LETTER TO THE EDITOR Factors that control the spatial and temporal distributions of phosphorus, nitrogen, and carbon in the sediments of a tropical reservoir Sheila Cardoso-Silva1 & Paulo Alves de Lima Ferreira2 & Rubens César Lopes Figueira2 & Daniel Clemente Vieira Rêgo da Silva3 & Viviane Moschini-Carlos1 & Marcelo L. M. Pompêo3 Received: 5 March 2018 /Accepted: 6 August 2018 /Published online: 29 August 2018 # Springer-Verlag GmbH Germany, part of Springer Nature 2018 Abstract The impacts of anthropic activities have had profound effects on the nitrogen (N) and phosphorus (P) cycles in many aquatic ecosystems. We investigated the spatial and temporal distributions of carbon (C), N, and P in the sediments of a tropical Paiva Castro Reservoir (São Paulo, Brazil), as well as their release and retention in the system. In 2010, surface sediments were collected at nine sites in the reservoir, and a core was obtained in the limnetic zone, in 2010. The core was dated using the 210Pb technique. The organic C content was estimated from organic matter concentration, which was measured by the loss-on- ignition method, and the concentrations of P and N were determined by spectrophotometry. Marked spatial heterogeneity in the Paiva Castro sediments associated with both natural variations in the water body and variations induced by human impacts was observed. Heterogeneity was evidenced by a decrease in the allochthonous contribution of organic matter (C/N) in the upstream- downstream direction and increases of N and P, mainly associated with water flows in the different compartments of the reservoir. In the core, C and N concentrations display significant positive correlations with increases in population and agricultural activities in the drainage basin through time. The C/P molar ratios in surface sediments are indicative of human impacts in the region, as C:P ratios in the sediment are low (7.8:1) compared to the Redfield ratio (C:P = 108:1). Predominance of oxic conditions at the sediment surface and particles sizes < 63 μm provided favorable conditions for P retention in the sediments, which helps prevent eutrophication. Approaches used in this research should be extended to other locations, especially in mesotrophic and oligotro- phic reservoirs, to provide information on historical impacts in such aquatic ecosystems. Keywords Anthropic impacts . Eutrophication . Nitrogen . Paleolimnology . Phosphorus . Reservoir . Sediment core Introduction Nitrogen (N) and phosphorus (P) are important elements in aquatic ecosystems, supporting the growth of algae and aquat- ic macrophytes (Hayakawa et al. 2015). Changes in the bio- geochemical cycles of these elements, caused mainly by hu- man activities, have enhanced nutrient delivery to water bod- ies and led to increases in productivity. Such cultural eutro- phication is now a global environmental problem (Wetzel 2001; Smith and Schindler 2009). This process is responsible for many ecological changes and socioeconomic costs, such as loss of biodiversity (Hautier et al. 2009), development of potentially toxic algae blooms (Moschini-Carlos et al. 2009; Cunha et al. 2017), and increased costs of water treatment (Smith and Schindler 2009). Environmental Science and Pollution Research (2018) 25:31776–31789 https://doi.org/10.1007/s11356-018-2923-0 Responsible editor: Philippe Garrigues Electronic supplementary material The online version of this article (https://doi.org/10.1007/s11356-018-2923-0) contains supplementary material, which is available to authorized users. * Sheila Cardoso-Silva she.cardosos@gmail.com 1 Environmental Sciences Program, São Paulo State University – UNESP, Sorocaba campus, Avenida Três de Março 511, Alto da Boa Vista, Sorocaba, SP 18087-180, Brazil 2 Chemistry Department, Institute of Oceanography, University of São Paulo, Praça doOceanográico, 191, São Paulo, SP 05508-120, Brazil 3 Ecology Department, Institute of Biosciences, University of São Paulo, R. do Matão, 14 - Butantã, São Paulo, SP 05508-090, Brazil http://crossmark.crossref.org/dialog/?doi=10.1007/s11356-018-2923-0&domain=pdf https://doi.org/10.1007/s11356-018-2923-0 mailto:she.cardosos@gmail.com Although there is a large body of research on this issue, uncertainties remain (Smith and Schindler 2009) and further studies that can contribute to a better understanding of the dynamics of the N, P, and C biogeochemistry in aquatic envi- ronments and the process that control trophic states of water bodies. Such studies should include sediment analysis, since bottom sediments act as a nutrient sink, but also as a potential source of nitrogen and phosphorus for the water column, un- der certain conditions (Wetzel 2001; Liu et al. 2009). For example, anoxic conditions in the hypolimnion can intensify denitrification and promote the release of nitrogen from sedi- ments (Hou et al. 