79
Physis (Fall 2013) 14:79-88
Mackenzie Mason • Oregon State University • masonmac@onid.oregonstate.edu
Anthropogenic influence on sedimentation and hydrocarbon concentration
80 more detrimental with increasing organic content of the sediment and decreasing grain size (Fabricius 2005). Exposure to sedimentation, especially when associated with a high concentration of fine sediment, can cause long-term effects in populations of corals by removing cohorts of young corals, which are highly susceptible to damage by sedimentation (Fabricius 2005). Few examples of recovery after sediment stress have been observed, most likely because sediment stress is usually accompanied by other stresses such as sewage (Rogers 1990). Many coastal communities near coral reefs do not have adequate methods for containing sewage on land, so the sewage is released into the ocean waters. Drainage of sewage into ocean waters provides an opening through which any terrestrial run-off produced by coastal communities can flow into the ocean.
Although the input of nutrients from terrestrial run-off has been the most commonly studied type of chemical pollution to coral reefs, hydrocarbons and heavy metals are also important (Fabricius 2005). The effects of hydrocarbon leaking into the ocean may not be immediate, as they embed themselves in marine tissues (Whitehead 2013). The toxicity of some polycyclic aromatic hydrocarbons (PAHs) may be enhanced by UV radiation. These PAHs can absorb UV light and become photo activated, then transfer their energy to molecular oxygen, forming reactive superoxide anions capable of oxidative damage (Bellas et al. 2008). Bellas et al.
(2008) performed an experiment using a battery of marine larvae in a variety of concentrations of intermediate weight PAHs and found that in their presence, UV light inhibited larval development. A study testing a variety of PAHs showed that most intermediate PAHs at natural levels do not cause oxidative damage under UV light, but at levels that could exist in areas polluted with oils cause oxidative damage to marine tissues (Fathallah et al. 2012).
Terrestrial run-off is known to cause harm to aquatic ecosystems, but the impact of contamination depends on chemical composition and the species composition of the affected area (Van Eal et al. 2012).
The distribution of waste products from the point of contamination is related to the velocity of the sea currents (Tomassetti and Porrello 2005). In aquatic ecosystems, organic pollutants are mostly concentrated in the sediments, which act like a sink for organic compounds and are an important exposure pathway to aquatic species (Van Eal et al. 2012). Small invertebrates living within contaminated sediment are excellent indicators of environmental stress because they are directly exposed to high concentrations of pollutants and are sessile or have limited range due to their small size (Tomassetti and Porrello 2005).
Benthic assemblages cannot easily avoid exposure to stressful conditions, therefore they can respond in a variety of ways including a quantitative composition change in the community (Tomassetti and Porrello 2005).
The island of Bonaire, Dutch Caribbean is known for its coral reefs and marine conservation efforts, but does not have a system for regulating terrestrial run-off. Following an observed disturbance near a specific drain in Bonaire, this study analyzed the contents and effects of the drainage. Corals, algae, and herbivorous fishes inhabit the area are affected by this drain, but it is also an area that is commonly used by people. The drain is located in the middle of an area of Bonaire popular to tourists next to swimming lanes and across the road from a restaurant. Because the drain of interest lies in a highly trafficked area, it is of great importance to the island of Bonaire due to its high reliance on tourism as income. The waters flowing through this drain have a visible effect on the ocean waters close to the drain, but the flow rate of the drain is currently unregulated. One week prior to the start of this study, construction work was conducted near the drain, and it was
81 expected that chemicals and organic material disturbed by the construction entered the water through the drain. This disturbance could have potentially caused the release of contaminants and fine sediment from the drain into the ocean; an effect that could be intensified by frequent rains in the upcoming rainy season.
The purpose of this study was to test for fine sediment and UV reactive hydrocarbons content near the unregulated drain and any effect that the run-off may have on the fauna living nearby. Analysis of the contents and effects of this particular drain in Bonaire answers the following questions: 1) How does the sediment composition differ in areas closer to the drain from areas far from the drain?
