Science of the Total Environment 482–483 (2014) 62–70

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Land-use impacts on fatty acid profiles of suspended particulate organic matter along a larger tropical river I.G. Boëchat a,⁎, A. Krüger b,1, R.C. Chaves a,c,2, D. Graeber d,3, B. Gücker a a

Department of Biosystems Engineering, Federal University of São João del-Rei, Praça Dom Helvécio 74, 36301-160 São João del Rei, MG, Brazil Leibniz-Institute of Freshwater Ecology and Inland Fisheries — IGB, Müggelseedamm 301, 12587 Berlin, Germany Graduate Program of Bioengineering, Federal University of São João del-Rei, Praça Dom Helvécio 74, 36301-160 São João del Rei, MG, Brazil d Department of Bioscience — Freshwater Ecology, Aarhus University, Vejlsøvej 25, 8600 Silkeborg, Denmark b c

H I G H L I G H T S • • • •

We investigated the efficiency of fatty acids (FAs) as indicators of land-use impacts. FA profiles suggested considerable inputs of domestic sewage. Saturated and sewage FAs were correlated with urbanization in a larger tropical river. FAs were the most efficient indicators of urban impact.

a r t i c l e

i n f o

Article history: Received 17 December 2013 Received in revised form 22 February 2014 Accepted 24 February 2014 Available online 15 March 2014 Keywords: River monitoring Suspended particulate organic matter Fatty acid biomarkers Urbanization Tropical rivers

a b s t r a c t Land-use change, such as agricultural expansion and urbanization, can affect riverine biological diversity and ecosystem functioning. Identifying the major stressors associated with catchment land-use change is a prerequisite for devising successful river conservation and restoration strategies. Here, we analyzed land-use effects on the fatty acid (FA) composition and concentrations in suspended particulate organic matter (SPOM) along a fourth-order tropical river, the Rio das Mortes. Thereby, we aimed at testing the potential of fatty acids in riverine suspended particulate organic matter (SPOM-FAs) as indicators of land-use change in tropical catchments, and at identifying major human impacts on the biochemical composition of SPOM, which represents an important basal energy and organic matter resource for aquatic consumers. River water SPOM and total FA concentrations ranged between 2.8 and 10.2 mg dry weight (DW) L−1 and between 130.6 and 268.2 μg DW L−1, respectively, in our study. Urbanization was the only land-use category correlating with both FA composition and concentrations, despite its low contribution to whole catchment (1.5–5.6%) and riparian buffer land cover (1.7–6.6%). Higher concentrations of saturated FAs, especially C16:0 and C18:0, which are the main components of domestic sewage, were observed at sampling stations downstream of urban centers, and were highly correlated to urbanization, especially within the 60 m riparian buffer zone. Compared to water chemical characteristics (inorganic nutrients, dissolved oxygen, pH, and specific conductance) and river habitat structural integrity, FA variables exhibited a higher variability along the investigated river and were more strongly correlated to urban land use, suggesting that SPOM-FA profiles may be an efficient indicator of urban land-use impacts on larger tropical rivers. High total FA concentrations in the SPOM of urbanized tropical rivers may represent high-energy biochemical subsidies to food webs, potentially leading to changes in functional ecosystem characteristics, such as bacterial and suspension-feeder production. © 2014 Elsevier B.V. All rights reserved.

1. Introduction

⁎ Corresponding author. Tel.: +55 32 33792541. E-mail address: [email protected] (I.G. Boëchat). 1 Tel.: +49 30 64181735. 2 Tel.: +55 32 33792541. 3 Tel.: +45 87158562.

http://dx.doi.org/10.1016/j.scitotenv.2014.02.111 0048-9697/© 2014 Elsevier B.V. All rights reserved.

Streams and rivers supply important ecosystem services, such as drinking water supply, fish production, opportunity for recreational activities, and the collection, transport and processing of pollutants and contaminants originating from the surrounding landscape (Millennium Ecosystem Assessment, 2005). Land-use changes in river basins, as a result of agricultural intensification and expansion, as well as urbanization, can affect various characteristics of river ecosystem

I.G. Boëchat et al. / Science of the Total Environment 482–483 (2014) 62–70

integrity, such as water quality, community structure, and primary and secondary production, organic matter decomposition, ecosystem metabolism and energy fluxes (Allan, 2004; Young et al., 2008; Rosa et al., 2013). Accordingly, one of the greatest challenges faced by aquatic ecologists nowadays is to identify the specific impact mechanisms of land-use change and to determine their effects on abiotic, biotic, and functional characteristics of lotic ecosystems. Besides altering channel morphology, habitat diversity and hydrodynamic features (Gücker et al., 2009), agricultural land use can lead to organic and nutrient enrichment in lotic systems, due to fertilization practices, cattle feces, and riparian clearcutting, with profound consequences for stream ecosystem functioning (Sweeney et al., 2004; Silva-Junior et al., 2014). Urbanization also affects the integrity of lotic ecosystems, by altering water quality, channel structure and habitat diversity, and biotic community structure among others (Chadwick et al., 2006). Untreated sewage discharge resulting from the absence of waste water treatment plants, is a common reality in many tropical countries, including Brazil, and is among the most severe problems related to urbanization, contributing substantial organic matter and nutrient loads to streams and rivers (Dodds, 2006; Smith et al., 2006; Gücker et al., 2006). Another frequent problem associated to the expansion of urban centers is the enhanced surface runoff generated by impervious surfaces, carrying substantial contaminant loads and causing hydrodynamic stress to the biotic community (Walsh et al., 2005). Larger rivers subjected to land-use change may exhibit changes in the quantity and quality of transported particulate organic carbon, depending on the type of land cover change (Young and Huryn, 1999). Two reviews on the global carbon cycle estimated that inland waters receive around 2.9 Pg C y−1 of organic carbon from the terrestrial landscape (Battin et al., 2008; Tranvik et al., 2009), of which approximately 27% are particulate organic carbon (Alvarez-Cobelas et al., 2012). Considering the small fraction of the earth surface covered by freshwater ecosystems, this rate is surprisingly high and may represent a large fraction of total terrestrial net ecosystem production (Cole et al., 2007). Recent estimates of total C dioxide emissions from inland waters of around 2.1 Pg C y− 1 (1.8 Pg C y−1 of those from running waters; Raymond et al., 2013) may point to a high organic C transformation potential by aquatic microbial communities. This transformation potential may depend on C lability and origin (Guillemette and Del Giorgio, 2011), and may affect both the quantity and the stoichiometric and biochemical quality of terrestrial organic C that is exported downstream. Thus, the roles of different stressors related to agricultural and urban land use in catchment organic matter transport and processing must be disentangled in order to understand the mechanisms of terrestrial– aquatic coupling and the role of larger rivers in assimilating and processing terrestrial organic material. Fatty acids (FAs) are an important class of lipids, because they represent biomarkers of aquatic organisms (Kaneda, 1991; Ahlgren et al., 1992) and their limitation, synthesis and metabolism have profound consequences for consumer physiology and trophic interactions (Müller-Navarra et al., 2004; Boëchat et al., 2005, 2006). Moreover, the role of FAs in ecosystems has gained attention in recent years, as a result of the necessity to couple ecosystem services to human health, and the importance of essential FAs from aquatic ecosystems for human nutrition has been emphasized (Lands, 2009). However, environmental studies involving FAs in aquatic ecosystems commonly focus on their use as markers for bacterial or algal presence and origin of organic matter (terrestrial, aquatic), as well as indicators of prey nutritional quality for consumers (Arts et al., 2009). FAs have rarely been tested as indicators of human impacts on ecosystems, but could be used as markers for urbanization impacts, such as sewage discharge (Jardé et al., 2005) or as markers for changes in key ecosystem processes, such as the response of primary production to enhanced nutrient availability in agricultural streams (Boëchat et al., 2011). In this study, we analyzed land-use effects on river SPOM-FAs in a larger tropical catchment. We aimed at identifying major human

