Marine Pollution Bulletin 75 (2013) 62–68
Contents lists available at ScienceDirect
Marine Pollution Bulletin journal homepage: www.elsevier.com/locate/marpolbul
Ecotoxicological assessment of sediments from an urban marine protected area (Xixová-Japuí State Park, SP, Brazil) G.S. Araujo a,b,⇑, L.B. Moreira c, R.D. Morais d, M.B. Davanso a, T.F. Garcia a, A.C.F. Cruz a,b, D.M.S. Abessa a a
Campus Experimental do Litoral Paulista, Universidade Estadual Paulista, Praça Infante Dom Henrique, s/n, CEP 11330-900 São Vicente, SP, Brazil Instituto Oceanográfico da Universidade de São Paulo, Praça do Oceanógrafo, 191, CEP 05508-120 São Paulo, Brazil c Instituto de Ciências do Mar – UFC, Universidade Federal do Ceará, Av. da Abolição, 3207 – Meireles, CEP 60165-081 Fortaleza, CE, Brazil d Lecotox Análises Ambientais – Universidade Positivo, Rua Prof. Pedro Viriato Parigot de Souza, 5300 bloco marrom sala 209, Campo Comprido, CEP 81280-330 Curitiba, PR, Brazil b
a r t i c l e
i n f o
Keywords: Contamination Amphipod Copepod Toxicity Marine protected area Whole-sediment TIE
a b s t r a c t This study aimed to evaluate the environmental quality of the marine portion of Xixová-Japuí State Park (XJSP), an urban marine protected area, which is influenced by multiple contamination sources, by using ecotoxicological and geochemical analyses. Sediments were predominantly sandy, with low CaCO3 and organic matter contents, and presented contamination by metals (Cd, Cu, Zn). Acute toxicity was detected in three tested samples, and copepod exposed to sediments from four stations exhibited lower fecundities, despite the absence of significant effects. Contamination and toxicity seemed to be associated, suggesting that the environment is degraded and presents risks to the biota. Whole sediment TIE indicated ammonia as a main responsible for toxicity, suggesting sewage is a main contributor to sediment degradation. As external contamination sources seem to be negatively influencing the sediment quality, the park conservation objectives are not being fully reached, demanding actions to mitigate impacts. Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction A protected area (PA) may be defined as ‘‘an area of land and/or sea especially dedicated to the protection and maintenance of biological diversity, and of natural and associated cultural resources, and managed through legal or other effective means’’ (IUCN, 1994). The PA concept can be used for both terrestrial and aquatic environments. In this context, considering the importance of coastal ecosystems, there is a concern for the establishment of marine protected areas (MPAs). A MPA, in its turn, can be defined as ‘‘any area of intertidal or subtidal terrain, together with its overlying water and associated flora, fauna, historical and cultural features, which has been reserved by law or other effective means to protect part or all of the enclosed environment’’ (Kelleher et al., 1995). However, some authors have showed that the simple establishment of a MPA is not capable to protect these areas, as there is a set of actions and policies that are required to effectively provide protection to the ecosystems (Jameson et al., 2002). Threats to the protection effectiveness of MPAs concern not only to internal factors (as fishing and overcrowding) but also to external factors, as presence of pollutant sources in the MPA vicinities (Kelleher et al., ⇑ Corresponding author at: Campus Experimental do Litoral Paulista, Universidade Estadual Paulista, Praça Infante Dom Henrique, s/n, CEP 11330-900 São Vicente, SP, Brazil. Tel.: 55 13 3569 7119/7133. E-mail address:
[email protected] (G.S. Araujo). 0025-326X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.marpolbul.2013.08.005
1995). MPAs located near human activities have experienced the phenomenon of transboundary movements of chemicals (Pozo et al., 2009; Terlizzi et al., 2004) and their negative effects to biota (Davanso et al., 2013; Rodrigues et al., 2013). Thus, knowing how contamination from external sources affects the environmental quality within MPAs becomes a critical aspect that must be addressed to the proper MPA management. In Brazil, there is a national system of protected areas (NSPA) that defines different categories of PAs, based on the degrees of desired protection, and imposes thus different restrictions to the uses of natural resources within such areas (Brasil, 2000). The NSPA also establishes guidelines and rules for managing PAs and enhancing their effectiveness; however, they are not totally followed for most of Brazilian PAs, especially the coastal PAs and MPAs. In this sense, Brazilian MPAs comprise a range of situations that may be used as examples of problems of management or lacking of protection effectiveness. One particular case is the Xixová-Japuí State Park (XJSP), which is located on the SW of Santos Estuarine System (São Paulo State, Brazil), inserted into a region that has been severely affected by environmental impacts, particularly unplanned urban settlement, industrialization and port activities (São Paulo, 2010). Thus, the marine portion of XJSP represents an interesting example to be evaluated, in terms of how external sources may affect the environmental quality within an urban MPA and if ecological processes are effectively protected when the surroundings do not have restrictions to impacting activities.
