Marine Pollution Bulletin 62 (2011) 1915–1919

Contents lists available at ScienceDirect

Marine Pollution Bulletin

journal homepage: www.elsevier .com/locate /marpolbul

Baseline

Eutrophication and hypoxia in four streams discharging in Guanabara Bay, RJ,Brazil, a case study

Valquiria Maria de Carvalho Aguiar ⇑, José Antônio Baptista Neto, Carlos Marclei RangelLaboratório de Geologia Marinha, Instituto de Geociências, Universidade Federal Fluminense, Av. General MiltonTavares de Souza, 24210-346 Niterói, RJ, Brazil

a r t i c l e i n f o

Keywords:NutrientsGuanabara BaySão GonçaloHypoxiaEutrophication

0025-326X/$ - see front matter � 2011 Published bydoi:10.1016/j.marpolbul.2011.04.035

⇑ Corresponding author. Tel.: +55 21 2629 5977.E-mail addresses: valquiria@io.usp.br, valquiria@i

com (V.M. de C. Aguiar).

a b s t r a c t

Four streams in the city of São Gonçalo, were sampled to evaluate their potential as sources of nutrientsto Guanabara Bay aiming to contribute with the government program to decrease the levels of pollutionin this area. Imbuaçu, Guaxindiba, Marimbondo and Brandoas streams were sampled in 2007, 2008 and2009. The streams revealed to be hipereutrophic with severe limitation of primary production by nitro-gen, as shown by the N/P molar ratio. Phosphate levels were abnormally high varying between 4.35 and130.82 lM, whereas nitrate and nitrite ranged from 0.06 to 54.05 lM and from 0.28 to 19.23 lM, respec-tively. The streams also presented severe hypoxia and anoxia, with oxygen values varying from non-detected to 3.72 ml l�1. Heavy loads of particulate suspended material were recorded in the studiedstreams, ranging between 6.00 and 400.00 mg l�1. The streams were considered inexorable sources ofnutrients, enhancing the severe eutrophication process in Guanabara Bay.

� 2011 Published by Elsevier Ltd.

Water quality of rivers is of crucial importance to the mainte-nance of biotic and ecological integrity. Rivers transport nutrientsto downstream habitats, and some of the impacts of nutrients inlarge rivers or small streams reach adjacent coastal water. Theenrichment of coastal urbanized areas by nutrients is an increasingand widely known problem all around coastal areas. Worldwide isdocumented that nutrient concentration in rivers are increasing inat least 50% (Dodds, 2006). The main consequences of this processare hypoxia/anoxia, increase of primary production followed byincrease of turbidity, decrease of phytoplankton diversity, amongothers which are direct consequences of eutrophication. The con-cept of eutrophication can be defined as an increase in the rateof supply of organic matter to an ecosystem, which most com-monly is related to nutrient enrichment enhancing the primaryproduction in the system (Newton et al., 2003).

In Brazil, the southeast coast experiences great pressure fromurbanization, intensive agriculture and aquaculture, besides theimplementation of the main industrial poles in the country. Guana-bara Bay, situated in the State of Rio de Janeiro, has been underanthropic influence for a long time, being considered one of thegreatest examples of coastal degradation in Brazil (Kjerfve et al.,1997; Knoppers et al., 1999; Crapez et al., 2000; Azevedo et al.,2004; Borges et al., 2007, 2009). The anthropic substances thatare discharged into Guanabara Bay are of several types, such as,heavy metals, pesticides, polycyclic aromatic hydrocarbons,

Elsevier Ltd.

geo.uff.br, vallaguiar@yahoo.