63

impacts on the biochemical quality of SPOM, which represents an important basal energy and organic matter source for aquatic consumers, and may affect several aspects of river ecosystem functioning, such nutrient spiraling, ecosystem respiration, secondary production and energy flux. We hypothesized that catchment land use from river's headwaters to its mouth not only leads to direct anthropogenic matter inputs, but also affects the structural integrity of the surrounding environment, thereby altering natural patterns of allochthonous matter inputs. As a consequence, FA composition and concentrations of SPOM should also change, with potential consequences for trophic linkages and energy fluxes. Moreover, we expected urban land use to directly and profoundly affect FA profiles, for instance due to direct inputs of cooking oils and fecal FAs by sewage discharge. Thus, we hypothesized that FAs would be good indicators of land-use impacts on tropical river ecosystem health, and could be a useful tool for monitoring and restoration efforts in tropical catchments, for which land-use maps are often not available or outdated.

2. Materials and methods 2.1. Study site The present study was conducted in the Rio das Mortes, a fourthorder tributary to the Rio Grande, in the upper Rio Paraná basin. The Rio das Mortes catchment covers two mesoregions in the Brazilian Federal State of Minas Gerais, the mesoregion of Barbacena and the mesoregion of São João del-Rei, with 27 cities (IBGE, 2007; see Fig. 1 for urban centers in the Rio das Mortes catchment). Land cover in the Rio das Mortes catchment is dominated by native vegetation (52.0% of total catchment area; Fig. S1), agriculture (mainly pasture, 30.2%; crops, 5.6%; open soil and burnt areas, 7.3%; and eucalyptus, 1.3%) and urban land cover (urban areas, 1.2%; roads, 2.0%; railways, 0.2%; and mines, 0.1%). Efficient sewage treatment is largely absent in the Rio das Mortes catchment. Thus, urban land cover had a much larger impact on the water quality of headwater streams of this catchment than its small contribution to total catchment area suggested (Silva-Junior et al., 2014). In order to test the hypothesis that different types of land use will drive changes in SPOM-FA composition and concentration, we selected 11 sampling stations, from the river's headwater to its mouth, including sites located upstream and downstream of urban centers, farmland and pasture areas (Fig. 1). Sampling stations were located in the middle of the river. All sites were sampled in triplicate in the middle of the water column during four campaigns carried out in May and September 2010, and March and June 2011, covering the dry and rainy season.

2.2. Land-use characterization of the Rio das Mortes catchment area Land use of the Rio das Mortes catchment was categorized into three land cover types, i.e. natural, agricultural and urban land cover, in a previous study (Silva-Junior et al., 2014). Riparian buffer zones of 60, 120, 180, 240, and 300 m width upstream of each sampling site were delimited using the open source software QGIS 2.0. Sub-catchment upstream of each study site was delimited based on hypsometric maps using QGIS 2.0 (Quantum GIS Development Team, 2013). Subsequently, land cover distributions for these sub-catchments as well as the riparian buffer zones were calculated from the previously compiled land-use map. As the relative contributions of each land cover type in riparian buffer zones of different width were highly correlated among each other, and with whole catchment land cover (all R2 N 0.99), only whole catchment land cover, and land cover in the 60 m and the 300 m riparian buffer zone were used in further analyses, e.g. in exploring the relationship between land use, water quality, and fatty acid composition and concentration.

64

I.G. Boëchat et al. / Science of the Total Environment 482–483 (2014) 62–70

Fig. 1. Hydrographic network, study sites (circles) and urban areas (in gray) in the Rio das Mortes catchment.

2.3. Water quality variables and structural integrity We measured river water physical and chemical variables, such as pH, specific conductance (SC), temperature (T), dissolved oxygen (DO) concentration and saturation (%DO) in situ using a multiparameter probe (556MPS, Yellow Springs Instruments, OH, USA). Water samples for chemical and biochemical analyses were taken in triplicate at each site in the middle of the water column in the river's thalweg using a Van Dorn bottle (Limnotec, Brazil). Concentrations of ammonium-nitrogen (NH4-N), nitrate + nitrite-nitrogen (NO3 + NO2-N), and soluble reactive phosphorus (SRP) were measured by flow injection analysis (FIALab 2500, FIALab, Bellevue, WA, USA) according to standard analytical procedures (APHA, 1999), after filtering samples through pre-combusted fiberglass filters (Whatman GF/F, 0.7 μm nominal pore size). Samples for suspended particulate organic matter (SPOM) and FA determination were filtered on the same type of pre-combusted glass fiber filter. Filters for SPOM were dried at 60 °C for 6 h. After weighting, filters for SPOM analyses were combusted at 520 °C for 12 h and weighted again to determine ash free dry mass (AFDM). SPOM was then calculated by the difference between precombusted and combusted filters (APHA, 1999). Filters for FA analyses were freeze–dried after filtration and stored at − 20 °C until analysis (Boëchat et al., 2005, 2007, 2011), which was performed within a few days after filtration to avoid further oxidation of fatty acids. River habitat structural integrity was analyzed following the German river habitat on-site survey methodology (Kamp et al., 2007). According to this method, each sampling station received a score, representing the structural integrity of the riverbed, riverbanks and a 100 m wide floodplain corridor of the river reach 500 m upstream of each sampling site. This survey method considers 6 main parameters (longitudinal profile, bank structures, etc.) defined by 14 functional units (sinuosity, constructions, etc.). The final survey score value varies from 1 to 7, the lower the score the higher the environmental integrity of the assessed river reach. 2.4. Fatty acid analysis In the laboratory, sampled river water was filtered onto precombusted and pre-weighted 25 mm glass fiber filters (GF-F, Whatman,