2014), as well as cause the release of phos- phorus (Fonseca et al. 2011), thereby contributing to eutrophication. Phosphorus is the principal limiting nutrient for the growth of algae in most water bodies (Oluyedun et al. 1991). Dissolved oxygen concentration at the sediment surface is important for retention of P in sediments, but other factors like temperature, redox conditions, pH, and the concentrations of other inorganic species such as sulfide and iron in the sedi- ment and the water column can influence P release (Ribeiro et al. 2008). An increase in temperature generally reduces phos- phorus adsorption by mineral particles in the sediments (Perkins and Underwood 2001), whereas low pH values pre- vent the release of phosphorus to the water column (Wetzel 2001). Under oxic conditions, dissolved phosphate is com- bined with iron and aluminum oxides in the sediment (Fonseca et al. 2011; Dittrich et al. 2013; Hayakawa et al. 2015). In contrast, under anoxic and reducing conditions, the phosphate is then released to the water column (Perkins and Underwood 2001; Wetzel 2001; Esteves 2011; Dadi et al. 2017). The presence of sulfate can increase the effects of an- aerobic conditions, a consequence of the double reaction of ferric iron with sulfate and sulfide to form ferrous iron and iron sulfide (Perkins and Underwood 2001). Under such an- aerobic conditions, increases in pH associated with sulfate reduction of ferric iron and polyphosphate breakdown lead to the release of phosphorus from the sediment (Perkins and Underwood 2001). Despite the potential of sediment analysis, to lead to a better understanding of biogeochemical cycles and trophic state conditions, caution is required in interpreting the data. As mentioned above, under anoxic conditions, phosphorus is released to the water column. Under eutrophic conditions, phosphorus release from the sediments will often be intense even if external loading is low, because of the intense internal loading, as observed in Paranoá Reservoir, Brazil, by Angelini et al. (2008). In such situation, phosphorus in sediments will not necessarily reflect the precisely trophic state in the water column. Therefore, it is important to concomitantly analyze other macronutrients such as nitrogen and carbon. For exam- ple, important information about the trophic state of the envi- ronment can be obtained frommeasures of C/N and N/Pmolar ratios in the sediments. And use of the C/N molar ratio as an indicator of organic matter sources is a common approach in geochemical studies (Ruttenberg and Gofii 1997) and can dis- tinguish between autochthonous or allochthonous sources of organic matter (Liu et al. 2010). Shifts in the N/P ratios, which are indicative of the limiting nutrient (Koerselman and Meuleman 1996), can affect the phytoplankton community composition, especially in relation to the cyanobacteria (Eillers et al. 2004). Caution in data interpretation is prudent when using sedi- ment profiles to infer the trophic state history of a water body. In addition to elements such as nitrogen and carbon, many studies have also employed biological indicators such as dia- toms or photosynthetic pigments (Boyle 2001; Smol 2008). Such bioindicators often provide a better understanding of past trophic state, especially in eutrophic environments. The use of biological variables as proxies for trophic state condi- tions requires accurate identification of taxonomic groups present, or the use of sophisticated techniques such as high- performance liquid chromatography (HPLC) to identify pig- ments. In comparison, geochemical analyses cost less and can be applied if the environment is not eutrophic (Boyle 2001). Chemistry of surface sediment samples and sediment cores can provide important information about both current and past environmental change, respectively (Boyle 2001; Rippey and Anderson 2008; Zan et al. 2012; Hou et al. 2014). We measured carbon, nitrogen, and phosphorus concentra- tions in surface sediments and a core from tropical Paiva Castro Reservoir, a system that fluctuates between mesotrophy (CETESB 2002–2009) and oligotrophy (CETESB 2002–2009; Pires et al. 2015). The spatial and tem- poral distributions of these elements were investigated, along with analysis of sediment release and retention of nutrients in this system. We expected that analyses of C, N, and P, along with analysis of variables responsible for nutrient retention and release, would shed light on natural and anthropogenic changes in the reservoir over the recent past. We also expected that the data would enable us to identify areas most susceptible to nutrient contamination and provide insights into the recent effects of human activities on the reservoir. Material and methods Study area The Cantareira system is the largest source of drinking water for the metropolitan region of São Paulo. It