2) Does the sediment that runs from the drain contain UV reactive hydrocarbon compounds? 3) Does the benthic fauna composition differ in areas closer to the drain from areas far from the drain? This study analyses the following hypotheses:
H1: Sediment collected from areas near the drain will contain a higher percentage of fine sediment than sediment far from the drain
H2: UV reactive hydrocarbon compounds will be present in water samples close to the drain
H3: There will be a difference in benthic community structure between areas close to the drain and areas far from the drain
If the sediment collected near the drain has a higher concentration of fine sediment, it will be more harmful to coral health than sediment far from the drain.
chemicals into the ocean environment and aims to call attention to this source of contamination in Bonaire. This study builds on current knowledge of anthropogenic release of known harmful
Materials and methods Study site
This study focuses on one particular drain in Bonaire, Dutch Caribbean. The drain is located next to the dive site Kas di Arte (12.155472 N, 68.27914 W) in the downtown area of Kralendijk, the most densely populated city on Bonaire. The drain is adjacent to swimming lanes, across the road from a few homes and a restaurant (La Barca) and down the street from a recreational park (Fig. 1). The drain runs under a main road, and the water runs through a cement-bottomed trough that collects run-off from the surrounding houses and restaurants (Fig. 1). During heavy rains, there is an increase in effluence through the drain and run-off into the ocean. The area under investigation begins 10 m from the drain and extends 30 m from the drain towards the reef. The substrate near the drain is rocky and has sparse patches of sand, becomes sandy 20 m from the drain, and the reef drop off starts 80 m from the drain. Corals are patchily distributed in the area near the drain, but are not clustered together and are mostly overgrown with algae. A rebar was hammered into the sediment 10 m, 20 m, and 30 m from the center of the drain and visited once a week for 5 weeks to record observations and take sediment and water samples for analysis in the laboratory.
Field observations
Effluence through the drain is highly dependent on rain and disturbances, therefore the number of days since the last rain or observed disturbance was recorded each week that samples were collected.
The waters surrounding the drain are naturally clear but there is a visible plume of dark deposition when high amounts of water flow through the drain. This study began one week after construction work, which caused a visible black ring around
82 the drain. Observations on the clarity of water surrounding the drain, as well as recent rains and time elapsed since the construction work were recorded.
Observations on water visibility, current, and waves were also recorded each week.
Granulometry
To analyze sediment composition and proportion of fine sediment at varying distances from the drain, sediment samples were taken each week within a 1-m radius of each rebar, starting from the 10-m marker and moving away from the drain.
In a simular study study performed by Duplisea and Drgas (1999), 8-cm cores were taken within 1m of a marker to a depth of greater than 12 cm. This study used a core sampler with a 15-cm diameter to take samples to a depth greater than 12 cm. Because the 10-m marker is in a rocky area, rocks often prevented sampling of sediment to the top of the core sampler
(~40 cm). In the case that sediment could not be sampled deeper than 12 cm, the samples at the 20-m and 30-m markers were taken to the same depth as the first marker. The samples were taken to the laboratory in plastic bags and placed in aluminum tins for several hours outdoors to remove excess water and subsequently placed in a dry oven set to 65.56°C for 48 h. When the samples were completely dry, they were weighed and placed in a 7-chamber sieve to be separated based on grain size. Each size of the sieve corresponded to different classifications of sediment. The chambers and their classifications are as follows: 2000 μm (rocks and shells), 1000 μm (coarse sand), 500 μm (medium sand), 250 μm (fine sand), 125 μm (very fine sand), 63 μm (silt and clay), <63 μm (fine clay) (Bian and Zhu 2009). The weight of sediment in each chamber was then recorded for data analysis.
Fig. 1 Map of Bonaire with the location of Kas di Arte in relation to the island and a close-up of the study site.
Dots indicate sampling sites
83 Bioassay
To test for the presence of UV reactive hydrocarbons, water samples were taken from within the benthos at the 10-m marker each week of sampling. Water was collected using a 60-ml syringe wrapped in 250-µm mesh to prevent sand clogging the syringe. The water was then placed in plastic air-filled bottles that were previously cleaned using 10% HCl and Milli-Q water. The water samples were taken back to the lab and diluted with filtered seawater to create a 1:3 dilution, a 1:1 dilution, and a 1:0 pure sample. The different concentrations of sampled water/filtered seawater along with a control of pure filtered seawater were used to perform a bioassay on Artemia sp. This shrimp is routinely used as a test organism for screening in ecotoxicological studies (Lu et al. 2012). The Artemia were hatched each week three days prior to the day of the bioassay. To perform the bioassay, 50 Artemia were placed into a 15-mL petri dish filled with 10 mL of water dilution. The Artemia were left for four hours and then exposed to a standing UV light. Eight petri dishes with Artemia were created in total, four of these received UV treatment, one dish for each solution (0:1,1:3,1:1,1:0), and four of these were placed under glass to shield them from UV light. The no-UV treatment was the negative control of this experiment and represents the percentage of individuals that died of natural or unrelated causes.