USA), freeze dried and stored at − 20 °C until analysis. Fatty acid methyl-esters (FAMEs) were obtained by extracting filtered samples in 15 mL of a 2:1 chloroform–methanol (v:v) solution for 3 h in the dark on ice, and under continuous shaking, after homogenizing the samples with ultrasonic sonication for 5 min in order to break cell walls of bacteria and algae (Boëchat et al., 2011). The method applied for lipid extraction does not require a pH buffer. Samples were centrifuged and the upper layer containing the lipid extract was transferred to 100 mL pear-shaped flasks. The extracts were freeze–dried under nitrogen flux and stored at −20 °C until further analysis. A 50-μL aliquot of a 0.4 mg mL− 1 internal standard solution (tricosanoic acid, 23:0, Supelco, Germany) was added to the samples. The FAs were methylated in a 5 mL sulfuric acid in methanol solution (5%) heated for 4 h at 80 °C. FAMEs were measured in a gas chromatography system (Agilent 6890, Germany) equipped with a mass selective detector (Agilent 5973-N, Germany) and a fused silica capillary column (CP Sil 88 for FAME, 100 m × 250 μm × 39 μm). The carrier gas (helium) was set to a constant flow rate of 0.2 mL min−1. The temperature of the programmed temperature vaporization inlet, operating in splitless mode, was 300 °C (initial temperature 70 °C, 720 °C min−1). The temperature of the detector interface was 280 °C. The following program was employed: 80 °C for 1 min, heating 4 °C min−1 until a temperature of 220 °C was reached and then maintained for 15 min. Fatty acid methyl-esters were identified by their retention times and mass spectra in full scan mode (SCAN), previously calibrated with fatty acid standards (FAME Mix 47885-4, PUFA n°1-47033 and PUFA n°3-47085-4, Supelco, Germany). FAMEs were quantified by selective ion monitoring (SIM) of the two most intensive ions of the molecular ion cluster (Boëchat et al., 2007). The concentrations of additional FAs were estimated by comparing the intensities of similarly FAs of the standard solution. Method detection limit and determination limit were 0.005 mg g− 1 and 0.01 mg g−1, respectively. We present FA data as absolute concentration (μg L−1) of all measured single FAs and FA classes — total fatty acids (ΣFA), saturated fatty acids (SFA), monounsaturated fatty acids (MUFA), polyunsaturated fatty acids (PUFA) as well as relative concentrations of the major single FAs. Fatty acids most common in domestic sewage (C15:0, C16:0, C16:1, C18:0, C18:1; Eganhouse and Kaplan, 1982; Quéméneur and Marty, 1994; Jardé et al., 2005) to total fatty acids (ΣSEW:ΣFA), bacterial

I.G. Boëchat et al. / Science of the Total Environment 482–483 (2014) 62–70

65

Table 1 Land cover (%) of 60 m and 300 m wide riparian buffer zones, and the whole sub-catchment upstream of each sampling station at the Rio das Mortes, as well as habitat structural integrity (SI) score of each site. The SI ranges from 1 to 7, the lower the score the higher the structural integrity. Land cover 60 m riparian buffer

300 m riparian buffer

Sub-catchment

Sampling station

% natural

% agricultural

% urban

% natural

% agricultural

% urban

% natural

% agricultural

% urban

SI-score

1 (headwater) 2 (CAM) 3 (BAR) 4 (PR) 5 (UT) 6 (TIR) 7 (ER) 8 (FLONA) 9 (RIT) 10 (IBIT) 11 (river mouth)

32.8 40.8 48.0 48.9 48.7 48.7 49.8 49.0 48.3 49.0 48.0

65.2 52.6 45.5 45.4 45.6 45.7 45.8 46.3 47.2 47.0 48.3

2.0 6.6 6.4 5.7 5.6 5.6 4.4 4.7 4.5 3.9 3.7

36.0 44.2 48.1 49.0 49.1 49.1 49.3 48.0 48.1 48.9 48.2

62.4 50.7 45.8 45.4 45.5 45.5 46.4 47.1 47.2 47.0 47.9

1.7 5.1 6.1 5.6 5.4 5.4 4.4 4.9 4.7 4.1 3.9

41.6 49.0 52.6 53.5 53.6 53.7 53.7 52.5 52.6 53.4 52.7

56.9 46.3 41.8 41.3 41.4 41.3 42.2 42.9 43.0 42.8 43.6

1.5 4.7 5.6 5.1 5.0 5.0 4.1 4.5 4.4 3.8 3.6

3 4 6 5 7 7 6 6 4 5 5

(iC15:0, aC15:0, C15:0, C17:0, C18:1ω7; Saliot et al., 2001) to total fatty acids (ΣBACT:ΣFA), even numbered to odd numbered fatty acids (even: odd), and ω6:ω3 fatty acid ratios were calculated.

3. Results

2.5. Statistical analyses

Natural areas (41.6 to 53.7% of catchment land cover) and agriculture (41.3 to 56.9%) were the prevalent types of land cover in the subcatchment upstream of each sampling station at the Rio das Mortes, followed by urban land use (1.5 to 5.6%). A very similar pattern was observed for the riparian buffer zones (Table 1). In general, natural and urban land covers had their highest contributions to total subcatchment land cover in the upper to middle course of the river, with the exception of the headwaters, but were lower in the most downstream river reaches, near to the river mouth. An inverse pattern was observed for agricultural land cover, both in the buffer zones and in entire sub-catchment (Table 1). The habitat structural integrity survey attributed the lowest scores (highest habitat integrity) to the two most upstream sites (stations 1 and 2, Table 1). Stations 5 and 6 in the mid-course of the river exhibited the highest scores, followed by stations 3, 7, and 8 (Table 1) and were thus classified as the most structurally impaired stations in our study. Sampling stations in the lower course (9, 10 and 11) had lower scores than mid-course stations (Table 1). None of the investigated sampling stations received score 1 or 2, suggesting that the entire river was subjected to some degree of structural environmental impact. Except for temperature and nutrient concentrations, no significant differences in water quality variables were found between sampling campaigns carried out in the dry and the rainy seasons (Table 2). Water temperature was higher in the rainy season compared to the dry season (Student t-test, P b 0.05). Concentrations of SPOM were higher in the rainy than in the dry season (Student t-test, P b 0.05), whereas SRP, NH4-N and NO3 + NO2-N, as well as DIN concentrations were higher in the dry season than in the rainy season (Student t-test, P b 0.05). Longitudinal variability in water quality features and SPOM concentrations was rather small (Table 2 and Fig. 2C), and tended to be most pronounced in upstream river reaches (sampling stations 1–

Parametric statistical procedures were conducted on log10 transformed data (except for relative data and ratios), whenever necessary to meet the assumptions of normality of data distribution and homoscedasticity. When those assumptions were not met by data transformation, non-parametric tests were used. Normality was tested for all variables by Shapiro–Wilk's test. We used one-way ANOVAs to test for differences in FA and water chemistry variables among sampling stations within each sampling campaign and Student t-tests to test for differences in FA and water chemistry data between the dry and the rainy seasons. Spearman rank correlation was calculated between FA data (FA composition, concentrations and ratios) and (1) land cover, both of subcatchments and riparian buffer zones, (2) water chemistry and (3) the structural integrity survey score on pooled data of all sampling campaigns. The decision to pool data of sampling campaigns was based on the fact that no significant differences were detected for single FAs, FA classes and FA ratios among sampling campaigns (one-way ANOVAs, P N 0.1). Nitrate + nitrite-N and ammonium-N were summed up to dissolved inorganic nitrogen (DIN) prior to statistical analysis. In order to evaluate longitudinal patterns in water quality, river structural integrity and FA variables in the Rio das Mortes catchment, we carried out a principal component analysis (PCA). In order to meet the assumption of PCA of having more cases than variables, some variables had to be removed from the analysis. Those included water quality variables not exhibiting significant correlations with land use, such as temperature and pH, as well as total PUFA concentration, as PUFAs were only found in very small concentrations in the SPOM of the Rio das Mortes. Ammonium-N and nitrate + nitrite-N concentrations were summed up to DIN. All tests were carried out using Statistica for Windows (v. 10.0, Statsoft, USA).