organochlorides, and, mostly, sewage without treatment. The laterreaches the bay through submarine outfalls and through riverineinput, such as Alcântara, Bombas, Brandoas, Imbuaçu, Marimbondo,Guaxindiba, Estrela, Paraíba do Sul, among others. Sewage reachingGuanabara Bay through riverine input is quite significant, due to thegreat lack of sanitation in the state, arriving in natura at this waterbody. The sewage charge contains great amounts of organic matterand nutrients, especially phosphorus (Khan and Ansari, 2005). Theorganic matter discharged into the bay can be deposited in the sed-iments or suffer bacterial decomposition releasing extra nutrient tothe water column, and increasing even more the primary produc-tion. Nutrient-enhanced eutrophication is a ‘‘driver’’ of hypoxia/an-oxia (Pearl, 1997), a common situation in Guanabara bay whichpresents hypoxia in the bottom layers and supersaturation ofdissolved oxygen in the surface (Kjerfve et al., 1997; Borges et al.,2009). The rivers and streams that discharge in this coastal baycross greatly urbanized areas, receiving all kinds of effluents. SãoGonçalo is the second biggest city in the State of Rio de Janeiro, wellknown for its lack of wastewater treatment plant and degradationof its rivers. Most of the rivers in this city are channeled, siltedand systematically receive solid and untreated liquid wastes, whichmake them open sewers, and, therefore, they represent a healthhazard for the marine life and human population. Besides the envi-ronmental impact on its rivers, this region is about to receive a hugepetrochemical complex, already under construction, which makesbaseline data extremely important, concerning future studies inthis area. Considering that there are several rivers and streamscrossing this region, the construction of a petrochemical complexwill increase the urbanization of this area, which, in the future,

1916 V.M. de C. Aguiar et al. / Marine Pollution Bulletin 62 (2011) 1915–1919

may impact the water quality and increase pollution issues inGuanabara Bay.

Four streams were sampled in the city of São Gonçalo; Marim-bondo, Imbuaçu, Brandoas and Guaxindiba. The three campaignsoccurred in the winter of 2007 and summer of 2008 and 2009, withthe exception of Guaxindiba that was sampled only in 2008 and2009. In Brazil, usually the summer is rainy and the winter isdry, therefore, two campaigns occurred in the wet season (2008and 2009) and one campaign in the dry season (2007). Six stationswere sampled in Brandoas and Imbuaçu, and four stations in Guax-indiba and Marimbondo (Fig. 1). Almost all the sampling pointswere allocated in channeled parts of the rivers that cross theurbanized area of São Gonçalo, exceptions were the points nearthe mouth, in Imbuaçu, Brandoas and Marimbondo. Since thereare no wastewater treatment plants, all the four rivers receive solidand untreated liquid wastes systematically.

Water samples for the analysis of dissolved oxygen (DO), chlo-rophyll-a and phaeophytin-a, particulate suspended matter (SPM),phosphate, nitrate and nitrite were collected with a Van Dorn bot-tle in the surface water of the four streams. Temperature and pHwere measured in situ with a Metrohm 827 pH meter. Samplesfor the analysis of nutrients and pigments were immediatelystored in ice and kept refrigerated until the arrival at the labora-tory, where each sample was filtered, through Millipore filters AP40. Aliquots of the filtered samples for determination of nutrientswere stored in polyethylene bottles (pre washed with HCl 10%and rinsed three times with Milli-Q water) and frozen at �20 �Cuntil the moment of analysis. The filters were dried in a muffleat 60 �C and the determination of suspended particulate matterproceeded according to Strickland and Parsons (1968). The filtersfor the determination of chlorophyll-a and phaeopigments were

Fig. 1. Sampling stations for Imbuaçu (I1–I6), Marimbondo (M1–M4), Guaxindiba(G1–G4) and Brandoas (B1–B6).

extracted with ketone 90% in the dark and the analysis was carriedon according to Strickland and Parsons (1968). The determinationof dissolved oxygen was performed through Winkler method de-scribed in Grasshoff et al. (1999) with a Metrohm burette. Dis-solved inorganic phosphorus, and dissolved nitrate and nitritewere analyzed with a Perkin Elmer spectrophotometer accordingto Grasshoff et al. (1999). Data analysis was complemented byprincipal component analysis (PCA), Spearman correlation andKruskall–Wallis analysis.

All the four streams were channeled and silted in the samplingbranches. Dark water and clandestine sewage pipes were observedalong the four streams, pointing to a systematically input ofdomestic wastes. In 2007, temperature values varied between22.60 and 29.50 �C. In 2008 and 2009, the maximum temperatureswere higher, with values from 25.80 to 32.80 �C and between 23.50and 30.50 �C, respectively, reflecting the difference between winterand summer campaigns (Table 1).

