Mangrove filtration of anthropogenic nutrients in the Rio Coco Solo, Panama

Mangrove filtration ofanthropogenic nutrients in the

Rio Coco Solo, PanamaBrenda B. Lin

School of Natural Resources and the Environment,University of Michigan, Ann Arbor, Michigan, USA, and

Jonathan DushoffDepartment of Ecology and Evolutionary Biology, Princeton University,

Princeton, New Jersey, USA

Keywords Ecology, Filtration, Panama

Abstract Measurements of the distribution pattern of several nutrients (ammonia, nitrite,nitrate, and phosphate) and indicators (dissolved oxygen and conductivity) were conducted alongthe river Coco Solo on the Caribbean coast of Panama. The project investigated the extent to whichmangrove forests could act as a vegetative buffer zone between disturbed freshwater sources andcoastal water. Upriver freshwater samples were collected in known areas of human disturbance toassess levels of the nutrients near points of origin and exhibited elevated concentrations ofnutrients. Samples were taken along the mangrove estuary to study the concentrations of nutrientsas they moved through the estuary into the ocean. Expected and observed values were compared tosee whether concentration levels of the chemicals exhibited normal dilution patterns. Graphs showthat the nutrient levels at the estuary points are significantly lower than expected through normaldilution, indicating the removal of nutrients through another mangrove mediated method. In thisway, mangrove forests can act as effective coastal buffers of anthropogenic effects to the oceanenvironment. Further studies must be done to determine the actual removal mechanisms ofnutrients in the mangrove estuarine system.

IntroductionThe mass movement of people to coastal urban areas has been a dominanttrend in the late twentieth century in both developed and developing nations(Heinrichsen, 1998). This demographic trend has resulted in severely pollutedterrestrial and aquatic ecosystems, in lower watershed, and in coastal areas(Bear and Pringle, 2000). In tropical riverine and coastal ecosystems,mangroves have been found to combat certain types of human disturbanceby cleansing waters of excess nutrients (Zuberer and Silver, 1979; Shermanet al., 1998; Corredor and Morell, 1994).

This study investigates the importance of mangroves in maintaining waterquality in the Bahıa Las Minas area in the Colon region of Panama. Thickmangrove forests border the coast, while most of the inland forest has beenremoved to make room for development. The rapid development and intense

The Emerald Research Register for this journal is available at The current issue and full text archive of this journal is available at

www.em eraldinsight.com/res earchregister www.em eraldinsight .com/1477-7835. htm

The authors would like to thank Princeton Environmental Institute and a Mellon Grant to SimonA. Levin for their support in this project.

Rio Coco Solo,Panama

131

Management of EnvironmentalQuality: An International Journal

Vol. 15 No. 2, 2004pp. 131-142

q Emerald Group Publishing Limited1477-7835

DOI 10.1108/14777830410523071

human disturbance translate to a considerable amount of pollution, whichthreatens to spread into the ocean. The poor water quality seen in inland waterscontrasts with the health of the coastal waters beyond the mangrove forests.This discrepancy may be due to cleansing by the mangrove ecosystem.

The danger of allowing a high nutrient level to build up in the ocean iseutrophication, specifically the production and dominance of blue green algalmats, which can lead to anoxia. Eutrophication may have far-reaching effectson the higher levels of food webs and can create irreversible damage toecosystems (Schlesinger, 1997). The importance of mangroves to water qualitywill be relevant to policy decisions about whether coastal ecosystems should bepreserved in this rapidly growing urban area.

Mangroves as vegetative buffersMangroves dominate the world’s coastlines in tropical areas. The mangroveecosystem holds and stabilizes the environment from erosion and acts as abuffer zone between land and sea. The sheltered slack-water conditions allowthe deposition of fine particles normally enriched with metals, organic matter,and minerals (Ramanathan et al., 1999). Mangrove prop roots support richecological communities (Keller and Jackson, 1993). Studies in the Colon-BahıaLas Minas region show that the prop roots of the red mangrove trees support127 species of animals, and 43 species of algae (Cubit et al., 1987).