3.1. Land-use characterization, structural integrity and water quality variables

Table 2 River water characteristics (median, min–max values in parentheses) of eleven sampling stations at the Rio das Mortes in the dry and the rainy season 2010–2011.

Dry season Rainy season

NO3 + NO2-N (mg L−1)

NH4-N (mg L−1)

DIN (mg L−1)

SRP (mg L−1)

pH

Temp. (°C)

DO (mg L−1)

SC (μS cm−1)

419⁎ (52–659) 254 (46–440)

217⁎ (34–473) 83 (29–205)

667⁎ (107–1124) 331 (75–567)

12⁎ (1.2–26) 3.9 (1.3–7.1)

7.3 (4.9–10.1) 7.5 (6.6–7.9)

17⁎ (14–20) 21 (19–22)

8.5 (7.6–9.5) 7.4 (6.7–7.9)

48 (17–61) 40 (17–50)

⁎ Significant differences at 95% (Student t-test, P b 0.05).

66

I.G. Boëchat et al. / Science of the Total Environment 482–483 (2014) 62–70 Table 3 Composition and mean absolute concentrations in μg L−1 (±1 SD) of single fatty acids (FA) and FA classes, as well as FA ratios (ΣBACT:ΣFA, ω6:ω3, even:odd FA) along a longitudinal gradient at the Rio das Mortes in 4 seasonal sampling campaigns. Fatty acids

Fig. 2. Longitudinal patterns in the absolute concentration of SPOM-FA classes (A), most abundant single fatty acids (B), and fatty acid ratios and SPOM concentration (C) in four seasonal sampling campaigns (mean + 1 SD) in the Rio das Mortes.

3). Indeed, no significant differences were found for nutrient and SPOM concentrations among sampling stations (one-way ANOVA, P N 0.05). Variability in water quality encountered along the river was related to urbanization. Despite its small contribution to total sub-catchment land cover, urban land use in the whole sub-catchment, as well as in the 60 and 300 m riparian zone, was the only land-use variable significantly correlated to water quality variables (i.e. DIN concentrations, Spearman rank correlation, r = 0.41, P b 0.05) and the structural integrity score (Spearman rank correlation, r = 0.35, P b 0.05). SPOM concentrations were positively correlated to the river structural integrity score (Spearman rank correlation, r = 0.32, P b 0.05) and negatively correlated to DIN concentrations (Spearman rank correlation, r = −0.51, P b 0.05). 3.2. Fatty acid composition and connection to land use The SPOM-FA composition in Rio das Mortes was dominated by saturated fatty acids (SFAs), mainly palmitic acid (C16:0) and stearic

C8:0 C9:0 C10:0 C11:0 C12:0 C13:0 IsoC14:0 C14:0 IsoC15:0 C14:1 AnteisoC15:0 C15:0 IsoC16:0 C15:1 C16:0 IsoC17:0 AnteisoC17:0 C16:1ω9c C16:1ω7c C17:0 CycloC17:0 C18:0 C18:1ω9 t C18:1ω9c C18:1ω7 t C18:1ω7c C18:2ω6 t C18:2ω6c C20:0 C20:1 C18:3ω3 C21:0 C22:0 C20:3ω6 C22:1ω9 t C22:1ω9c C20:3ω3 C23:0 C22:2ω6 C24:0 C20:5ω3 C24:1 C26:0 C28:0 Total FA Sum SFA Sum MUFA Sum PUFA ω6:ω3 ΣSEW:ΣFA ΣBACT:ΣFA Even:odd

Sampling campaign May 2010

Sept. 2010

March 2011

June 2011

0.6 (0.3) 1.7 (0.7) 1.0 (0.2) 0.9 (0.2) 2.0 (0.4) 0.7 (0.1) 2.0 (3.7) 7.5 (1.4) 1.9 (2.7) 5.1 (0.5) 3.9 (6.0) 3.9 (0.5) 0.8 (0.4) 0.5 (0.1) 32.8 (8.8) 0.3 (0.1) 0.2 (0.1) 7.5 (2.9) 2.5 (1.3) 1.7 (0.3) 1.2 (0.2) 19.5 (3.9) 1.5 (0.6) 9.0 (1.9) 0.6 (0.5) 0.2 (0.1) 0.4 (0.1) 3.0 (0.8) 1.5 (0.2) 1.5 (0.8) 0.3 (0.1) 0.3 (0.0) 1.5 (0.2) 0.2 (0.1) 0.7 (0.6) 14.6 (8.0) 0.2 (0.1) 0.5 (0.1) 0.4 (0.1) 2.0 (0.4) 0.6 (0.3) 1.5 (1.9) 1.3 (0.6) – 136.5 (28.8) 88.0 (22.1) 42.6 (11.2) 4.3 (1.9) 7.6 (4.8) 0.5 (0.1) 0.09 (0.04) 6.1 (2.2)

– 1.0 (0.4) 0.9 (0.2) 0.8 (0.1) 2.5 (0.8) 0.7 (0.1) 1.2 (0.5) 10.4 (3.9) 1.3 (0.4) 7.6 (2.3) 3.0 (1.2) 5.0 (1.6) 1.4 (0.4) 0.7 (0.3) 49.6 (22.6) 0.2 (0.1) 0.4 (0.3) 8.8 (4.4) 3.6 (2.3) 2.3 (0.7) 1.6 (0.9) 25.4 (14.7) 1.1 (1.4) 14.5 (6.1) 2.0 (0.9) 0.1 (0.0) 0.6 (0.3) 4.7 (2.4) 2.0 (0.5) 2.2 (0.9) 0.6 (0.4) 0.4 (0.1) 1.9 (0.4) – 1.7 (0.5) 23.2 (7.4) – 0.6 (0.2) 0.6 (0.1) 2.7 (0.8) 0.3 (0.2) 0.6 (0.5) 1.3 (0.5) 1.5 (0.7) 189.3 (53.6) 117.2 (43.5) 65.4 (11.4) 6.6 (2.9) 8.2 (3.2) 0.5 (0.1) 0.08 (0.01) 6.4 (2.3)

– 0.3 (0.1) 0.6 (0.1) 0.6 (0.1) 1.7 (0.8) 0.5 (0.1) 1.0 (0.3) 7.6 (7.2) 2.0 (3.4) 4.7 (1.8) 3.0 (5.0) 3.4 (1.3) 1.0 (0.5) – 40.3 (28.1) 0.1 (0.0) 0.3 (0.3) 4.6 (1.4) 3.0 (2.6) 1.8 (0.8) 1.0 (0.6) 23.3 (13.4) 1.5 (1.7) 15.3 (13.4) 1.2 (0.6) – 0.4 (0.1) 4.3 (1.2) 2.0 (0.6) 1.2 (0.4) 0.6 (0.3) 0.4 (0.1) 2.0 (0.6) 0.2 (0.0) 1.3 (0.3) 18.6 (7.4) – 0.6 (0.2) 0.4 (0.1) 2.8 (0.9) – 0.3 (0.3) 1.1 (0.6) 1.3 (0.6) 156.8 (82.8) 99.5 (60.6) 51.6 (22.3) 5.4 (1.5) 9.1 (2.2) 0.5 (0.1) 0.08 (0.02) 6.9 (1.2)