The pH varied greatly along the streams and among the sam-pling periods (Table 1). The pH values varied between 6.68 and8.64 in 2007, from 6.70 to 7.76 in 2008 and from 6.57 to 8.09 in2009. The higher pH values were observed in the stations closerto Guanabara Bay, at the mouth of the streams, showing the influ-ence of marine water. Brandoas and Marimbondo presented thelower pH values in the campaigns occurred in 2008 and 2009,which is probably a consequence of higher input of organic matterin these streams, decreasing the pH during remineralizationprocess.

The streams were also heavily loaded with suspended particu-late matter (Table 1), varying between 60.00 and 400.00 mg l�1

in 2007, with the highest values in Brandoas and Imbuaçu, espe-cially in the stations closer to Guanabara Bay (B6 and I6). In2008, PSM decreased varying from 6.00 to 80.00 mg l�1, with themaximum values encountered in Brandoas and Imbuaçu. In thesubsequent year, values of PSM varied from 8.00 to 362.00 mg l�1,and the highest concentrations were registered in Marimbondo.

Analysis of dissolved oxygen was not performed in the cam-paign of 2007 due to technical problems. Severe hypoxia was ob-served in all the sampled streams (Table 1), and anoxia wasregistered in Marimbondo, Imbuaçu and Brandoas. The causes ofhypoxia/anoxia are related to increasing nutrient loads, althoughphysical factors may influence the timing and extent of this condi-tion (Conley et al., 2007). Brandoas stations presented values ofoxygen between 0 and 3.72 ml l�1 in the year of 2008, with thehigher values in the most internal stations, whereas in the rest ofthem, hypoxia and anoxia were characterized. In the following yearthe Brandoas water were completely anoxic, with non-detectabledissolved oxygen at all stations. The same pattern was observedin Marimbondo, with oxygen values between 0 and 2.32 ml l�1 in2008 and total anoxia in the subsequent year. In Guaxindiba, thewater were considered hypoxic in both years, with concentrationsbetween 0.50 and 2.92 ml l�1 in 2008 and between 0.59 and1.85 ml l�1 in 2009. In Imbuaçu the values of dissolved oxygen var-ied from 0.71 to 1.90 ml l�1 in 2008 and from 0 to 1.21 ml l�1 in2009, characterizing hypoxia in both periods. Anoxia in Imbuaçuwas observed only in the year of 2009, in the two stations closerto Guanabara Bay (I5 and I6). Hypoxia is a complex phenomenonthat arises from the convergence of several factors, some of whichare altered by human activities. Anoxia is rare in the water columnof natural rivers and streams, for they are relatively shallow andhave significant greater rates of atmospheric exchange comparedto lentic systems. Therefore, the biota of these systems will hardlydeplete the dissolved oxygen in the water column, unless a sub-stantial input of organic matter and nutrients occur, to supportvery rapid rates of heterotrophic activity (Dodds, 2006). In Imbu-açu, hypoxia was probably a consequence not only from the inputof domestic effluents, but also from the mangrove intense produc-

Table 1Mean, standard error and range for temperature, pH, dissolved oxygen, suspended particulate material, chlorophyll-a, phaeophytin-a, nitrate, phosphate and nitrite in the foursampled streams between 2007 and 2009.

Streams Statistics T (�C) pH DO (ml l�1) SPM(mg l�1)

Chlo-a(lg l�1)

Phae-a(lg l�1)

NO�3(lM)

NO�2(lM)

PO3�4

(lM)

Brandoas Mean 26.91 7.30 0.82 107.28 15.00 33.01 5.82 1.59 66.83SE 0.26 0.10 1.33 22.33 5.35 7.44 2.13 0.33 7.30Min–max 25.00–29.10 6.68–8.18 0–3.72 21.88–400.00 0–35.60 2.14–80.99 0.06–23.02 0.28–5.18 38.88–130.82