Four petri dishes with 50 Artemia and dilutions (0:1, 1:3, 1:1, 1:0) of filtered seawater with four drops of motor oil served as a positive control and were placed in the UV treatment. One petri dish with 100% motor oil solution was placed under glass. This positive control experiment was replicated once. After exposing the Artemia sp. to UV light for 14 minutes and 59 seconds, they were left in a dark room overnight. The following day, the numbers of dead Artemia sp. in each petri dish were counted and recorded
for data analysis. Toxicity results were expressed as percentage mortality adjusted relative to the negative control, following Abbot’s formula:
%Toxicity= (It-I0)/(1-I0) x100 where It denotes the observed mortality of the UV treatment and I0 represented the natural mortality of negative control (Lu et. al 2012).
Endobenthos
To test the effect of the pollution from the drain on some of the organisms living within the surrounding ecosystem, the number of fauna living within the benthos was determined using endobenthos technique. A core sample was taken each week near each of the markers with the 15-cm core sampler to a depth of >12 15-cm. The samples were brought to the surface in plastic bags and transferred to a plastic container if the bag was leaking. The benthos samples were treated with a relaxing solution (MgCl2) then brought back to the laboratory. The samples were stained using a Rose Bengal/ethanol(10%) solution then placed in the refrigerator for 1-2 days before analysis. The samples were run through a 500-μm sift and placed in a plastic container with fresh water.
Fauna within the samples were extracted using forceps and placed in 70% Ethanol solution for counting and analysis under a microscope. One 100-mL dish of sand from each sample was analyzed under a microscope to determine if fauna remained in the sand after the initial extraction.
Data analysis Granulometry
The grain size proportions of each sample were organized into a bar graph for visual comparison of the three sampling sites.
The percentage of silt and clay were each organized into one bar graph to analyze
84 the differences between the three groups and compare the change over time to the rain and disturbance observations.
Bioassay
Scatter plots were made comparing the percent of dead individuals versus the concentration of sampled water for the UV treatment and the negative control for each week’s bioassay experiments and analyzed with linear regression analysis. One scatter plot comparing the percent toxicity of each week’s experiments versus the percent sampled water was made and each week was analyzed with linear regression. A scatter plot of the percent dead individuals versus percent motor oil solution was made for the positive control and analyzed with linear regression. A bar graph of the percent toxicity in pure sampled wáter each week was made to analyze the difference in the toxicity over time and compare to rain and other disturbances.
Endobenthos
A table of the number of organisms counted in the different samples (10-m, 20-m, 30-m) was created to compare the sites over the four weeks against the rain and disturbance observations.
Results
Due to the open and unregulated nature of the drain and the shape of the cement trough leading into the drain, rain and disturbances have a large impact on the
run-off coming from the drain.
Observations on the appearance of the water around the drain, recent disturbances, and the strength of waves and currents were recorded (Table 1). No samples were taken into the lab during the first week of observation, so this week is labeled as week 0. Weeks 1-4 correspond to the weeks during which samples were brought into the lab for analysis. A black plume around the drain was observed until week 3 after the construction work, and observed again during week 4 of sampling after a heavy rain event (Table 1).
Granulometry
The percent of silt in the sediment samples was highest in the sediment 10 m from the drain each week except for week 2 when the 30 m site contained the highest percent of silt (Fig. 2). The sediment 20 m from the drain contained nearly the same percent of silt each week (Fig. 2). The sediment 30 m from the drain contained very little silt during week 1 and no silt during week 3, but more silt during weeks 2 and 4 (Fig. 2). The site which contained the highest percent of clay every week was the 10-m sampling site (Fig. 3). The 30-m sampling site contained no clay during week 3 and very little clay during week 4 (Fig. 3). The percent of clay present in the samples from the 20-m site varied little each week except week 3, when there was no clay present (Fig. 3).