– 0.5 (0.2) 0.9 (0.2) 0.8 (0.1) 2.3 (0.6) 0.6 (0.1) 1.2 (0.5) 8.1 (2.8) 1.2 (0.4) 6.7 (2.1) 2.4 (0.8) 4.4 (1.4) 1.4 (0.3) – 48.1 (19.5) 0.2 (0.1) 0.3 (0.2) 8.2 (6.4) 4.0 (1.5) 2.1 (0.8) 4.8 (2.4) 27.9 (12.5) 0.7 (0.4) 15.3 (9.3) 1.5 (0.9) – 0.5 (0.2) 5.3 (3.1) 2.0 (0.8) 0.8 (0.3) 0.6 (0.4) 0.4 (0.1) 2.0 (0.7) 0.2 (0.1) 0.6 (0.8) 3.6 (5.1) – 0.6 (0.2) 0.3 (0.1) 2.6 (1.2) – 0.1 (0.1) 1.3 (0.6) 1.4 (0.7) 169.3 (68.6) 120.4 (50.8) 42.5 (19.4) 6.4 (3.8) 10.1 (4.3) 0.6 (0.1) 0.08 (0.01) 8.1 (1.4)

– — not detected.

acid (C18:0), as well as the MUFAs erucic acid (C22:1ω9) and oleic acid (C18:1ω9) (Table 3). Noticeable minor FAs were C14:0 and the MUFAs C14:1, C16:1ω9c, and C18:2ω6 (Table 3). PUFAs were found in only very small concentrations in the SPOM of Rio das Mortes, but ω6 PUFAs were always in higher amounts than ω3 PUFAs (ω6:ω3 ratio varied from 7.6 to 10.1). Fatty acids common in domestic sewage (ΣSEW) were detected in high concentrations at all sampling stations, resulting in ΣSEW:ΣFA ratios between 0.5 and 0.6 (Table 3). Bacterial FAs (ΣBACT) were also detected at all sampling stations, e.g. iso- and ante-isoforms of C15:0, as well as C16:1ω7 and C18:1 (Table 3). The ΣBACT:ΣFA ratio varied between 0.08 and 0.09 in all samplings. The even:odd FA ratio varied between 6.1 and 8.1 (Table 3). Significantly higher absolute concentrations of total FAs and SFAs (Fig. 2A), as well as higher relative concentrations of the single SFAs

I.G. Boëchat et al. / Science of the Total Environment 482–483 (2014) 62–70

67

Table 4 Significant correlation coefficients (P b 0.05, Spearman rank correlation) of urban land use (in 60 and 300 m riparian buffer zones and in the whole catchment) with fatty acid variables, DIN concentrations, and survey structural integrity score.

Total FA SFA PUFA ω6 ω3 C16:0 C18:0 C18:1ω9 ΣSEW:ΣFA DIN IS-score

% urbanization (60 m)

% urbanization (300 m)

% urbanization (sub-catchment)

0.47 0.56 0.49 0.50 0.47 0.56 0.61 0.36 0.49 0.38 n.s.

0.33 0.42 0.40 0.41 0.40 0.42 0.46 n.s. 0.35 0.40 0.37

0.32 0.42 0.39 0.40 0.39 0.41 0.45 n.s. 0.35 0.40 0.35

n.s. — correlation not significant at 95% confidence level.

C16:0 and C18:0 (Fig. 2B), and a higher ΣSEW:ΣFA ratio (Fig. 2C) were detected at the second sampling station than at all other stations at the Rio das Mortes in all sampling campaigns (one-way ANOVA followed by Tukey HSD-test, P b 0.05). The MUFA C22:1ω9, the third most abundant FA in three out of four sampling campaigns, had its highest relative concentration at stations 1, 3 and 7 (Fig. 2B). No significant effect of seasonality was detected neither for FA composition, nor for concentrations of single FAs, FA classes or ratios (one-way ANOVA followed by Tukey HSD-test, P N 0.05). Spearman rank correlations showed that the contribution of urban land cover to total sub-catchment land cover was significantly correlated to total FA concentration (r = 0.32, P b 0.05; Table 4), as well as the concentration of SFA (r = 0.42, P b 0.05) and PUFA (r = 0.39, P b 0.05; Table 4). Sub-catchment urban land use was also significantly correlated to the absolute concentration of the most abundant FA, C16:0 (r = 0.41, P b 0.05), C18:0 (r = 0.45, P b 0.05), and the ΣSEW:ΣFA ratio (r = 0.35, P b 0.05) (Table 4). The percentage of urban land use in the 60 and 300 m riparian buffer zone was, in general, correlated to the same FA concentrations and ratios as for the urban land use in the whole sub-catchment. However, correlation coefficients were highest for urban land use in the 60 m riparian buffer zone (r = 0.47, 0.56, 0.49, 0.56, 0.61, and 0.49 for total FA, SFA, PUFA, C16:0, C18:0, and ΣSEW:ΣFA, respectively, P b 0.05; Table 4). Principal component analysis (PCA) suggested that the headwater and the second station differed from downstream river stations when all investigated variables were considered together. The first axis of the PCA explained 37% of the variability among sampling stations, and the variables causing this separation were mostly FA variables, which separated the second station from all other stations, mainly because of high concentrations of C18:0 and C16:0, total FA, SFA and C18:1ω9c (Table 5, Fig. 3). The second axis of the PCA explained around 19% of the variability among sampling stations, basically due to the percentage of urban land use, agriculture, DIN concentrations and the habitat SIscore (Table 5, Fig. 3). On this axis, the headwater station was separated

Table 5 Variables and their scores (in parentheses), as well as eigenvalues and explained variation extracted by the first two axes of the PCA for sampling sites, based on riparian land cover, river water characteristics, site structural integrity, and fatty acid variables.