Imbuaçu Mean 27.18 7.52 0.86 88.53 47.30 41.03 15.04 4.01 46.29SE 0.60 0.09 0.18 25.11 20.79 10.37 5.18 1.10 6.21Min–max 23.20–31.00 7.05–8.20 0–1.90 8.00–40.00 5.34–190.71 0.89–102.80 2.80–54.05 0.38–15.36 4.35–86.39

Marimbondo Mean 28.35 7.48 0.50 97.38 11.05 44.73 3.94 1.69 58.81SE 0.80 0.16 0.28 28.38 2.12 15.16 1.27 0.52 9.32Min–max 22.60–32.40 6.74–8.64 0–2.32 20.00–362.00 5.34–16.02 9.35–120.68 0.71–11.82 0.40–6.37 4.80–128.32

Guaxindiba Mean 28.60 6.87 1.45 27.70 7.92 416.73 10.40 4.65 37.66SE 1.06 0.07 0.31 4.68 0.29 387.74 3.64 2.21 4.67Min–max 25.40–32.80 6.57–7.20 0.50–2.95 6.00–44.64 7.63–8.21 28.99–804.47 2.79–20.20 0.44–19.23 17.29–59.21

V.M. de C. Aguiar et al. / Marine Pollution Bulletin 62 (2011) 1915–1919 1917

tion, with a higher demand of dissolved oxygen for the decompo-sition of organic matter especially at the stations situated closerto the mouth. Hypoxia also typically occurs during periods of verylow discharge, or in rivers with limited flushing rates. Hypoxia wasobserved during the rainy season in both years (2008 and 2009),when the fluvial discharge is higher, which lead us to suppose thatanoxia in the studied streams was caused mostly by the degrada-tion of anthropic organic matter.

The concentrations of nutrients found in the studied streamswere so high that are hardly found in polluted aquatic ecosystems.Natural concentrations of phosphate and nitrate in streams usuallyvary between 0.02 and 2.60 lM and between 0.81 and 3.22 lM,respectively (Meybeck and Helmer, 1989).

Phosphate concentrations were abnormally high (Table 1). Inthe year of 2007 the concentrations of phosphate varied from22.49 to 130.82 lM, with the highest value recorded in the innerstation of Brandoas (B1). In the same period nitrate and nitrite var-ied from 0.06 to 27.63 lM and from 0.32 to 10.80 lM, respectively.The highest values of nitrogen forms were exhibited by station 2 inImbuaçu stream.

In 2008, although the concentrations of phosphate decreased,they were still considered very high, ranging between 1.37 to78.23 lM, with the highest value registered in station 2 in Brand-oas. Nitrate concentrations varied from not detected, at Brandoas(B3), to 54.05 lM, in station I2, located in Imbuaçu. Concentrationsof nitrite for the same period varied between 0.28 (B3) and15.36 lM (I1).

In 2009 concentrations of phosphate varied between 16.80 and58.83 lM, with the highest concentration in a inner station inBrandoas (B2), whereas nitrite varied between 0.9 and 19.23 lM,with highest contents in inner stations of Imbuaçu (I1) and Brand-oas (B1).

Such high concentrations of nutrients undoubtedly reflectanthropic input in the studied streams. The concentrations ofphosphorus found in this study are hardly comparable to contentsfound in other polluted aquatic systems, such as Wonokromo river,where Jennerjahn et al. (2004) found concentrations of phosphorusup to 3.0 lM. Kucuksezgin et al. (2006) registered concentrationsof phosphate between 0.01 and 10 lM in Izmir Bay (Turkey), whichreceives the discharge of several streams and domestic sewageoutlets. Pereira-Filho et al. (2001) registered phosphate between0.13 and 3.23 lM in river Camboriú estuary (Brazil). Aguiar andBraga (2007) found phosphate concentrations between 0.40 and19.17 lM in Santos estuary, a well known polluted coastal areain Brazil. The contents of phosphate found in this study are onlysimilar to the ones found by Tas et al. (2006) in the heavily pollutedGolden Horn estuary (Turkey), between 0.3 to 135 lM.