is formed by five reservoirs that are connected by 48 km of artificial under- ground tunnels, canals, and pumps (Whately and Cunha 2007; SABESP 2018). This artificial cascade system was de- veloped in two main stages. The first involved construction of the Paiva Castro, Atibainha, and Cachoeira Reservoirs, which Environ Sci Pollut Res (2018) 25:31776–31789 31777 began operation in 1974, with a flow of 11 m3 s−1. The second stage was completed in 1981, with inclusion of the Jaguari and Jacareí Reservoirs, when the flow increased to 33 m3 s−1 (Whately and Cunha 2007). The Paiva Castro Reservoir, the last in this artificial cascade system, is located in the Alto Tietê hydrographic basin and was formed by damming of the Juquery River. Effluents from the Mairiporã city sewage treatment station are released into the Paiva Castro upstream area. The reservoir lies at an alti- tude of 745 m and has a drainage area of 314 km2 and a mean flow of 4.4 m3 s−1 (SABESP 2018). Rapid population growth (Whately and Cunha 2007) has led to increased nutrient inputs into this reservoir (Giatti 2000; Silva 2002). Regular applica- tions have been made of algicides such as copper sulfate and hydrogen peroxide which have been used to control algae blooms resulting from these excessive nutrient inputs. Sampling Two samplings were carried out at the Paiva Castro Reservoir. In March 2010, a gravity corer (Ambühl and Bührer sam- pler—Ambühl and Bührer 1975) was deployed twice in the deepest part of the reservoir near the dam (C) (Fig. 1). Each 32-cm core was sliced at 2-cm intervals down to 26-cm depth, and at 1-cm intervals thereafter. Samples were stored in sealed plastic bags and kept in thermally insulated bags, until analy- sis in the laboratory. One core was used for determination of iron, manganese, phosphorus, and nitrogen, and for 210Pb dat- ing. The other core was used for determination of organic matter content and particle size. The core slices were num- bered in ascending order. The second sampling was completed on 13 July 2010 and involved collection of surface sediment samples for measure- ments of biogeochemical variables. Nine sites were sampled, with seven along the main axis of the reservoir and two (sites 5 and 6) in a lateral arm from where water is extracted for the public water supply (Fig. 1). At each sampling site, reservoir depth was first determined, followed by two deployments of an Ambühl and Bührer sam- pler (Ambühl and Bührer 1975). The first collected sample was used for in situ determination of the redox potential (EH) (electrode: Digimed Model DMP-CP1; base: Gehaka Model PG1400), pH, and temperature (T) in the uppermost centimeter of the sediment, and dissolved oxygen (DO)— 5 cm above sediment line (YSI 556 MPS multi-parameter probe). The topmost 3 cm of the sediment column was then used for N and P analyses. The second sample collected, the first 3 cm of the sediment column was removed for grain size and organic matter analyses. The sediment samples were transferred to sealed plastic bags, which were stored in ther- mally insulated bags prior to analysis in the laboratory. In all sampling and stations, depth was determined throwing a mea- suring tape with a weight at its end. Laboratory analyses Sedimentation rate and 210Pb dating A portion of each sample from the cores was dried at 45 °C to constant weight to assess percent dry weight. Dry material was then ground with a glass mortar until homogenized and was then used for the elemental analysis. Samples for mea- surements of 210Pb and 137Cs activities were sealed and stored for at least 20 days to enable 214Pb and 214Bi to reach secular equilibrium with in situ 226Ra. Samples were counted for 50,000 s with a high-efficiency, low-background detector (EG&G Ortec Model GMX 25190P) and the spectra were analyzed using Maestro 2 software. The CIC (Constant Fig. 1 Paiva Castro Reservoir and sediment sampling sites (1–9, surface sediment; C, core sites). The sampling sites were georeferenced using a Garmin Model 72 GPS, according to the UTM coordinate system, datum SAD69, and central meridian 45°00 (based on Frascareli et al. 2018) 31778 Environ Sci Pollut Res (2018) 25:31776–31789 Initial Concentration) model was used for the calculating dates and sedimentation rate calculations (Robbins and Edgington 1975). Geochronology and sedimentation rate and results were presented in Cardoso-Silva et al. (2016a). Vertical profiles of 137Cs and ln (210Pbxs) (unsupported) can be observed in Fig. 2a, b. The 210Pb total shows an exponential decay throughout the core, which is due to its radioactive decay, stabilizing around the value of 210Pb supported. Two different curves of 210Pb were observed: (1) from 0- to 16-cm depth and (2) from 16- to 32-cm depth. As the core displayed two dif- ferent curves, this system presented two periods with different