Bioassay
The numbers of dead individuals in the
Table 1 Observations on recent rain/disturbance each week and surface observations of the water around the drain and strength of waves/current
Week Days since last
disturbance Type of disturbance Surface Observations Waves/current
0 10 Construction Thick black plume Mild
1 17 Construction Light black plume Mild
2 24 Construction No plume Strong
3 5 Light rain No plume Strong
4 0 Heavy rain Thick Black plume Mild
85
Fig. 2 Percent of the sediment at each of the sampling sites that consisted of silt (63-125 μm) each week
Fig. 3 Percent of the sediment at each of the sampling sites that consisted of clay (<63 μm) each week
UV treatment and the no-UV treatment counted in each concentration of contaminated water were analyzed for the first 3 weeks. The data for the bioassay during week 4 and the second test in motor oil solution were not included due to contamination in the Artemia hatching tank. The positive control in motor oil solution showed a positive correlation (R2=0.93903) for the UV treatment and 10% of the individuals died in the 100%
no-UV treatment (Fig. 4). A scatter plot was constructed comparing the percent toxicity vs the percent sampled water for all three weeks (Fig. 5). Week 1’s data showed a positive correlation (R2=0.91859), week 2’s data showed positive correlation (R2=0.90494), and
week 3’s data showed a positive correlation (R2=0.98387) (Fig. 5). The percent toxicity in the 0% sampled water solution had a value of -21% during week 1 and -11.1% during week 2. A bar graph was constructed to show the percent
0 0.5 1 1.5 2 2.5 3 3.5
week 1 week 2 week 3 week 4
Percent sediment 63-125 μm
10m from drain 20m from drain 30m from drain
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
week 1 week 2 week 3 week 4
Percent sediment <63 μm
10m from drain 20m from drain
30m from drain Fig. 4 Percent of dead individuals with respect to the percent of motor oil solution for the UV treatment and the no-UV treatment with linear regression (R2=0.939)
Fig. 5 Percent sample water vs percent toxicity with linear regressions for week 1 (R2=0.91859), week 2 (R2=0.98387), and week 3 (R2=0.90494)
Fig. 6 Percent toxicity of pure sampled water each week
0 10 20 30 40 50
0 25 50 75 100
% Dead individuals
% Motor oil solution UV Treatment
No UV Treatment
-30 -20 -10 0 10 20 30
0 20 40 60 80 100
% Toxicity
% Sampled water Week 1 Week 2 Week 3
0 5 10 15 20 25 30
week 1 week 2 week 3
% Toxicity
86 toxicity in the 100% contaminated solutions each week (Fig. 6). The percent toxicity was 26.1% during week 1, 21.7%
during week 2, and 19.6% during week 3 (Fig. 6).
Endobenthos
The number of organisms counted from the sites were not consistently different from each other throughout the sampling period. The sample from the site 10 m from the drain contained 10 organisms during week 1, 39 during week 2, 20 during week 3, and 12 organisms during week 4 (Table 2). The highest number of organisms counted was 39 from the 10-m sampling site during week 2, and the lowest number of organisms counted was 7 from the 20-m sampling site the same week (Table 2). No organisms greater than 500 μm were observed in any of the petri dishes of sand taken from each sample.
Discussion
The sediment sampled near the 10 m marker contained an observably higher proportion of silt and clay each week than the sediment sampled near the 20-m and 30-m markers (Fig. 2, Fig. 3). The percentage of clay in the sediment 10 m from the drain was highest during weeks 1 and 4 of the study, which were the weeks when disturbances had the greatest affect on the drain (Fig. 3). The sediment contained 1.14% clay in the first week of sampling, 2 weeks after construction work, and decreased to a value near 0.45% until
week 4 when, due to heavy rain, a value of 1.34% was measured (Fig. 3). These observations suggest that the abundance of clay in the sediment near the drain is highest when a recent disturbance has occurred. The percentages of silt showed a more uneven distribution between the three sampling sites than the percentages of clay, but the 10-m sampling site also showed a higher percentage of silt during weeks that were affected by disturbances (Fig. 3). The differences in percentages of silt and clay during weeks without disturbance and absence of fine sediment in the 20-m and 30-m sites during week 3 could be due to the strong currents and waves observed during these weeks. Fine sediment is more likely to become suspended in fast moving fluid than coarser, heavier sediment and is more likely to disperse over long distances (Bian and Zhu 2009). The presence of fine sediment in the 20-m and 30-m sampling sites was observed again after heavy rain.