The elevated concentrations of phosphorus may have an inter-nal source as well, since anoxia in the water column allow therelease of phosphorus from sediments making the bottom waterrich with this nutrient.

Nitrate and nitrite were also considered very elevated along thestudied streams (Table 1). Nitrate varied from 0.06 to 27.63 lM in2007 and from 0.71 to 54.05 lM in 2008. Nitrite ranged between0.32 and 10.80 lM in 2007, from 0.28 to 15.36 lM in 2008, andfrom 0.99 to 19.23 lM in 2009. In both sampling periods the higherconcentrations of nitrate occurred in Imbuaçu in station I1 (2007),and in station I6 (2008). Among the studied streams, Imbuaçu wasnitrate richest, and the higher contents of nitrite were registered inGuaxindiba followed by Imbuaçu. Smaller streams tend to presentmore temporal and spatial variability than rivers. Generally,responses of primary production in rivers should be less than insmall streams because of increased turbidity and light limitation.

Despite the high concentrations of phosphate, phytoplanktonicbiomass, measured through chlorophyll a, was very low (Table 1),with the exception of some blooms observed in Imbuaçu, with con-centrations of 190.71 lM in station 5 in 2007 and 106.80 lM instation 6 in 2008, and in Brandoas, with a concentration of35.60 lM. In Imbuaçu, the phytoplanktonic blooms were observedcloser to the mouth in the mangrove area, suggesting that besidesthe anthropic discharge, the litter fall contribution must also beconsidered, whereas in Brandoas, the bloom was registered inthe urban area, which lead us to conclude that the primary produc-tion was solely due to anthropic influence. On the other hand, thephaeophytin-a concentrations were quite elevated in most stationsthroughout the sampling periods (Table 1), varying from 0.89 to120.68 lM in 2007, from 9.35 to 71.71 lM in 2008 and between2.14 and 804.47 lM in 2009. The high concentrations of phaeophy-tin-a accompanied by the low levels of oxygen, point to the severedegradation scenario in the studied streams. The dark water pres-ent in the streams may act as a limitation for primary production,but the strongest factor is the nitrogen limitation as shown by N/Pratio, calculated as the sum of nitrate and nitrite/phosphate (Table2), in accordance with the belief that phosphorus is more likely tobe deficient, and limit the primary production of any region of theearth’s surface proposed by Hutchinson (1967) and therefore, themost common cause of eutrophication in freshwater lakes, reser-voirs and streams and headwater of estuarine systems (Correll,1999).

According to the classification proposed by Carlson (1977) andadapted by Toledo et al. (1983) for tropical ecosystems, the trophicstate of streams can be defined through the concentrations ofphosphorus.Based on the following equation the trophic index(TI) was calculated for each stream in the present study:

Table 3Kruskall–Wallis test for significant differences among the campaigns 2007–2009 andamong the four streams.

Variables Campaigns Streams

p value p value

Temperature 0.0011 0.0702pH 0.0085 0.2640Suspended particulate material 0.0000 0.4997Dissolved oxygen 0.0281 0.2553Clorophyll a 0.2304 0.1924Phaeophytin a 0.0001 0.9941Phosphate 0.0000 0.1274Nitrite 0.0161 0.1002Nitrate 0.6662 0.4643

1918 V.M. de C. Aguiar et al. / Marine Pollution Bulletin 62 (2011) 1915–1919

TIðPÞ ¼ 10 6� Inð80;32=PÞIn2

� �� �

The concentrations of phosphorus were transformed in lg.l�1 asrequested by the formula. According to Toledo et al. (1983),24 < TI 6 44 describe oligotrophic water; 44 < TI 6 54, mesotro-phic; 54 < TI 6 74, eutrophic and TI P 74, hipereutrophic.

Results showed that, along the sampling period, all stationswere considered hipereutrophic (Table 2), with no expressive dif-ferences among the years, pointing to systematic anthropic input,mostly through domestic sewage, which contains approximately50% of human waste and 20–30% of detergents (Khan and Ansari,2005).