sedimentation rates. For sedimentation rate assessment, the natural logarithm (ln) of the 210Pbxs data was estimated for both phases. Then, regression lines for the vertical profiles of 210Pbxs curves where calculated, for the distinct periods. Having the values of a, the corresponding sedimentation rates were evaluated. The analyzed core displayed an average sed- imentation rate of 0.91 cm/year (± 0.19) from 0- to 16-cm depth, and 0.26 cm/year (± 0.19), from 16- to 32-cm depth. Based on sedimentation rates, the years of deposition for each sedimentary layer were estimated (Fig. 2c). Samples were analyzed every 2 cm. Some samples presented values below the limit of detection, and for this reason, they are not repre- sented in the graphic analysis (Fig. 2a, b). Nutrients, organic carbon, grain size, sulfate, and metals Total phosphorus was determined according to the calcination method described by Andersen (1976); after calcination, a solution of HCI 1Mwas added and the sediments were heated in water bath for1 h. The phosphorus was then determined spectrophotometrically (spectrophotometer—Micronal B572) with the molybdenum blue method described by Strickland and Parsons (1960). Total nitrogen was determined by digestion in concentrated sulfuric acid, followed by distil- lation with boric acid (APHA 2005). Grain size analysis employed the Atterberg system and the beaker method, proposed by Piper (1947) and modified by Meguro (2000). Organic matter (OM) was determined gravimetrically by weight loss on ignition (Meguro 2000) at a temperature of Environ Sci Pollut Res (2018) 25:31776–31789 31779 Fig. 2 Vertical profiles of 137Cs (a) and ln (210Pbxs) (unsupported) (b) in core sediments (adapted fromCardoso-Silva et al. 2016a) and depth and age data by CIC model (c) 500 °C. Organic carbon was estimated from OM, assuming that OM is 58% carbon by weight (Meguro 2000). Manganese and iron were also measured in the core, using US EPA method 3050B EPA SW-846 series (US EPA 1996). This method evaluates the concentration of pseudo total metals through the sediment digestion with HNO3, HCl, and H2O2 additions under heating at 90 °C. Manganese and iron were investigated to determine past redox conditions at the sediment/water interface. According to Mackereth (1966), the ratios of these elements can be used to determine whether there was oxygen in the hypolimnion. Under oxic conditions, Fe and Mn are highly insoluble. Under anoxic and reducing conditions, however, the solubilities of both elements in- crease, and they are mobilized to the water column (Ladwig et al. 2017). Manganese is more soluble than iron, and mobil- ities of these elements are closely associated with the reducing potential of the environment (Mackereth 1966; Boyle 2001; Smol 2008). Sulfide and Mn in superficial sediments were determined according to the analysis of acid-volatile sulfide (AVS) and simultaneously extracted metals (SEM), following the recom- mendations of Allen et al. (1991). Samples were stored at 4 °C before analysis (using three replicates) of the metals by atomic absorption spectrometry (AAS). Sulfide was determined spec- trophotometrically with Hach test kits (sulfide 1, cat. no. 1816-32 3; sulfide, cat. no. 1817-33), as described by Cardoso-Silva et al. (2016b). The concentrations of AVS and Mn were expressed as milligram/kilogram of dry sediment. All metals analyzed were obtained by an atomic absorption spectrophotometer—flame mode—Thermo model S. Data analysis Data were analyzed using basic descriptive and multivariate statistical analyses, performed using the PAST program (Hammer et al. 2001). Nutrient accumulation rates in the core were calculated as described by Cochran et al. (1998): Fx ¼ RxρxCx where Fx is the nutrient accumulation rates for the xth depth interval (mg cm−1 year−1); Rx is the 210Pb-derived sedimenta- tion rate for the xth interval (cm year−1); ρx is the dry bulk density of the xth interval (g cm−3); and Cx is the nutrient concentration for the xth interval (mg g−1). The use of nutrient accumulation rates core is more appropriate than simply reporting the concentrations of nutrients or other contami- nants, because such concentrations are determined by the ac- cumulation of the nutrient/contaminant of interest, relative to accumulation of other sediment components. The C/N, C/P, N/P, Fe/P, and Fe/Mn molar ratios were calculated to obtain information concerning the origin of the organic matter, redox conditions, and the potential for release of phosphorus from the sediment over time. Principal compo- nent analysis (PCA) was applied to the surface sediment data in order to elucidate the spatial distribution patterns and geo- chemistry of the nutrients. Cumulative mass water flow mass was obtained by digital processing, using ArcGIS software. Firstly, the reservoir was delimited from a Landsat