G.S. Araujo et al. / Marine Pollution Bulletin 75 (2013) 62–68
The pollution assessment in aquatic environments must recognize that, after reaching aquatic systems, pollutants are spread and transported by currents and waves, increasing the range of their distribution (Salomons and Forstner, 1984). Additionally, they tend to remain in the water column for limited periods, precipitating to the bottom and accumulating in sediments (Ingersoll, 1995), from where they can be released back to the water column or affect benthic biota. As sediments represent substrate for a wide range of species and feeding sites for many predators, sediment quality assessment constitutes a reliable strategy for evaluating environmental quality. Chemical analyses are useful to identify and quantify the contaminants, whereas ecotoxicological approaches may evaluate potential effects on aquatic biota (Costa et al., 2008). In addition, advanced techniques, as the toxicity identification evaluation (TIE) approach, have been employed to determine the chemical compounds responsible for toxicity. Sediment TIEs involve a suite of procedures that are designed to decrease, increase, or transform the bioavailable fractions of sediment contaminants to assess their contributions to sample toxicity (Burgess et al., 2000). Considering the contamination sources that can potentially influence the marine portion of an urban MPA, this study aimed to assess the quality of sediments from XJSP marine portion and consequently to evaluate if this MPA is effectively protected from external impacts. To achieve that, geochemical analyses and ecotoxicological assays with benthic organisms were employed together with whole-sediment TIE (phase I), in order to identify the main chemical groups responsible to the toxicity. 2. Materials and methods 2.1. Study area As previously mentioned, the XJSP is located in the metropolitan region of ‘‘Baixada Santista’’ (MRBS), which is highly urbanized, and comprises the Port of Santos (the largest of Brazil) and the industrial complex of Cubatão, which includes major steel and petrochemical plants. The park has 901 ha, from which 600 ha comprise its terrestrial portion and 301 ha comprise the marine zone (Fig. 1). In its turn, marine portion comprises the entire area surrounding the coastline till 200 m onshore (São Paulo, 2010); starting at the São Vicente Estuary (at Santos Bay) and reaching the beach at Praia Grande, after contouring Ponta do Itaipu (Fig. 1). Historically, MRBS occupation was marked by huge conurbation between its municipalities and high levels of air, soil and water pollution (Lamparelli et al., 2001). MRBS also has a close interface with the metropolitan region of São Paulo (MRSP), which is just 50 km away, intensifying the pressures on its natural resources, particularly those related to seasonal tourism, which may triplicate the MRBS population (São Paulo, 2010). Multiple potential sources of contaminants are present in the MRBS, according to the State Environmental Agency – CETESB (Lamparelli et al., 2001); and comprise the Port of Santos, marinas, domestic and industrial landfills, industrial effluents, sewage outfalls, urban drainage, among others. Literature showed contamination and toxicity in both waters and sediments from this region (Torres et al., 2009; Lamparelli et al., 2001), including the west portion of Santos Bay in the immediate vicinity of XJSP (Abessa et al., 2008a) and nearby areas (Abessa et al., 2005, 2008b). 2.2. Sediment sampling The study area includes the marine portion of XJSP (Fig. 1), with six stations located within the park boundaries and potentially subject to the influence of contaminant sources: P1 was positioned in a low energy area and under pressure of diffuse sources; P2 and
63
P3 were under potential influence of the sewage submarine outfall of Praia Grande; P4 and P5 were located within Santos Bay, both under possible influence of pollution sources within the São Vicente estuary and the sewage submarine outfall of Santos; and P6 was in the mouth of São Vicente estuary, which receives contributions from marinas, urban drainage, industrial and domestic landfills and sewage discharges. The geographic coordinates of the stations are shown in Fig. 1. The reference sediment was collected at Engenho d’’água Beach, Ilhabela-SP (23°480 S–45°220 W), where amphipods used in the acute toxicity test were collected. The choice for using this area as reference is due to the absence of clean areas in Santos Estuarine System (Abessa et al., 2008b). Sediment samples (2-cm surficial layer) were collected in August 8, 2011, with a 0.026 m2 stainless steel ‘‘Van Veen’’ grab sampler, and immediately chilled. In the laboratory, the aliquots for the ecotoxicological assays were refrigerated at 4 °C, the samples for sedimentology were dried at 60 °C for three days, and those for chemical analyses were frozen. 