Many previous studies have shown that vegetation facilitates biologicaldegradation of chemicals in water systems (Dillaha et al., 1989; Alberts et al.,1981; Doyle et al., 1977). Areas of vegetative buffering include wetlands, saltmarshes, and stream habitat. As water passes through these habitats,vegetative structures cause loose sediment and nutrients to be trapped in thisarea. Vegetation also creates conditions for microorganisms to flourish in theroot zone where they can actively take up nutrients. Studies that have removedvegetative buffers along waterfronts have shown a decrease in water quality(Barling and Moore, 1994).

The Bahıa Las Minas is a northwest-facing bay, approximately sixkilometers wide, situated immediately east of the Atlantic entrance to thePanama Canal (see Figure 1). The convoluted bay is fringed with densemangrove forests growing in a variety of habitats, including exposed coastalsites behind reef flats, margins of tidal channels, and river dominated estuaries(Duke, 1996). Many rivers and creeks run through this region, passing throughmultiple villages before reaching the thickly bordered mangrove coast. Thewaters directly surrounding Colon have long lost their protective mangrovelining to make room for port space. As a result, the waters directly outside thecity have undergone a large amount of eutrophication (D’Croz and Robertson,1997). The mangrove forests in the Bahıa Las Minas area have remained, forthe most part, intact. Many signs point to the mangrove forest as the main

MEQ15,2

132

source for maintaining water quality and marine health in the oceans of thecoast of Colon.

This study addresses the question of mangrove filtration by examining thechemical concentration of several pollutants. We compare our data to a simpledilution model, and find evidence that mangroves are indeed extracting somepollutants from the water system.

Materials and methodsClimatically, Panama experiences two conditions: a dry season, which occursbetween December and April, and a wet season, which occurs from May toNovember. The dry season consists of low rainfall, strong onshore northerlytrade winds, and high mean water level over reef flats. The wet season has

Figure 1.Map of the Bahıa LasMinas region outliningthe sampled Rıo CocoSolo and the sampled

river, estuary, andocean points

Rio Coco Solo,Panama

133

variable winds (onshore mainly), a lower mean water level, and 2,000-4,000mmof rain per year (Keller and Jackson, 1993). Because of this seasonal pattern,there are distinctive seasonal cross-shelf variations in the water and planktonthat affect the coral reefs. Rainfall and run-off are frequently mentioned as themain source of environmental gradients for coral reef ecosystems nearcontinental lands (Birkeland, 1987). Run-off from these freshwater sources isimportant for plankton, as it is the source of land-derived inorganic nutrients inthe coastal zone. Higher plankton production is expected to occur near shore,and to be correlated with peak rainfall and runoff (D’Croz, 1999). Therefore,higher amounts of inorganic nutrients will enter into the oceanic saltwatersystem during the rainy season. Light penetration is decreased both by theabundance of plankton during the upwelling, and by sediments released withfreshwater discharge from rivers during the wet season.

Rio Coco Solo is typical of many rivers that run through the Bahıa Las Minasregion. The river is composed of tributaries that begin in an upland protectedforest near Lake Gatun. Each tributary runs through several small villages, andeventually crosses the Transıstmica highway. Once the tributaries join into oneriver, it passes along the side of the hospital, local airport, and Department ofPublic Works. The river enters into a forested and grassland area beforeentering the thick forest of the coastal mangroves.

The mangrove forest is covered with heavy sediment. The river does nothave a channel in the mangrove forest, but rather runs through the mangroveroots and sediment. The sediment in this area is too loosely packed to supporthuman weight, so a key portion of the river could not be sampled in this study.This un-sampled distance extends for approximately 150 meters until themangrove forest becomes less dense, and a shallow river channel begins to formagain. The channel continues to become deeper as it moves toward the ocean. Atthe mouth, the river is about one meter deep, but still heavy with sedimentationand soft mud. This area is dominated by Rhizophora mangrove. Because of theshallowness of the water, we assumed the river was vertically well mixed.