Variable contribution

Eigenvalue Extracted variation (%) Cumulative extracted variation (%)

1st axis

2nd axis

ΣFA (0.97) SFA (0.98) C16:0 (0.98) C18:0 (0.96) C18:1ω9c (0.93) ΣSEW:ΣFA (0.67) 5.87 36.7 36.7

% agric. (0.84) % urban (−0.78) SI-score (−0.75) DIN (−0.64) C22:1ω9c (0.41) 3.03 18.9 55.7

Fig. 3. PCA of sampling sites, based on riparian land cover, river water characteristics, structural integrity, and fatty acid variables in the Rio das Mortes. Single fatty acids and fatty acid classes and ratios as in Table 3; SPOM: suspended organic matter; DIN: dissolved inorganic nitrogen; SRP: soluble reactive phosphorus; DO: dissolved oxygen concentration; SI score: structural integrity index score.

from all other stations, mostly based on the high percentage of agriculture, low percentage of urban land cover, the lowest habitat structural integrity score, the lowest DIN concentrations and high concentrations of C22:1ω9. Based on the investigated water quality, structural integrity and FA variables, all other sampling stations were similar according to the PCA. 4. Discussion Fatty acid composition and concentration of SPOM varied along the Rio das Mortes, with notable peaks of some FA markers at some stations, suggesting that land use may indeed affect the quality of SPOM in this larger tropical river. However, despite its low relative contribution to total catchment area, urban land use was the only statistically significant predictor of both FA composition and concentrations. This relation between FA variables and urban land use was mainly due to high concentrations of the saturated palmitic acid (C16:0) and stearic acid (C18:0) at stations located directly downstream of larger urban centers, such as the cities of Barbacena (station 2) and Barroso (station 3). Palmitic and stearic acids have previously been described as the main FAs in both treated and raw sewage (Quéméneur and Marty, 1994; Réveillé et al., 2003) and our study expands this knowledge by demonstrating their persistence in higher concentrations downstream of the more urbanized areas of the investigated tropical river system, and thus their suitability as biochemical indicators of sewage discharge in situ. Additionally, considering the generally high ΣSEW:ΣFA ratio at all

68

I.G. Boëchat et al. / Science of the Total Environment 482–483 (2014) 62–70

stations, the highest ΣSEW:ΣFA ratio of all stations occurred at the most urbanized station (station 2), and the lowest ΣSEW:ΣFA ratio of all stations at the least urbanized station (station 1). Accordingly, sewage appears to be an important source of SPOM-FAs in the investigated tropical catchment. Interestingly, 440,000 inhabitants of the total population of the investigated catchment (around 520,000 inhabitants) were located in urban areas (IBGE, 2007). The mesoregion of Barbacena (part of station 2's sub-catchment) had the highest population density in the catchment, with around 220,000 inhabitants (IBGE, 2007). Downstream of Barbacena, population density diminished (IBGE, 2007). Together with the dilution effect of less urbanized tributaries, this may have led to lower concentrations of all FAs and FA ratios – but especially of sewage markers – in the middle and lower course of the Rio das Mortes. This distribution pattern of the human population may also have been responsible for the found PCA pattern, with all stations, except for the first and second ones, being indistinguishable on the PCA axes. Urbanization also exerted major impacts on other indicators of water quality, structural integrity and ecosystem functioning of the Rio das Mortes and its tributary streams. For instance, urban land use was correlated to DIN concentrations and the habitat integrity score in the present study, and to instantaneous secondary production of macroinvertebrates in the main channel of the Rio das Mortes (unpublished data). In headwater streams of the Rio das Mortes catchment, small fractions of urban land use (b8.3% of total sub-catchment area) increased stream ecosystem respiration, but diminished leaf decomposition rates (Silva-Junior et al., 2014). Thus, it is likely that, despite the larger contributions of pasture and agriculture in continental Brazil, the condensed character of urbanization is more deleterious to water quality and ecosystem functioning of small and large rivers than other land uses; a fact already well described for temperate catchments (Paul and Meyer, 2001). Raw domestic sewage can contain large amounts of SPOM, of which lipids can contribute up to 20–25%, with concentrations ranging from 40 to 100 mg L− 1 (Loehr and De Navarra, 1969), mainly derived from kitchen waters and human feces. Vegetable oils are rich in C16:0, C18:1ω9 (oleic acid), C18:2ω6 (linoleic acid) whereas animal fat contain large amounts of C16:0, C18:0, C18:1ω9 and cholesterol (Wolfe, 1968; Segura, 1988). Human feces may contain 4–23% of lipids and their fatty acid composition is dominated by C18:1ω9, C18:0 and C16:0 (William et al., 1960). Interestingly, urban land use in the riparian zone was highly correlated to the concentrations of C16:0, C18:0 and C18:1ω9 as well. The presence of a high even:odd ratio with a dominance of chains containing 16 and 18 carbons is typical of higher plant or animal contribution to the fatty acid composition (Rao et al., 1990), and corroborates the presence of vegetable oils and human feces in SPOM. Interestingly, the second axis of our PCA suggested a connection between the even:odd ratio and the percentage of urbanization in the Rio das Mortes catchment. Thus, even though SPOM total concentration was not significantly correlated to urban land use, the FA profile was clearly correlated to urbanization and proved to be a powerful tool for assessing land-use effects, that can be used whenever other more commonly applied methods fail. The most common FA found in our study was the rather ubiquitous saturated palmitic acid (C16:0), the most abundant FA found in animals, plants and microorganisms (Gunstone et al., 2007). Palmitic acid is mainly used to produce soaps, cosmetics, and release agents, in the form of sodium palmitate, which is obtained by saponification of palm oil. The second most abundant FA was C18:0, a SFA commonly found in vegetable oil. Both saturated C16:0 and C18:0 FAs have been related to sewage in previous studies (Quéméneur and Marty, 1994; Réveillé et al., 2003; Jardé et al., 2005), and are commonly associated with human fecal discharge and kitchen oils, indicating that considerably increased concentrations of those FAs in natural waters may be good indicators of urban land use in the surrounding catchment, even though they are ubiquitous. Interestingly, the first axis of the PCA suggested