image 7, an image was then generated with the combination of different bands to produce a classification according to the reflectance color emitted by the reservoir. For each color of reflectance, a code was established that ranged from 1 to 255, the higher the cumulative mass water flow the greater the numerical value. The applied geoprocessing model was based on methodology developed for the determination of the land use classification by reflectance and false color ArcGis 10.3 software as de- scribed in Ortega et al. (2016). To illustrate increasing or decreasing trends along the main axis of the reservoir, some of the graphical representations exclude sites 5 and 6, which are located in a dendritic side arm of the reservoir. These sites, however, are included in the PCA results and in the tables. Results and discussion General characteristics of the sediments Sediments of the Paiva Castro Reservoir display uniform coloration, with no clear temporal or spatial differences. Both surficial and deeper sediments are largely inorganic, with OM contents < 12% (Esteves 2011). Silt and clay particles (< 63 μm) predominated along the reservoir (Fig. 3a), except for the superficial sediments in the higher energy fluvial area (He et al. 2011; Molinaroli et al. 2009). To highlight downstream increase in clay pro- portion, only the seven sampling points located in the longitudinal axis of the reservoir are shown in Fig. 3a. The core indicates that the sand contents have decreased since the operation of the reservoir began (Fig. 3b). Silt and clay fractions dominate, especially in areas with low- er water velocity, as observed in reservoir limnetic regions (He et al. 2011). The reservoir depth presented a gradient of increase in the upstream-downstream direction (Fig. 3a, Table 1). Spatial heterogeneity of the superficial sediments Geochemistry of nutrients and carbon Concentrations of nitrogen, phosphorus, and organic carbon in the superficial sediments of the Paiva Castro Reservoir were spatially heterogeneous, with coefficient of variation values ≥ 49% (Table 1). Nitrogen and phosphorus concentrations tended to increase downstream, i.e., in the direction of the 31780 Environ Sci Pollut Res (2018) 25:31776–31789 dam (Fig. 4a, b), while the levels of carbon tended to decrease in the upstream-downstream direction (Fig. 4c). Two principal components (PCs) explained 77.21% of the total variance (Fig. 5). The main variables influencing PC1 were nitrogen (− 0.97) and phosphorus (− 0.92), responsible for the arrangement of the sites located downstream, and the carbon content (0.83), which influenced the fluvial and tran- sition area sites and the site where there was extraction of water for the public water supply (sample 6). Higher levels of organic carbon were found in the areas with higher water flow, especially in the fluvial region, under greater influence of the main reservoir-forming river, and in the catchment area for public water supply. This revealed the predominance of allochthonous (Figs. 4d and 5) organic matter, with C/N ratios of 75.74 (sample 1), 27.6 (sample 2), and 110.31 (sample 6 which lies outside the main channel). The spatial distribution of nutrients in surficial sediments could be explained by higher water flow in the upper part of the reservoir. Greater turbulence in the fluvial and transition regions, together with shallower depths, would create an unstable depositional environment, as observed by Hou et al. (2014) in Dalinouer Lake, China, and by Bartoszek and Tomaszek (2007) in Solina–Myczkowce Reservoirs, in Poland. In the dam area, however, the higher sedimentation rate (Cardoso-Silva et al. 2016a) favors greater accumulation and retention of nutrients in the sediments (Jia et al. 2017). The association between phosphorus and the clay and silt particles (Fig. 5) indicates the importance of these sediment frac- tions in binding the element. Fine-grained sediments have large surface areas that provide binding sites available for P adsorption (Wetzel 2001). Particles < 63 μm predominated along the reser- voir (silt, 59.1 ± 18.0%; clay, 27.8 ± 15.1%; Table 1), with the relative proportion of clay particles increasing in the downstream direction (Fig. 4b). Lowest water flows were observed in the limnetic zone at sites 8 and 9 (Table 1), resulting in greater accumulation of fine-grained sediments in this region, following the general pattern observed for reservoirs (Zan et al. 2012; Cardoso-Silva et al. 2016b). The second PC was mainly influenced by the variables dis- solved oxygen (0.87), pH (− 0.65), and water flow (0.53), which were responsible for the arrangement of the upstream reservoir sites. Higher water flows could explain the high oxygen concen- trations at these sites. Highest DO concentrations were found in the dam area, although the sediments throughout the reservoir