Water samples were collected from the Rio Coco Solo in August 2000. Oneocean sample and 15 estuary samples were collected on August 23. Because thewater column is vertically well mixed, only one sample was taken from eachpoint along the estuary. Before each water sample was taken, the bottle waswashed out three times in the river water. The estuary was entered by boatfrom the bay. One-liter samples of water were taken at different points alongthe estuary, as well as at the mouth of the river and 30 meters beyond themouth, in order to obtain the oceanic reading. Ten freshwater samples werecollected on August 24. Tributaries of the river were sampled at multiple inlandpoints in order to assess the amount of pollution coming from the variousfreshwater points leading to the estuary. Samples were taken from variouslocales such as roadsides and villages where the river was easily accessible.Most freshwater samples were taken from areas with human inhabitants.

MEQ15,2

134

Samples were stored in plastic ice chests and kept dark in order to preventchemical changes. Bottles were taken out of the ice chest and returned to roomtemperature before testing. Chemical analysis used chemicals and testingmethods supplied by Hach Chemical Company based on the DrelSpectrophotometer 2000.

Salinity tests were done with a conductivity meter. Conductivity, dissolvedoxygen, and ammonia were tested prior to filtering. Because of the heavy layerof mud in some of the mangrove estuary samples, nitrite, nitrate, andphosphate could only be tested after the water had been filtered.

Mathematical analysisNutrient levels in the estuary are expected to be lower than in the river for tworeasons: removal by the mangrove ecosystem, as discussed above, and dilutionof river water by ocean water with lower nutrient concentrations. To test forthe effects of the mangroves, we thus compared our results with those thatwould be expected from dilution alone. Expected concentrations due to dilutionwere determined by assuming that sample salinity reflects the proportion ofsaltwater and freshwater in the sample. Thus the expected concentration in anestuary sample is given by:

Ce � Co

Cr � Co¼

Se � So

Sr � So:

where Ce is the expected concentration of nutrient in a sample, if there were noremoval by mangroves, Co is the concentration of the chemical in the ocean, Cr

is the concentration in the river, Se is the salinity of the sample, Sr is the salinityof the river and concentration at the freshwater source, and So is the salinity ofthe ocean (Boyle et al., 1974).

Since all the quantities except Ce are known, the expected level of thechemical can be easily calculated. By applying this equation to each sampledpoint along the estuary, we obtain a linear relationship between the expectednutrient concentrations to the conductivity when plotted. An observedconcentration below the line then means that less of a given chemical waspresent than would be expected from dilution alone, and thus some process wasremoving the chemical. An observed concentration above the line means thatmore chemical was found than would be expected under pure dilution.

Certain points were grouped and averaged based on their conductivities.Expected levels were then calculated from the averages. The results from thesepoints are very similar to those found in the ungrouped data. Standard errorwas calculated for each point and error bars were added to aid in showing thepronounced difference between the corresponding observed and expectedlevels at each sample point along the estuary. We added 0.005 variance as aconservative correction for measurement error.

Rio Coco Solo,Panama

135

ResultsThe data show that levels of ammonia, nitrite, nitrate, and phosphate are notconserved through the estuary system (Figures 2-5). The levels of thesechemicals are significantly lower than would be expected in a closed system,approaching detection limits in most observations. This is most apparent in theammonia and nitrite results. Nitrate had higher concentrations at some points,but still not as high as one would expect under dilution alone. Phosphate levelsshow a greater degree of random scatter, but all of the observed points, exceptfor the last, are below the expected value of a conservative system.

Measurements of oxygen were taken to ensure that all of the water sampleswere in the same redox regime. An ANOVA (analysis of variance) of dissolvedoxygen was not significantly different, indicating that the water samples werein the same redox regime. Copper and iron were also measured, and were foundnot to be significant.

DiscussionTesting of the water samples involved many chemicals, but analysisconcentrated primarily on ammonia, nitrate, nitrite, and phosphate. Thesechemicals could be traced from known pollutants such as waste contaminationor sewage run-off. Oxygen content of the samples was measured to make surethat both the freshwater and estuary samples were in the same redox regime.