that the high concentrations of both C16:0 and C18:0 found in the SPOM of the second sampling station, were the variables exhibiting the highest correlations with urban land cover. It is important to note that both palmitic and stearic acids are the initial precursors in the synthesis of long chain fatty acids such as linoleic and linolenic acids, which are in turn precursors in the synthesis of essential PUFAs by some organisms (Berg et al., 2002). Although not essential, a high availability of palmitic and stearic acids in river SPOM may represent a subsidy for FA synthesis to mixotrophic protists and suspension-feeding consumers. Moreover, lipids provide more energy than carbohydrates and proteins. Therefore, higher lipid concentration may represent an additional energy source, even though the quality of the available FAs is not comparable to that of e.g. essential PUFAs. Erucic acid, the third most abundant FA found, is the cis isomer of brassidic acid, generally derived from rapeseed, wallflower seed, or mustard seed. End industrial applications of erucic acid include the synthesis of lubricants, heat transfer fluids, surfactants, slip agents, emollients, cosmetics and coatings. It is also used in the synthesis of polyesters, plastics and nylons. Erucic acid has also been reported in aquatic ecosystems as a component of leaf wax coating resulting from terrestrial sources (Rieley et al., 1991). Generally, even-numbered saturated FAs with chain lengths higher than C20 are related to terrestrial sources of organic matter in the water column and sediments of aquatic systems (Canuel et al., 1997). The high concentration of erucic acid in SPOM of the headwater station may suggest a substantial contribution of terrestrial higher plants and terrestrial detritus to the food web of upstream segments of the river. Moreover, the second axis of the PCA suggested a connection between high concentrations of erucic acid and agricultural land cover. The fourth most abundant FA found was C18:1ω9c (oleic acid), which is a major component of human feces (William et al., 1960). Concentrations of oleic acid did not differ among stations along the investigated tropical river, pointing to sewage impacts from the headwaters to the river mouth. In three out of four sampling campaigns, its absolute concentration was highest at the urbanized station 2, coinciding with peaks of both saturated C16:0 and C18:0 FAs. The low concentrations of PUFAs in river SPOM either indicated a small contribution of algae and other autochthonous autotrophic organisms, a fact also suggested by the low chlorophyll-a concentrations in this river (b 2 μg L−1 during the present study, unpublished data). Additionally, these FAs are highly labile and subject to rapid losses by bacterial degradation, diagenesis and/or via grazing (Cranwell, 1981; Hu et al., 2006). The ω3: ω6 ratio has been suggested as a marker for the contribution of autochthonous versus allochthonous matter in aquatic systems. A higher ω3: ω6 ratio is normally associated to a high contribution of sestonic algae (Ravet et al., 2010) whereas a low ω3: ω6 ratio is frequently reported as an indicator of allochthonous input from the upstream terrestrial system (Torres-Ruiz et al., 2007). Thus, the high ω6: ω3 ratio found in our study suggests that most PUFAs in the SPOM of Rio das Mortes are probably originated from cooking oils and other plant byproducts, as domestic sewage seems to be the main allochthonous source of fatty acids in Rio das Mortes SPOM. The high concentrations of FAs common in sewage at all sampling stations, as e.g. indicated by the high ΣSEW:ΣFA ratio, may also point to a high bacterial activity, due to bacteria discharged concomitantly with raw sewage, in the SPOM of the Rio das Mortes. This hypothesis deserves further investigation, e.g. planktonic respiration measurements, in order not only to prove potentially high processing rates of transported detritus along tropical river systems, but also to allow for more realistic estimates of SPOM, and thus organic carbon exports to downstream reservoirs and the ocean in tropical areas. To our knowledge, this is the first study to investigate the effects of catchment land cover on the SPOM-FA composition in a tropical river. Land use impacts were substantial in the investigated tropical catchment, as indicated by the high degree of structural and water quality degradation already in the river's headwaters. Moreover, spatial

I.G. Boëchat et al. / Science of the Total Environment 482–483 (2014) 62–70

patterns in fatty acid composition in SPOM were driven by urbanization, whereas nutrient concentrations and other water quality variables did not exhibit any spatial patterns at all, suggesting that natural longitudinal patterns in river organic matter and solute dynamics, e.g. those suggested by the River Continuum Concept (Vannote et al., 1980), may have been overridden by the pronounced land use effects in the investigated tropical river catchment. Urbanization appeared to be the main driver of FA composition and concentration, and could be an important stressor of riverine food web interactions. Urbanization related impacts were mainly due to the absence of waste water treatment, a general problem in developing countries that contributes to the eutrophication of lentic systems and running waters. Interestingly, FA variables were more strongly associated to urban land cover than water quality and habitat integrity variables, suggesting that SPOM-FAs may be very efficient indicators of land-use change, especially in urbanizing tropical catchments without waster water treatment. Spatial changes in FA composition and concentration as found in this study can affect food web interactions and population growth, as FA composition is a recognized determinant of resource nutritional quality for consumers in aquatic food webs (Boëchat and Adrian, 2006; Ravet et al., 2010). By affecting the biochemical quality of available resources, even low levels of catchment urbanization, without efficient waste water treatment, may have important impacts on riverine energy fluxes and the maintenance of food web structure and interactions. 5. Conclusions Urbanization was the main land-use category affecting both FA composition and concentrations along a 4th order tropical river. Higher concentrations of palmitic and stearic acids, as well as a high sewage to total FA ratios were important markers of sewage discharge, a major impact on the water quality of the studied tropical river. High total fatty acid concentrations in the SPOM of urbanized river sections may affect the quantity and quality of available organic matter and, therefore, lead to changes in functional ecosystem characteristics, such as bacterial and suspension-feeder production. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.scitotenv.2014.02.111. Conflict of interest The authors declare no conflict of interest. Acknowledgments We thank R.C.S. Silva, A.P.C. Carvalho, A.T.B. dos Santos, and E. M. Soares for their help with sampling and chemical analyses. A.B.M. Paiva is acknowledged for providing data on habitat structural integrity and S. Hille for assistance with GIS analyses. Two anonymous reviewers provided helpful comments on an earlier version of the manuscript. This study was supported by the Fundação de Amparo à Pesquisa no Estado de Minas Gerais (FAPEMIG; APQ-01619-09). D. Graeber was supported by the Danish Council for Independent Research (ECOGLOBE project, Natural Sciences, 09-067335). References Ahlgren G, Gustafsson IB, Boberg M. Fatty acid content and chemical composition of freshwater microalgae. J Phycol 1992;28:37–50. Allan JD. Landscapes and riverscapes: the influence of land use on stream ecosystems. Ann Rev Ecol Syst 2004;35:257–84. Alvarez-Cobelas M, Angeler DG, Sánchez-Carrillo S, Almendros G. A worldwide view of organic carbon export from catchments. Biogeochemistry 2012;107:275–93. APHA (American Public Health Association). Standard methods for the examination of water and wastewater. 20th ed. Washington, DC: American Public Health Association, American Water Works Association, Water Environment Federation; 1999. Arts MT, Brett MT, Kainz MJ. Lipids in aquatic ecosystems. New York: Springer; 2009.

