching of corals. Our results suggest that stratification caused by the prolonged periods of low winds and warm temperatures that accompany El Nino events can result in significant increases in damaging UV radiation over the reefs. We hypothesize that this increased exposure to UV, in concert with warmer waters, places intense stress on the corals that contributes to extensive bleaching.

· We elucidated possible biological sources of CDOM in waters close to coral reefs. Changes in these biological sources, such as seagrasses and mangroves, caused by climate change and human activities can have long-term detrimental effects on corals by perturbing UV protective substances in the ocean water.

· Algorithms that relate remotely observed optical properties to UV penetration over coral reefs were developed as part of this project. These algorithms can be used in the future to provide regional scale assessments of corals UV exposure.

· We developed a sensitive and rapid technique to assess thymine dimers in tissues of aquatic organisms. This technique is more sensitive and convenient than other techniques available. This technique now can be utilized for assessing UV effects in other organisms and will be particularly useful to investigators with small samples that contain limited amounts of DNA or very low levels of DNA damage.

· We developed new techniques for immunolocalization of thymine dimers in coral, overcoming numerous technical problems not solved by previous investigators, and determined that, for P. porites, thymine dimers are localized in the zooxanthellae and not the coral tissue. We envision that the immunolocalization techniques will be of interest to numerous coral biologists. The finding that thymine dimers are localized in the zooxanthellae settles controversy as to where DNA damage would occur.

· We measured thymine dimers in coral for the first time and showed that levels of DNA damage in the field and the laboratory were very low compared to other aquatic organisms. This indicates that corals have excellent repair systems and/or pigment protection. Although it was anticipated that corals were well protected from UV damage under normal solar irradiance, it was unknown what responses might be observed with increasing irradiance. Because DNA is an important target of UV-B damage, our findings would support the possibility that other portions of the light spectrum, or other molecular targets, are responsible for observations from numerous parts of the world that solar irradiance is an added stressor in coral bleaching. Future research can focus on this topic and provide important data guiding our understanding of multiple stressors on failing coral ecosystems.

· We confirmed under controlled laboratory conditions that UV radiation can induce coral bleaching and that induction of bleaching and pigmentation is remarkably variable, even within coral colonies.

· We determined that the response of pigmentation, DNA dimer formation, zooxanthellae densities, and protein content do not correspond in a linear manner to a visible bleaching response.

· This project provides a description of an artificial UV dosing system that can induce coral bleaching and will aid in bleaching research efforts by eliminating sole reliance on field collections and inherent variability in UV dose, within colony response variability, and variability in field bleaching events.

· Our integrated approach, evaluating UV exposure and effects in a coordinated manner, has permitted a more quantitative understanding of potential levels of UV-induced damage. We found that UV irradiance over reefs may increase in warm stratified conditions but that coral are well protected against dimer formation during UV exposure. They do, however, exhibit bleaching in response to increasing UV exposure, and the mechanism for this is uncertain.

We have contributed to the management of coral reefs in the following ways:

· By identifying methods for remote sensing of corals UV exposure to coral reefs, we have laid the groundwork for a remote sensing based "UV/hotspots" system that potentially can be used to alert managers of corals that conditions are favorable for extensive corals bleaching. High water temperatures do not always presage major coral bleaching events, but high water temperatures coupled with high UV exposure almost always lead to extensive bleaching. The prediction of major bleaching events by the hotspot network is enhanced by the inclusion of a time factor in its warning procedure. That is, the hotspot must prevail for a period of weeks before a warning of bleaching is issued. A prolonged period of hotspot development also is a good indicator of strong stratification of the ocean at that location. Stratification promotes increased UV exposure over a period of time. However, the exact relationship between length of hotspot development and increased UV exposure is poorly understood. Thus, the addition of a remote sensing capability for UV exposure, coupled with the current hotspot method, likely would enhance the ability to forecast bleaching events.

· Our findings that CDOM plays a key role in controlling harmful UV exposure should help managers plan strategies to optimize coral health, e.g., by protecting and enhancing the health of seagrasses and mangroves that produce UV-protective substances.

· Our integrated effort has contributed to debate regarding multiple stressors on coral systems within the EPA and the research community. For example, although consensus is emerging that increasing sea surface temperatures are a dominant cause of bleaching, there has been little assessment of the magnitude of added stressors. Dr. Zepp's finding that thermal stratification can result in increased UV exposure will garner significant attention. Other stressors on coral systems are under investigation in the EPA and elsewhere. These include coral disease, ocean pollution, and sedimentation. Currently, global change experts are using Global Circulation Models to predict the future extent and magnitude of thermal bleaching and remote sensing of thermal anomalies for nearer term predictions. As we begin to understand the potential magnitude of added stressors, then we will know whether current predicitions are accurate. It is possible that information of this nature will eventually contribute to international debate on global carbon dioxide limitations, as well as local decisions regarding how to best increase reef ecosystem resilience through management of secondary stressors.