Figure 2.Graph of observedammonia concentrationsand concentrationsexpected from dilutionhypothesis

MEQ15,2

136

Freshwater oxygen levels measure about 8.0 mg/L, and oxygen levels in salinewater are usually somewhat lower. Although the estuary samples had loweroxygen readings than normally expected, they remained oxic, and thereforestayed in the same redox regime as the freshwater samples. The lower dissolvedoxygen levels may be a result of heavy levels of organic matter in the mangrovesystem. The conductivities and oxygen readings show that the water is in theproper redox zone to support this assumption (Boto and Bunt, 1981).

Our data show that significantly more chemical is removed from the waterthan would be expected by normal dilution. Many other studies have been doneon the mangrove system showing the removal of chemicals through variousmethods (Zuberer and Silver, 1979; Corredor and Morell, 1994; Sherman et al.,1998). Active biological degradation due to bacteria and other pathogens canremove large amounts of nutrients, and the mangrove root system facilitatesthe growth and development of such organisms. Low concentrations ofnutrients in wetland water columns may also be due to the short-term removalof such chemicals as phosphate from the surface water by adsorption onorganic matter and bottom sediments (Nwankwor and Anyaogu, 2000). In somecases certain chemicals, such as ammonia, may also be volatilized, and lost tothe atmosphere.

Figure 3.Graph of observed nitrite

concentrations andconcentrations expectedfrom dilution hypothesis

Rio Coco Solo,Panama

137

NitrogenObserved levels of ammonia, nitrite and nitrate were all significantly lower inthe estuary than would be expected from dilution alone. This is not a surprisingresult and is presumably due to nitrogen uptake by mangroves and theirassociated bacteria (Zuberer and Silver, 1979). Several studies have shown thatmangrove sediment microbial communities are capable of taking up largeamounts of nitrate added to the water system by sewage effluent in controlledstudies (Corredor and Morell, 1994; Nedwell, 1975).

Ammonia oxidizes to nitrite and then to nitrate through biologicallymediated processes. This process, together with decomposition of organicmatter, drastically lowers the dissolved oxygen levels of the micro layer at thesediment-water interface. This process may explain the low levels of dissolvedoxygen found in the estuary (Babu et al., 2000). This chemical progression mayalso explain some of the patterns seen in the nitrogen data: nitrate is still foundin the lower estuary zone, while ammonia and nitrite are completelyundetectable by the time the river reaches the first sampled estuary point.

Rivera-Monroy and others (RiveraMonroy et al., 1995; RiveraMonroy andTwilley, 1996) argue that the uptake of nitrogen in the riverine mangrovesystem is not necessarily due to a nitrogen sink via denitrification and bacteriauptake, but may be due to retention of nitrogen in mangrove sediments.

Figure 4.Graph of observednitrate concentrationsand concentrationsexpected from dilutionhypothesis

MEQ15,2

138

PhosphatePhosphate typically exists in a single, stable oxidation state in both solids andliquids, and has no significant atmospheric fluxes. Its ability to adsorb anddesorb readily from sediment creates a buffer system in the water, which tendsto maintain near-constant concentration levels of phosphorus (Froelich, 1988).

Phosphate levels are usually higher in estuaries than in fresh water(Froelich, 1988). This is contrary to our measurements in the Coco Solo riversystem. The high levels of phosphate from the freshwater zone of the riveralmost disappeared in the estuary zone. This may be because the natural levelof phosphate in the estuary is lower than the high concentrations found infreshwater samples. Because of the ability of phosphate to adsorb and desorbfrom sediment to balance the phosphate levels in the river, it is possible thatmost of the phosphate was adsorbed into the soil sediment. The phosphateconcentration at the last estuary point was much higher than the other estuarypoint.

ConclusionsThe presence of a mangrove forest along the coast appears to act as avegetation buffer between the inland freshwater zone and the ocean ecosystem.

Figure 5.Graph of observed

phosphateconcentrations and

concentrations expectedfrom dilution hypothesis

Rio Coco Solo,Panama

139

The trees facilitate an environment where the nutrients can be removed fromthe aquatic system. The roots provide a vast surface area for biologicalorganisms to settle and absorb extra waste nutrients. The extensive rootstructure also functions in holding soil together and catching sediments frominland waters as they travel through the mangrove buffer area. These twofunctions are probably most evident in the highest estuary zone, between thelast freshwater sample point and the beginning of the first estuary samplepoints in the mid-estuary, where the river runs directly through the mud androots of the trees. The river water directly interacts with the bacteria, and thesediment is easily trapped within this zone.