2.3. Sediment properties The sediment grain size distribution was analyzed by a twostep sieving (Mudroch and MacKnight, 1994): the first step consisted in wet sieving of more than 100 g of previously dried sediments through a fine mesh (0.062 mm) to separate fine particles (silt and clay); the difference between the initial and the final weights represented the mud fraction. The second step consisted on the dry sieving of the material retained on the 0.062 mm mesh into a set of sieves (U scale) in order to separate different classes of sands. The classification method was based on the scale established by Wentworth (1922). The calcium carbonate contents determination followed the protocol described by Grant-Gross (1971), which was based on the sample digestion with hydrochloric acid (HCl) 30 volumes. The analysis of organic matter (OM) content in the sediment samples followed the ignition method (GrantGross, 1971). 2.4. Chemical analyses Metals from sediments were extracted with acqua regia (HCl:HNO3 1:3) according to USEPA 3050b protocol (USEPA, 1996). One gram aliquots of whole sediment were transferred to flasks; then 20 ml acqua regia were introduced into each flask and the mixture was heated to 90 °C for 40 min. After cooling, extracts were separated and transferred to volumetric flasks and the final volumes were adjusted to 25 mL by adding 2% HCl. The contents of Cd, Cu and Zn in the sediment samples were read by flame Atomic Absorption Spectrophotometry (model AA 6800 Shimadzu). Analysis was done in duplicate. Quantification Limits for Cd, Cu and Zn were 0.02 lg g1, 1.0 lg g1 and 1.3 lg g1, respectively. To verify the used methods, Estuarine Sediment BRCÒ 667 was evaluated, and respective recovery values were 98.14% for Cd, 119.42% for Cu and 77.68% for Zn. 2.5. Acute sediment toxicity tests The acute toxicity test with sediments was performed with the amphipod Tiburonella viscana (Melo and Abessa, 2002; ABNT, 2008). This 10-day test used 3 replicates per sediment sample. Each test chamber (1L polyethylene vessel) was prepared with approximately 2 cm of test sediment and 750 ml seawater, and received 10 amphipods. During the test, the system was maintained at 25 ± 2 °C, with constant lighting and aeration, and the animals were not fed. The number of survivors in each chamber was counted after 10 days. Missing organisms were considered dead (Melo and Abessa, 2002). At the beginning and the end of the
64
G.S. Araujo et al. / Marine Pollution Bulletin 75 (2013) 62–68
Fig. 1. Localization of Xixová-Japuí State Park and its municipalities, showing the park zonation, the sediment sampling stations and their geographical coordinates. Used datum: SAD 69.
experiments, physical–chemical parameters of overlying water were measured, as follows: pH (digital pHmeter Lutron PH-206), temperature (digital thermometer), dissolved oxygen concentration (Digimed DM4P), salinity (211 hand refractometer) and concentration of ammonia (Thermo Orion 952 electrode coupled to pH meter). 2.6. Chronic sediment toxicity tests The sediment chronic toxicity tests used the copepod Tisbe biminiensis as test-organism, based on the protocol developed by Araújo-Castro et al. (2009), with adaptations (photoperiod of 16:8 h, whole sediment) proposed for tests with other harpacticoids (Lotufo and Abessa, 2002). Aliquots of the collected samples were introduced into high density 30 ml polyethylene test-chambers (4 replicates per sample), in order to form a layer of about
0.5 cm. Then, 20 ml of filtered sea water were added to each test chamber. Ten healthy ovigerous females were introduced into each replicate. The entire test system was incubated at 25 ± 2 °C for 7 days. At the end of the experiment, the contents of each replicate were fixed by the addition of 1 ml formaldehyde (4%) and RoseBengal dye (0.1%). After two days, the sub-lethal effects (reproduction) were analyzed, by counting offspring (nauplii and copepodits) and adults. 2.7. Statistical analysis The results of ecotoxicological assays were first checked for homocedasticity and normality by Bartlett’s test and Shapiro–Wilks test, respectively. Then, data from the acute sediment toxicity test, that passed for homocedasticity were analyzed by paired Student-t’ test. Data from the chronic toxicity test did not pass for
G.S. Araujo et al. / Marine Pollution Bulletin 75 (2013) 62–68
65
homocedasticy and were analyzed by a non-parametric approach (Kruskall-Wallis followed by Dunn’s test). Geochemical data was interpreted also by observing exceedances to Canadian Sediment Quality Guidelines (Smith et al., 1996) – considering Threshold Effect Levels (TEL) and Probable Effect Levels (PEL) – and also by comparison with local sediment quality values – SQVs (Choueri et al., 2009). Ecotoxicological and geochemical data were also integrated by using a qualitative interpretation and Principal Component Analysis (PCA). Statistical and multivariate techniques were run with the help of PAST free software (Hammer et al., 2001), after log-transformation of data prior to the analysis. Fig. 3. Fecundity of Tisbe biminiensis exposed to sediments from XJSP. + indicates sign of toxicity.