· Findings from this study (and the subsequent studies) are being incorporated into the Tropical Marine chapter of Climate Change Adaptation Methods and Strategies: A User's Manual. WWF (Hansen LJ, in press, 2003). This is a biome-based assessment of options to increase resilience in protected and managed ecosystems. The three dominant factors for all biomes relate to preserving resources (creation of MPAs and networks of MPAs), reduction of all non-climate stresses (coastal degradation of mangroves and seagrasses, pollution, destructive fishing), and identification of population variability for tolerant populations (MAAs). Additional related activities regarding multiple stressors and coastal habitat condition (mangroves, seagrass, terrestrial runoff) are underway in Dr. Hansen's ongoing research program. These include extensive interactions with local government in American Samoa.

Outreach and Teaching. Drs. Anderson, Zepp, and Hansen have been active in numerous outreach and teaching activities. Our full final report to the EPA details 10 invited outreach seminars, as well as numerous intern training activities. In addition, interactions with graduate students are discussed.

Awards. The ORD UV-Corals Interaction Team (R. Zepp, W. Fisher, D. Santavy, L. Hansen, L. Oliver, and B. Levinson) won an EPA Bronze Medal for "creative research on UV exposure in coral assemblages and its relevance to global climate change and coral bleaching." The award was presented in August 2002, at the EPA in Cincinnati, OH. In addition, Dr. Lara Hansen was honored with two EPA Superior Accomplishment Awards in 2000, as well as an EPA "On the Spot" Award in 1999.

UV Exposure Database. We are providing data sets obtained with the Satlantic profilers, UV irradiance measurements at a moored location at Sombrero Tower close to the reef tract in the Lower Keys, and UV-B and UV-A irradiance measurements made continuously at the Mote Marine Laboratory, Summerland Key, FL. The data are provided as Excel and ASCII files on a CD. We plan to put the data on the ERD-Athens Web Site at Exit , and they also will be made available at the BML Web Site.

Journal Articles on this Report : 6 Displayed | Download in RIS Format

Other project views: All 25 publications 6 publications in selected types All 6 journal articles
Type Citation Project Document Sources
Journal Article Anderson S, Zepp R, Machula J, Santavy D, Hansen L, Mueller E. Indicators of UV exposure in corals and their relevance to global climate change and coral bleaching. Human and Ecological Risk Assessment 2001;7(5):1271-1282. R826939 (2000)
R826939 (2001)
R826939 (Final)
not available
Journal Article Anderson S, Jackson S, Machula J, Hansen L, Oliver L, Zepp R, Cherr G, Brown H. Diurnal variation in thymine dimers in the coral Porites porites using a sensitive immunoassay. Marine Biology. R826939 (Final)
not available
Journal Article Rogers JE, Oliver LM, Hansen LJ. Symbiodinium spp. isolates from stony coral: isolation, growth characteristics and effects of UV irradiation. Journal of Phycology 2001;37(S3):43-43. R826939 (Final)
not available
Journal Article Stabenau E, Zepp RG, Bartels E, Zika RG. Role of seagrass (Thalassia Testudinum) as a source of chromophoric dissolved organic matter in Coastal South Florida. Marine Ecology-Progress Series 2004;282:59-72. R826939 (1999)
R826939 (2000)
R826939 (2001)
R826939 (Final)
not available
Journal Article Zepp RG, Callaghan TV, Erickson DJ. Interactive effects of ozone depletion and climate change on biogeochemical cycles. Photochemical & Photobiological Science 2003;2(1):51-61. R826939 (Final)
not available
Journal Article Zepp RG, Anderson SL, Stabenau E, Patterson KW, White E, Hansen L, Bartels E. Factors influencing geographic and seasonal variations in light exposure of coral assemblages in the Florida Keys. Limnology and Oceanography. R826939 (Final)
not available

Supplemental Keywords:

global climate, stratospheric ozone, marine, ecological effects, ecosystem indicators, aquatic, ecology, zoology, genetics, remote sensing, Atlantic Coast, Gulf Coast, southeast., RFA, Scientific Discipline, Air, Geographic Area, Ecosystem Protection/Environmental Exposure & Risk, Ecology, Ecosystem/Assessment/Indicators, Ecosystem Protection, climate change, State, Chemistry, Ecological Effects - Environmental Exposure & Risk, Monitoring/Modeling, Ecological Risk Assessment, Ecological Indicators, anthropogenic stresses, ecological effects, environmental monitoring, anthropogenic stress, ecological exposure, biomarkers, stressors, UV effects, thermal stratification, coral bleaching, Florida Keys, remote sensing data, coastal zone, climate, natural stressors, multiple stressors, CISNet Program, biomonitoring, ecosystem indicators, water quality, chromophoric dissolved organic matter, Florida, photobleaching, thymine dimers, Global Climate Change

Relevant Websites: Exit

Progress and Final Reports:

Original Abstract
  • 1999 Progress Report
  • 2000 Progress Report