69

Battin T, Kaplan L, Findlay S, Hopkinson C, Marti Packman A, Newbold J, et al. Biophysical controls on organic carbon fluxes in fluvial networks. Nat Geosci 2008;1:95–100. Berg J, Tymoczko J, Stryer L. Biochemistry. 5th ed. New York: WH Freeman & Co Ltd; 2002. Boëchat IG, Adrian R. Biochemical composition of algivorous freshwater ciliates: you are not what you eat. FEMS Microbiol Ecol 2005;53:393–400. Boëchat IG, Adrian R. Evidence for biochemical limitation of population growth and reproduction of the rotifer Keratella quadrata fed with freshwater protists. J Plankton Res 2006;28:1027–38. Boëchat IG, Schuran S, Adrian R. Supplementation of the protist Chilomonas paramecium with a highly unsaturated fatty acid enhances its nutritional quality for the rotifer Keratella quadrata. J Plankton Res 2005;27:663–70. Boëchat IG, Weithoff G, Krüger A, Gücker B, Adrian R. A biochemical explanation for the success of mixotrophy in the flagellate Ochromonas sp. Limnol Oceanogr 2007;52: 1624–32. Boëchat IG, Krüger A, Giani A, Figueredo CC, Gücker B. Agricultural land use affects the nutritional quality of stream microbial communities. FEMS Microbiol Ecol 2011;77: 568–76. Canuel EA, Freeman KH, Wakeham SG. Isotopic compositions of lipid biomarker compounds in estuarine plants and surface sediments. Limnol Oceanogr 1997;42:1570–83. Chadwick MA, Dobberfuhl DR, Benke AC, Huryn AD, Suberkropp K, Thiele JE. Urbanization affects stream ecosystem function by altering hydrology, chemistry, and biotic richness. Ecol Appl 2006;16:1796–807. Cole JJ, Prairie YT, Caraco NF, McDowell WH, Tranvik LJ, Striegl RG, et al. Plumbing the global carbon cycle: integrating inland waters into the terrestrial carbon budget. Ecosystems 2007;10:171–84. Cranwell PA. Diagenesis of free and bound lipids in terrestrial detritus deposited in a lacustrine sediment. Org Geochem 1981;3:79–89. Dodds WK. Eutrophication and trophic state in rivers and streams. Limnol Oceanogr 2006;51:671–80. Eganhouse RP, Kaplan IR. Extractable organic matter in municipal wastewaters. 2. Hydrocarbons: molecular characterization. Environ Sci Technol 1982;16:541–52. Gücker B, Brauns M, Pusch MT. Effects of wastewater treatment plant discharge on ecosystem structure and function of lowland streams. J N Am Benthol Soc 2006;25: 313–29. Gücker B, Boëchat IG, Giani A. Stream morphology controls ammonium retention in tropical headwaters. Freshw Biol 2009;54:2069–85. Guillemette F, Del Giorgio PA. Reconstructing the various facets of dissolved organic carbon bioavailability in freshwater ecosystems. Limnol Oceanogr 2011;56:734–48. Gunstone FD, Harwood JL, Dijkstra AJ. The lipid handbook. 3rd ed. Boca Raton,Florida: CRC Press; 2007. Hu J, Zhang H, Peng P. Fatty acid composition of surface sediments in the subtropical Pearl River estuary and adjacent shelf, Southern China. Estuar Coast Shelf Sci 2006;66: 346–56. IBGE — Instituto Brasileiro de Geografia e Estatística. Contagem da População. Ministério do Planejamento, Orçamento e Gestão, Governo Brasileiro, Brasília; 2007. Jardé E, Mansuy L, Faure P. Organic markers in the lipidic fraction of sewage sludges. Water Res 2005;39:1215–32. Kamp U, Binder W, Hölzl K. River habitat monitoring and assessment in Germany. Environ Monit Assess 2007;127:209–26. Kaneda T. Iso-and anteiso-fatty acids in bacteria: biosynthesis, function, and taxonomic significance. Microbiol Rev 1991;55:288–302. Lands WEM. Human life: caught in the food web. In: Arts MT, editor. Lipids in aquatic ecosystems. Berlin: Springer Verlag; 2009. Loehr RC, De Navarra Jr CT. Grease removal at a municipal treatment facility. J Water Pollut Control Fed 1969;41:R142-R154. Millennium Ecosystem Assessment. Millennium Ecosystem Assessment Synthesis Report. Washington DC: Island Press; 2005. Müller-Navarra DC, Brett M, Park S, Chandra S, Ballantyne AP, Zorita E, et al. Unsaturated fatty acid content in seston and tropho-dynamic coupling in lakes. Nature 2004;427: 69–72. Paul MJ, Meyer JL. Streams in the urban landscape. Ann Rev Ecol Syst 2001;32:333–65. Quantum GIS Development Team. Quantum GIS Geographic Information System. Open Source Geospatial Foundation Project. http://qgis.osgeo.org, 2013. Quéméneur M, Marty Y. Fatty acids and sterols in domestic wastewaters. Water Res 1994; 28:1217–26. Rao KS, Dominic R, Singh K, Kaluwin C, Rivett DE, Jones GP. Lipid fatty acid, amino acid and mineral compositions of five edible plant leaves. J Agric Food Chem 1990;38:2137–9. Ravet JL, Brett MT, Arhonditsis GB. The effects of seston lipids on zooplankton fatty acid composition in Lake Washington, Washington, USA. Ecology 2010;91:180–90. Raymond PA. Global carbon dioxide emissions from inland waters. Nature 2013;503: 355–9. Réveillée V, Mansuy L, Jardé É, Garnier-Sillam É. Characterisation of sewage sludgederived organic matter: lipids and humic acids. Org Geochem 2003;34:615–27. Rieley G, Collier RJ, Jones DM, Eglinton G. The biogeochemistry of Ellesmere Lake, U.K.-I: source correlation of leaf wax inputs to the sedimentary lipid record. Org Geochem 1991;17:901–12. Rosa RS, Aguiar ACF, Boëchat IG, Gücker B. Impacts of fish farm pollution on ecosystem structure and function of tropical headwater streams. Environ Pollut 2013;174:204–13. Saliot A, Mejanelle L, Scribe P, Fillaux J, Pepe C, Jabaud A, Dagaut J. Particulate organic carbon, sterols, fatty acids and pigments in the Amazon River system. Biogeochemistry. Environ Pollut 2001;54:79–103. Segura R. Preparation of fatty acid methyl esters by direct transesterification of lipids with aluminium chloride methanol. J Chromatogr 1988;441:99–118. Silva-Junior EF, Boëchat IG, Moulton TP, Gücker B. Leaf decomposition and ecosystem metabolism as functional indicators of land use impacts on tropical, agricultural streams. Ecol Indic 2014;36:195–204.

70

I.G. Boëchat et al. / Science of the Total Environment 482–483 (2014) 62–70

Smith VH, Joye SB, Howarth RW. Eutrophication of freshwater and marine ecosystems. Limnol Oceanogr 2006;51:351–5. Sweeney BW, Bott TL, Jackson JK, Kaplan LA, Newbold JD, Standley LJ, et al. Riparian deforestation, stream narrowing, and loss of stream ecosystem services. Proc Natl Acad Sci U S A 2004;101:14132–7. Torres-Ruiz M, Wehr JD, Perrone AA. Trophic relations in a stream food web: importance of fatty acids for macroinvertebrate consumers. J N Am Benthol Soc 2007;26:509–22. Tranvik LJ, Downing JA, Cotner JB, Loiselle SA, Striegl RG, Ballatore TJ, et al. Lakes and reservoirs as regulators of carbon cycling and climate. Limnol Oceanogr 2009;54: 2298–314. Vannote RL, Minshall GW, Cummins KW, Sedell JR, Cushing CE, et al. The river continuum concept. Canadian Journal of Fisheries and Aquatic Sciences 1980;37:130–7.

Walsh CJ, Roy AH, Feminella JW, Cottingham PD, Groffman PM, Morgan II RP. The urban stream syndrome: current knowledge and the search for a cure. J N Am Benthol Soc 2005;24:706–23. William JA, Sharma A, Morris JL, Holman RT. Fatty acids composition of feces and fecaliths. Proc Soc Exp Biol Med 1960;105:192–5. Wolfe IP. Manuel d'Analyse des Corps Gras. Paris: Azoulay; 1968:552. Young RG, Huryn AD. Effects of land use on stream metabolism and organic matter turnover. Ecol Appl 1999;9:1359–76. Young RG, Matthaei CD, Townsend CR. Organic matter breakdown and ecosystem metabolism: functional indicators for assessing river ecosystem health. J N Am Benthol Soc 2008;27:605–25.

Land-use impacts on fatty acid profiles of suspended particulate organic matter along a larger tropical river.

Land-use change, such as agricultural expansion and urbanization, can affect riverine biological diversity and ecosystem functioning. Identifying the ...
678KB Sizes 0 Downloads 3 Views