3.2. Sediment chemistry 2.8. Whole-sediment TIE Whole-sediment TIE – Phase I treatments were prepared based on the protocol developed by USEPA (2007). To perform these techniques, P1 and P4 were selected as both sediments exhibited toxicity (Figs. 2 and 3) and high concentrations of Cu and Zn (Table 1). To remove ammonia, treatment with Ulva lactuca was employed: algae were collected in clean areas and maintained during four days under constant aeration and photoperiod of 16:8 h (light: dark). Then, 0.6 g of the algae was placed into the testchambers 24 h prior to the beginning of the test, to achieve equilibrium, and then removed before this. Fenton reagent was used to remove metals, by using the proportion of 1 mmol of Fe+2 to 100 mmol of H2O2, i.e., by adding 0.036 g of Fe+2 and 2.7 ml of H2O2 to each test-chamber. The reaction was maintained for 24 h to achieve equilibrium, and then the reagent was removed immediately before the introduction of test-organisms. For organic toxicants, treatment consisted in the use of coconut charcoal. The charcoal was maintained for 18 h with deionized water in a vacuum system; then 0.6 g of hydrated charcoal was added to each test-chamber, 24 h before the addition of the organisms. Then, sediments were tested for chronic toxicity to the copepod Nitocra sp, according to the protocol developed by Lotufo and Abessa (2002). Data were analyzed by a two-way analysis of variance (ANOVA), at 5% significance level (USEPA, 2007).
3. Results 3.1. Sediment properties The sediments of XJSP consisted mainly of fine sands, and their CaCO3 contents ranged between 4.93% and 7.81%. The OM contents (Table 1) were relatively low; the higher levels were observed in sediments from Santos Bay (P4, P5 and P6), between 0.48% and 0.91%, which may indicate a contribution of estuarine inputs.
Chemical analyses indicated levels of Cd exceeding TEL (0.68 lg g1) for P1, P3 and P4 samples, and exceedance of local Sediment Quality Values (SQV – 0.75 lg g1) for P4 sediments. Concentrations ranged from below detection limits to 0.96 lg g1. For copper, concentrations exceeding PEL (108 lg g1) were recorded for P1, P2, P5 and P6 sediments, and TEL (18.7 lg g1) was exceeded at P3 and P4; concentrations in all sediments also exceeded local SQVs (69 lg g1), ranging from 94.45 to 203.76 lg g1. Finally, for zinc, concentrations ranged between 46.30 and 148.12 lg g1, and samples from P1 and P4 presented concentrations above TEL (124 lg g1), whereas local SQVs (61.1 lg g1) were exceeded for sediments from P1, P2, P4 and P6. 3.3. Acute sediment toxicity In the acute sediment toxicity test, salinities of overlying waters ranged from 33 to 36; DO levels were high, between 4.56 and 5.76 mg/L; pH values ranged between 7.50 and 8.24 and unionized ammonia levels ranged between 0.038 and 0.177 mg/L. Conditions in the test chambers were considered acceptable, according to Melo and Abessa (2002). Organisms exposed to control sediment exhibited high survival (88%), within the acceptable range for the species (Melo and Abessa, 2002; ABNT, 2008), as well as those exposed to sediments from P3 and P6 (Fig. 2). On the other hand, survivals of organisms exposed to the sediments from P1, P2 and P4 were significantly low. Animals exposed to P5 sediments exhibited a relatively low survival rate (56.6%), but due to the high standard deviation, no statistical difference from the control was observed; in this case, probably such low mean survival indicates a sign of toxicity. 3.4. Chronic sediment toxicity In the chronic sediment toxicity test, physical–chemical salinities of overlying waters ranged from 33 to 35; pH values varied between 6.67 and 7.84 and unionized ammonia concentrations ranged from 0.069 to 0.129 mg/L. Dissolved Oxygen levels ranged from 1.55 to 6.73 mg/L; low values (