There is a large physical distance between the last freshwater sample andthe first estuary sample, as can be seen by the large conductivity differencebetween these two points. If samples had been collected within the highestestuary zone, there would have been more continuity between the freshwaterand estuary samples along the salinity gradient. The first estuary sample hadalready been cleansed in the high estuary zone, resulting in extremely lowreadings.

Loading capacitiesSeveral authors have proposed the use of mangrove ecosystems as “naturalsewage treatment plants” (Nedwell, 1975; Odum and Johannes, 1975). They seethe mangrove forests as a low-cost solution for the pressing needs of sewagetreatment in tropical coastal communities. Colon is an example of such a use.They act as a natural treatment plant for the numerous villages with poorsewage systems and prevent the eutrophication that these pollutants wouldotherwise cause. This natural buffer system requires no human maintenanceand performs the function of water cleansing completely free of cost for themunicipality.

Although the functions of the mangrove forest are both economic andeffective, buffer strips such as these forests should only be considered as asecondary conservation practice for these waters. Primary protection mustcome from controlling the production of pollutants at the source and protectingthe water system from contamination.

In areas where rapid development is considered more important thanenvironmental protection, the mangrove forests are especially necessary inprotecting the health of the oceanic waters. Although there is very little reasonto believe that the mangrove forest will become saturated in the near future,long-term solutions should be focused on policy that calls for standards insewage construction and regulations in waste removal.

Contamination from sewage effluent and industrial pollution should also becontrolled in freshwater zones because of the possible effects of pathogenic andtoxic pollution to the people living near the river. Although no study has beenconducted on the concentration levels of toxic substances or pathogenic

MEQ15,2

140

bacteria, it is suspected that concentrations may be high, and therefore harmfulto humans.

Further work is needed to determine whether nutrient-enriched groundwaters reach the reefs. The study shows that the chemical level entering theocean ecosystem from the mangrove estuary areas are quite low due tobiological degradation, adsorption, and sedimentation of chemicals in the upperestuary zone. More studies will need to be done on the sediments of the estuaryto confirm the belief that much of the nutrient becomes trapped throughsediment filtration.

References

Alberts, E.E., Neibling, W.H. andMoldenhauer, W.C. (1981), “Transport of sediment nitrogen andphosphorus in runoff through cornstalk residue strips”, Soil Science Society of AmericaJournal, Vol. 45, pp. 1177-84.

Babu, K.N., Ouseph, P.P. and Padmalal, D. (2000), “Interstitial water – sediment geochemistry ofN, P and Fe and its response to overlying waters of tropical estuaries: a case from thesouthwest coast of India”, Environmental Geology, Vol. 39, pp. 633-40.

Barling, R.D. and Moore, I.D. (1994), “Role of buffer strips in management of waterway pollution– a review”, Environmental Management, Vol. 18, pp. 543-58.

Bear, K. and Pringle, C.M. (2000), “Special problems in urban river conservation: the encroachingmegalopolis”, in Boon, P.J. and Davies, B. (Eds), Global Perspectives on River Conservation:Science, Policy, and Practice, John Wiley, New York, NY, pp. 381-98.

Boto, K.G. and Bunt, J.S. (1981), “Dissolved-oxygen and pH relationships in northern Australianmangrove waterways”, Limnology and Oceanography, Vol. 26, pp. 1176-8.

Boyle, E., Collier, R., Dengler, A.T., Edmond, J.M., Ng, A.C. and Stallard, R.F. (1974), “Chemicalmass-balance in estuaries”, Geochimica et Cosmochimica Acta, Vol. 38, pp. 1719-28.

Corredor, J.E. and Morell, J.M. (1994), “Nitrate depuration of secondary sewage effluents inmangrove sediments”, Estuaries, Vol. 17, pp. 295-300.