This data suggests that fine sediment flows through the drain after a disturbance and that this fine sediment is suspended into the water column and swept away with currents. Although a difference in sediment composition between the 3 sites is observed, the hypothesis regarding fine sediment proportions cannot be supported using statistical analyses due to the time-dependent nature of this study.
The results of the bioassay showed a positive relationship between percent sampled water and percent toxicity with above 90% goodness-of-fit each week.
Increasing mortality with decreasing dilution of the sample water suggests that UV reactive hydrocarbons present in the sample water caused deaths in the Artemia sp. The elevated mortality in the negative control with 0% sample water compared to the UV treatment in 0% sample water during the first two weeks suggests that unrelated factors caused death in the Artemia sp. The results of the positive control confirm that the percent of dead individuals increases with increasing
Table 2 Numbers of organisms counted in the benthos samples collected at each of the sites every week
Number of organisms counted Week 10 m from
drain
20 m from drain
30 m from drain
1 10 10 13
2 39 7 24
3 20 18 13
4 12 26 10
87 concentration of hydrocarbons. Analysis of the percent toxicity in the 100% sample water dishes over time showed that toxicity was highest during the first week of sampling and decreased during weeks 2 and 3 (Fig. 6). This observation follows a similar trend to the percentages of clay in the sediment samples from the same site over time. The water samples contained some of the small particles that were absorbed from the benthos layer and a higher toxicity was observed when more fine sediment (<63 μm) was present. UV reactive hydrocarbon adsorption into marine sediment varies with sediment size, and small particles (<50 μm) have the highest capacity to carry hydrocarbons (Hiraizumi et. al 1979). The water collected from near the drain that resulted in phototoxicity in the Artemia sp. is likely to contain UV reactive hydrocarbons that flowed from the drain along with fine sediment.
The results of the endobenthos organism count did not show any trend that suggests that the pollution from the drain is correlated to a change in the abundance of benthic fauna. However, the results did show that the numbers of organisms found in the samples taken from the 10-m sampling site were lowest during weeks of high disturbance and high fine sediment percentage. The lower numbers observed during these weeks could be because increased sediment load containing UV reactive hydrocarbons from the drain caused the death of the less tolerant organisms living within the benthos. Given a longer period of study time, identification and analysis of the species diversity of the fauna may provide stronger evidence to analyze the impact of the run-off on the benthic organisms.
Future studies analyzing the relationship between benthic communities and the effluence from this drain over time could determine what effect this drain has on benthic organisms.
The data collected during this study suggests that the run-off from the drain
contains fine sediment and UV reactive hydrocarbons and is sensitive to rain and other disturbances. The fine sediment released from the drain can have negative impacts on the surrounding ecosystem because of its impact on turbidity and its ability to adsorb harmful chemicals. The hydrophobic nature of UV reactive hydrocarbons causes them to have a higher affinity for molecules that are less polar than water, and hydrocarbons will settle into other materials when suspended in water. Hiraizumi et al. (1979) showed that hydrocarbons suspended in water had a higher affinity for fine sediment than other sediment sizes, but had the highest affinity for the membranes of zooplankton. The chemical properties of hydrocarbons, such as their chemical stability and hydrophobicity, make them persistent in the environment and give them the ability to accumulate easily into the tissues of biota, to enrich throughout food chains, and to eventually cause toxicological effects (Van Ael et. al 2012). Fine sediment released from the drain is likely to contain hydrocarbons that will transfer to organisms living within the benthos and lead to bioaccumulation throughout trophic levels.
The area surrounding the drain up to a distance of 20 m contains corals and algae that are growing on the rocky substrate.