The results were tested for significant differences, consideringthe particularities between them. The Kruskall–Wallis analysisrevealed significant differences between the three sampling peri-ods (2007–2009) for temperature, suspended particulate material,dissolved oxygen, phaeophytin-a, phosphate and nitrite (Table 3).The same test did not reveal significant differences for the samevariables among the streams (Table 3), suggesting that the anthro-pic input is equivalent in the studied streams.

Principal component analysis considered the three samplingperiods altogether and revealed 49.61% of total variance. The firsttwo components explained only 29.51% of the total variance(Fig. 2), with minimum differences between them. The first axis ac-counted for 20.10% of the total variance, and correlated positivelywith temperature, pH, SPM, chlorophyll-a, nitrite and nitrate, beingthis last one the most significant component. The second axis ac-counted for 20.10% of the total variance, and correlated positivelywith pH, chlorophyll-a, SPM and phosphate, being this nutrientthe most significant component for the secondary variance. Bothaxis showed the main influence of nitrate and phosphate in thesample variance, reflecting a direct effect of anthropic input inthe streams, and the decrease of dissolved oxygen related to it.

Spearman correlation (Table 4), exhibited significant positivecorrelation between pH and chlorophyll-a, suggesting a slight risein pH during photosynthesis, what indeed was observed in thesampling points where phytoplanktonic blooms were observed.Suspended particulate matter correlated significant and positivelywith phosphate and with phaeophytin a, and negatively with dis-solved oxygen, suggesting that the increase in primary production

Table 2Nitrogen and phosphorus molar ratio and trophic index in Brandoas, Marimbondo,Guaxindiba and Imbuaçu streams.

Stream Station 2007 2008 2009

N/P TI N/P TI N/P TI

Brandoas 1 0.03 133 0.70 116 0.04 117Brandoas 2 0.01 131 0.29 126 0.04 121Brandoas 3 0.09 122 0.04 122 0.05 117Brandoas 4 0.02 132 0.10 117 0.06 117Brandoas 5 0.07 130 0.04 118 0.05 120Brandoas 6 0.07 122 0.04 119 0.05 115Marimbondo 1 0.10 133 0.03 117 0.03 119Marimbondo 2 0.05 121 0.03 123 0.03 118Marimbondo 3 0.39 107 0.12 119 1.33 85Marimbondo 4 0.04 129 0.11 116 0.02 117aGuaxindiba 1 – – 0.30 114 0.07 113aGuaxindiba 2 – – 0.87 112 0.60 113aGuaxindiba 3 – – 0.06 121 0.27 104aGuaxindiba 4 – – 0.18 120 0.06 116Imbuaçu 1 0.18 127 4.13 103 0.07 118Imbuaçu 2 0.07 126 0.14 112 0.07 116Imbuaçu 3 0.04 124 0.10 112 0.06 114Imbuaçu 4 0.18 121 – – 0.08 114Imbuaçu 5 0.26 125 0.78 110 0.10 109Imbuaçu 6 0.52 125 – – 2.25 84

a Not sampled in 2007.

due to the excess of phosphorus, is followed by degradation cor-roborating the high contents of phaeophytin a and SPM causinglow levels of DO in the streams. This hypothesis is corroboratedby the significant and positive correlation between phosphateand phaeophytin a. The positive and significant correlation ofnitrate with chlorophyll a and dissolved oxygen, confirms the lim-itation of primary production by nitrogen. Denitrification is a formof anaerobic microbial respiration in which nitrate is reduced tonitrous oxide or dinitrogen, and is a major sink for bioavailablenitrogen (Childs et al., 2002). Denitrification can also alter stoichi-ometric nutrient ratios and drive systems to nitrogen limitations(McCarthy et al., 2008). Water hypoxia or anoxia favors denitrifica-tion if nitrate is available for facultative bacteria, since they can gettheir oxygen directly by taking nitrogen out of water or by taking itoff from nitrate molecules and therefore, in depleted oxygen water,nitrate becomes the primary oxygen source for microorganisms.Denitrification also requires carbon to occur, and the organic mat-ter present in raw wastewater is usually enough for this transfor-mation, which make the studied streams proper sites for thisprocess, since they receive a great amount of domestic input.Despite the favoring conditions for denitrification, made by anoxiaand high concentrations of nitrate, this process does not seem toproceed all the way to N2, since NO�2 appears in elevated

Fig. 2. Principal factor analysis (2-factors) performed with the data from thestreams for 2007, 2008 and 2009.