Cubit, J.D., Getter, C.D., Jackson, J.B.C., Garrity, S.D., Caffey, H.M., Thompson, R.C., Weil, E. andMarshall, M.J. (1987), “An oil spill affecting coral reefs and mangroves on the Caribbeancoast of Panama”, in STRI (Ed.), The 1987 Oil Spill Conference, STRI, Panama City,pp. 401-6.

D’Croz, L. and Robertson, D.R. (1997), “Coastal oceanographic conditions affecting coral reefs onboth sides of the isthmus of Panama”, Proceedings of the 8th International Coral ReefSymposium, Vol. 2, pp. 2053-8.

Dillaha, T.A., Reneau, R.B., Mostaghimi, S. and Lee, D. (1989), “Vegetative filter strips foragricultural nonpoint source pollution-control”,Transactions of the Asae, Vol. 32, pp. 513-9.

Doyle, R.C., Stanton, G.C. andWolf, D.C. (1977), Effectiveness of Forest and Grass Buffer Filters inImproving the Water Quality of Manure Polluted Run-off, Paper no. 77 2501, AmericanSociety of Agricultural Engineers, St Joseph, MI.

Duke, N. (1996), “Mangrove reforestation in Panama: an evaluation of planting in areasdeforested by a large oil spill”, Restoration of Mangrove Ecosystems, International Societyof Mangrove Ecosystems, Okinawa, pp. 209-32.

Froelich, P.N. (1988), “Kinetic control of dissolved phosphate in natural rivers and estuaries – aprimer on the phosphate buffer mechanism”, Limnology and Oceanography, Vol. 33,pp. 649-68.

Rio Coco Solo,Panama

141

Heinrichsen, D. (1998), Coastal Waters of the World: Trends, Threats, and Strategies, IslandPress, Washington, DC.

Keller, B.D. and Jackson, J.B.C. (1993), Long-term Assessment of the Oil Spill at Bahia Las Minas,Panama, Synthesis Report, Volume I: Executive Summary OCS Study MMS 93-0047, USDepartment of the Interior, Mineral Management Service, Gulf of Mexico OCS Region, NewOrleans, LA, p. 129.

Nedwell, D.B. (1975), “Inorganic nitrogen-metabolism in a eutrophicated tropical mangroveestuary”, Water Research, Vol. 9, pp. 221-31.

Nwankwor, G.I. and Anyaogu, C.N. (2000), “Hydrologic characteristics of a small tropical riverinewetland at Ulakwo, Imo State, Nigeria”, Hydrogeology Journal, Vol. 8, pp. 646-53.

Odum, W.E. and Johannes, R.E. (1975), “The response of mangroves to man-induced stress”,Tropical Marine Pollution, Vol. 12, Elsevier Oceanographic Series, Elsevier, Amsterdam,pp. 52-62.

Ramanathan, A.L., Subramanian, V., Ramesh, R., Chidambaram, S. and James, A. (1999),“Environmental geochemistry of the Pichavaram mangrove ecosystem (tropical),southeast coast of India”, Environmental Geology, Vol. 37, pp. 223-33.

RiveraMonroy, V.H. and Twilley, R.R. (1996), “The relative role of denitrification andimmobilization in the fate of inorganic nitrogen in mangrove sediments (TerminosLagoon)”, Limnology and Oceanography, Vol. 41, pp. 284-96.

RiveraMonroy, V.H., Day, J.W., Twilley, R.R., Veraherrera, F. and Coronadomolina, C. (1995),“Flux of nitrogen and sediment in a fringe mangrove forest in Terminos Lagoon, Mexico”,Estuarine Coastal and Shelf Science, Vol. 40, pp. 139-60.

Schlesinger, W.H. (1997), Biogeochemistry: An Analysis of Global Change, Academic Press,San Diego, CA.

Sherman, R.E., Fahey, T.J. and Howarth, R.W. (1998), “Soil-plant interactions in a neotropicalmangrove forest: iron, phosphorus and sulfur dynamics”, Oecologia, Vol. 115, pp. 553-63.

Zuberer, D.A. and Silver, W.S. (1979), “N2-fixation (acetylene-reduction) and the microbialcolonization of mangrove roots”, New Phytologist, Vol. 82, p. 467.

MEQ15,2

142