Fine sediment released from the drain can increase turbidity and settle on the membranes of corals, decreasing light available to the corals. Sediment containing hydrocarbons that settles on corals is likely to transfer the chemicals into their membranes, making them vulnerable to phototoxicity. Damage to the corals causes the algae to outcompete the corals for space, and reduce coral cover in the area.
Pollutants released from the drain can cause damage to the marine ecosystems in many ways, and this damage can also have detrimental impacts on humans. Decreased coral cover in an area popular to tourists is important to prevent in Bonaire because of
88 its high dependence on tourism as a source of income. Bioaccumulation of toxic chemicals through trophic levels of the marine ecosystem can negatively impact humans who consume fish that have accumulated harmful chemicals.
Regulating the source of effluence can lower the amount of pollution that has an effect on marine life and human populations. This study and others like it aim to call attention to the negative impact of pollution on marine ecosystems because the only solution to anthropogenic impacts on ecosystems is awareness and action by the citizens inhabiting offshore areas.
Acknowledgements First and foremost, I would like to thank Dr. Enrique Arboleda for being a wise and patient advisor and giving me guidance as I formulated a plan for my project. I would like to thank Yannick Mulders for his insight into my project and the constructive comments he made while editing my paper. I would also like to thank my research partner, Kyra Creger, for helping me collect samples every week and helping me analyze those samples in the lab. I would also like to thank Megan Beazley and Kevin McFadden for helping me with lab work and Dr. Rita Peachey for giving me my project idea. I give special thanks to CIEE and Oregon State University for making it posible to do this research.
References
Bellas J, Saco-Álvarez L, Nieto O, Bieras R (2008) Ecotoxicological evaluation of polycyclic aromatic hydrocarbons using marine invertebrate embryo-larval bioassays. Mar Poll Bull 53:493-502
Bian B, Zhu W (2009) Particle size distribution and pollutants in road-deposited sediments in different areas of Zhenjiang, China. Environ Geochem Health 31:511-520
Dubinsky Z, Stambler N (1996) Marine pollution and coral reefs. Glob Chan Bio 2:511-526 Duplisea DE, Drgas A (1999) Sensitivity of a
benthic, metazoan, biomass size spectrum to differences in sediment granulometry. Mar Ecol Prog Ser 177:73-81
Fathallah S, Medhioub MN, Kraiem MM (2012) Photo-induced toxicity of four polycyclic aromatic hydrocarbons (PAHs) to embryos and larvae of the carpet shell clam Ruditapes decussatus. Bull Env Cont Tox 88:1001–1008
Fabricius KE (2005) Effects of terrestrial runoff on the ecology of corals and coral reefs: review and synthesis. Mar Poll Bull 50:125-146 Hiraizumi Y, Takahashi M, Nishimura H (1979)
Adsorption of polychlorinated biphenyl onto sea bed sediment, marine plankton, and other adsorbing agents. Environ Sci Technol 13:580-584
Kline DI, Kuntz NM, Breitbart M, Knowlton N, Rohwer F (2006) Role of elevated organic carbon levels and microbial activity in coral mortality. Mar Ecol Prog Ser 314:119-125 Lu Y, Xu X, Li T, Xu Y, Wu X (2012) The use of a
brine shrimp (Artemia salina) to assess the water quality in Hangzhou section of Beijing-hangzhou grand canal. Bull Environ Contam Toxicol 88:472-476
Rogers CS (1990) Responses of coral reefs and reef organisms to sedimentation. Mar Ecol Prog Ser 62:185-202
Tomassetti P, Porrello S (2005) Polychaetes as indicators of marine fish farm organic enrichment. Aqua Int 13:109-128
Van Ael E, Covaci A, Blust R, Bervoets L (2012) Persistent organic pollutants in the Scheldt estuary: Environmental distribution and bioaccumulation. Environ Int 48:17-21
van Dam JW, Negri AP, Uthicke S, Meuller JF (2011) Chemical pollution on coral reefs:
exposure and ecological effects. In: Sánchez-Bayo F, van den Brink PJ, Mann RM (eds) Ecological impacts of toxic chemicals.
Bentham eBooks, pp 187-211
Whitehead A (2013) Interactions between oil-spill pollutants and natural stressors can compound ecotoxicological effects. Int and Comp Bio 53:635-647