Table 4Results of Spearman correlation analysis (p < 0.050).

T pH DO SPM Clor a Phae a NO�3 PO3�4

NO�2

T 1.00 �0.26 �0.04 �0.18 0.00 �0.12 �0.08 �0.28 �0.12pH 1.00 0.05 0.44 0.29 0.33 0.29 0.21 0.07DO 1.00 �0.32 �0.17 �0.09 0.72 �0.10 0.10SPM 1.00 0.14 0.45 0.03 0.49 �0.19Clor a 1.00 0.23 0.35 �0.05 0.25Phae a 1.00 0.12 0.43 �0.06NO�3 1.00 �0.14 0.49

PO3�4

1.00 �0.47

NO�2 1.00

V.M. de C. Aguiar et al. / Marine Pollution Bulletin 62 (2011) 1915–1919 1919

concentrations along all the streams. The explanation for the highcontents of nitrite in the water may be the presence of sulfideswhich are highly favored by anoxia, inhibiting denitrification andreleasing intermediates such as NO�2 and N2O into the water(Childs et al., 2002).

The studied streams revealed themselves as major sources ofnutrients, through raw sewage, to Guanabara Bay, showing astrong suppression of primary production contrasting with phyto-planktonic blooms along the sampling area, coupled with heavyloads of phosphate, nitrate and nitrite. Besides the economicimportance of Guanabara Bay, this area is about to become thestage of important world events in a near future, what might bea chance to adopt measures to decrease the levels of pollution inthis ecosystem. The studied area urges for environmental monitor-ing and recovery and further investigation of pollutants sources toGuanabara Bay, since it is has already collapsed from the ecologicalpoint of view.

Acknowledgments

The authors would like to thank the financial support of Fun-dação de Amparo à Pesquisa do Rio de Janeiro (FAPERJ) Processn. 151.155/07, and to Universidade Federal Fluminense (UFF),and also to Paula Falheiros, for the map.

References

Aguiar, V.M.C., Braga, E.S., 2007. Seasonal and tidal variability of phosphorus along asalinity gradient in the heavily polluted estuarine system of Santos/S.Vicente –São Paulo, Brazil. Marine Pollution Bulletin 54, 464–488.

Azevedo, L.A., Bruning, I.M.R.A., Moreira, I., 2004. Hydrocarbon contamination inmussels from Guanabara Bay. Marine Pollution Bulletin 49, 1109–1126.

Borges, A.C., Dias, J.C., Machado, W., Patchineelam, S.R., Sella, S.M., 2007.Distribuição espacial de ferro, cobre e chumbo em sedimentos de manguezalem um gradiente de degradação na Baía de Guanabara (Estado do Rio deJaneiro). Química Nova 30 (1).

Borges, A.C., Sanders, C.J., Santos, H.L.R., Araripe, D.R., Machado, W., Patchneelam,S.R., 2009. Eutrophication history of Guanabara Bay (SE, Brazil) recorded byphosphorus flux to sediments from a degraded mangrove area. Marine PollutionBulletin 58, 1739–1765.

Carlson, R.E., 1977. A trophic state index for lakes. Limnology and Oceanography 22(2), 361–369.

Childs, C.R., Rabalais, N.N., Turner, E., Proctor, L.M., 2002. Sediment denitrification inthe Gulf of Mexico zone of hypoxia. Marine Ecology Progress Series 240, 285–290.

Conley, D.J., Cartensen, J., Aertebjerg, G., Christensen, P.B., Dalsgaard, T., Hansen,J.L.S., Josefson, A.B., 2007. Long-term eutrophication and impacts of hypoxia inDanish coastal waters. Ecological Applications 17 (5), 165–184.

Correll, D.L., 1999. Phosphorus: a rate limiting nutrient in surface waters. PoultryScience 78, 674–682.

Crapez, M.A.C., Tosta, Z.T., Bispo, M.G.S., Pereira, D.C., 2000. Acute and chronicimpacts caused by aromatic hydrocarbons on bacterial communities at BoaViagem and Forte do Rio Branco Beaches, Guanabara Bay, Brazil. EnvironmentalPollution 108, 291–295.

Dodds, W.K., 2006. Eutrophication and trophic state in rivers and streams.Limnology and Oceanography 51 (1), 671–680.

Grasshoff, K., Ehrhardt, M., Kremling, K., 1999. Methods of Seawater Analysis, thirded. Weinhein.

Hutchinson, G.E., 1967. A Treatise on Limnology: Introduction to Lake Biology andthe Limnoplankton, vol. 2. John Wiley & Sons, New York, pp. 11–15.

Jennerjahn, T.C., Ittekkot, V., Klopper, S., Adi, S., Nugroho, A.P., Sudiana, N., Yusmal,A., PrihartantoGaye-Haake, B., 2004. Biogeochemistry of a tropical river affectedby human activities in its catchment: Brantas River estuary and coastal watersof Madura Strait, Java, Indonesia. Estuarien Coastal and Shelf Science 60, 503–514.

Khan, F.A., Ansari, A.A., 2005. Eutrophication: an ecological vision. The BotanicalReview 71 (4), 449–482.

Kjerfve, B., Ribeiro, C.A., Dias, G.T.M., Filippo, A., Quaresma, V.S., 1997.Oceanographic characteristics of an impacted coastal bay: Baía de Guanabara,Rio de Janeiro, Brazil. Continental Shelf Research 17, 1609–1643.

Knoppers, B., Carmouze, J.P., Moreira-Turq, P.F., 1999. Nutrient dynamics,metabolism and eutrophication of laggons along the East Fluminense Coast,State of Rio de Janeiro, Brazil. Environmental Geochemistry of Coastal LaggonSystems, Rio de Janeiro, Brazil, Série Geoquímica Ambiental 6, 123–154.

Kucuksezgin, F., Kontas, A., Altay, O., Uluturhan, E., Darilmaz, E., 2006. Assessmentof marine pollution in Izmir bay: nutrient, heavy metal and total hydrocarbonconcentrations. Environmental International 32, 41–51.

McCarthy, M.J., McNeal, K.S., Morse, J.W., Gardner, W.S., 2008. Bottom-waterhypoxia effects on sediment-water interface nitrogen transformations in aseasonally hypoxic shallow bay (Corpus Christi Bay, TX, USA). Estuaries andCoasts 31, 521–531.

Meybeck, M., Helmer, R., 1989. The quality of rivers-fom pristine stage to globalpollution. Global and Planetary Change 75, 283–309.

Newton, A., Icely, J.D., Falcao, M., Nobre, A., Nunes, J.P., Ferreira, J.G., Vale, C., 2003.Evaluation of eutrophication in the Ria Formosa coastal lagool, Portugal.Continental Shelf Research 23, 1945–1961.

Pearl, H.W., 1997. Coastal eutrofication and harmful algal blooms: importance ofatmospheric deposition and groundwater as ‘‘new’’ nitrogen and other nutrientsources. Limnology and Oceanography 42, 1154–1165.

Pereira-Filho, J., Schettini, C.A.F., Rörig, L., Siegle, E., 2001. Intratidal variation andnet transport of dissolved inorganic nutrient, POC and chlorophyll a in theCamburiú River Estuary. Estuarine Coastal and Shelf Science 53, 249–257.

Strickland, J.D., Parsons, T.R., 1968. A Practical Handbook of Seawater Analysis.Tas, S., Okus, E., Aslan-Yilmaz, A., 2006. The blooms of a cyanobacterium Microcystis

cf. aeruginosa in a severely polluted estuary, the Golden Horn, Turkey.Continental Shelf Research 68, 593–599.

Toledo Jr., A.P., Talarico, M., Chinez, S.J., Agudo, E.G., 1983. Aplicação de modelossimplificados para a avliação de processo da eutrofização em lagos ereservatórios tropicais. Congresso Brasileiro de Engenharia Sanitária, Camboriú.