displayed oxic characteristics, which was corroborated by the low values for sulfides and EH (Table 1), as observed by Cardoso-Silva et al. (2016b). Oxic conditions, together with high Mn concentrations in the region and the pH values close to neutrality (pH 6.9 ± 0.5), should favor retention of phosphorus in the sediments (Ribeiro et al. 2008;Wang et al. 2016), but there was no positive correlation between these variables and phos- phorus content. Lowest values for phosphorus and nitrogen in the sediment were found in the upstream reservoir area and could be related to high flow and turbulence associated with the shallower depth in the area which promotes an unstable environ- ment for phosphorus and nitrogen sedimentation (Bartoszek and Tomaszek 2007; Smal et al. 2013; Hou et al. 2014). C/N, C/P, and N/P molar ratios The C/N molar ratios suggested the predominance of autochtho- nous organic matter in the sediments and C/N values decline significantly, downstream matter towards the dam (Fig. 4d). Allochthonous organic matter typically has C/N ratios > 20, whereas autochthonous organic material is indicated by ratios between 4 and 10 (Meyers 1994; Hedges and Oades 1997). The phytoplankton community generally shows low molar C/N ratios (5–10), because of its high content of N-rich protein and absence of organic C-rich cellulose (Meyers 1994). The alloch- thonous material recorded in areas with higher water flow is associated with both the discharge of effluents in the upper res- ervoir area and the surrounding terrestrial vegetation in the region where there is the extraction of water for public supply. The general pattern expected in reservoirs is a greater allochthonous input in the fluvial area, with a tendency to decrease towards the dam, where there is a predominance of autochthonous 0 20 40 60 80 100 0 6 10 14 18 22 26 28 30 Grain size % De pt h cm Clay Silt Sand RO a b Fig. 3 Reservoir depth (Z) (a) and percent of particles of different sizes in the superficial sediments (a) and the sediment core (a) from the Paiva Castro Reservoir. RO, period of reservoir operation. Distances in ms from site 1. Only data for the sampling sites located in the longitudinal axis of the reservoir are shown. 1 to 9 sampling sites Environ Sci Pollut Res (2018) 25:31776–31789 31781 production that mainly originates from the phytoplankton com- munity, with C/N values between 5 and 10 (Holligan et al. 1984). The C/P molar ratios were also higher for the upstream area, with a tendency to decrease towards the dam (Fig. 4e). Themean value was 38.1 ± 60.2, with high spatial variability indicated by the coefficient of variation of 157.8% (Table 1). With respect to the Redfield C:P ratio (108:1), only sites in the fluvial area (sam- ple 1, 154:1) and in the catchment area for public water supply (sample 6, 120:1) presented values greater than or equal to this ratio. This is probably because these sites have the highest sand contents, with proportionally fewer binding sites for phosphorus. The reservoir mean for the Redfield C:P ratio was 7.8:1, with a minimum of 3.3:1 at site 7. Data obtained for the superficial sediments revealed anthropic contamination, likely caused by effluent discharges, because the relative high concentrations of phosphorus compared to the carbon concentration. The N/P ratio also showed an increase towards the dam (Fig. 4f). Different from the findings of Pompêo et al. (2017), the N/P ratio of 0.5 ± 0.9 (Table 1) suggests that nitrogen is the limiting element. For aquatic ecosystems, N/P > 16 generally indicates that phosphorus is the limiting nutrient, whereas N/P < 14 indi- cates nitrogen limitation (Elser et al. 2009). Ratios between 14 and 16 indicate that the environment is limited by either nutrient or plant growth is limited by N and P together (Elser et al. 2009). Temporal changes in the sediment core Geochemistry of nutrients and carbon The sediment core showed no significant changes over time in terms of the contents of C, N, and P concentrations, or the relationships between them, as shown by the low coefficients of variation (Table 2; Fig. 6). Nutrients (N, P) and carbon accumulation rates increased significantly between the years 1979 and 1987, a consequence of the increase in bulk sedi- mentation rate (Table 2; Fig. 6). According to Cardoso-Silva et al. (2016a), this increase could be attributed to two main events. Firstly, in 1981, two additional downstream reservoirs in the region entered into operation and the flow rate in the system increased from 11 to 33 m3/s, directly affecting the dynamics in the water body. Secondly, between 1975 and 1980, the first major urban expansion in the Juqueri River Basin occurred, following the construction of the Paiva Castro Reservoir (EMPLASA 2006). Urbanization is known to alter the rate of sedimentation in a water body (Moreira et al. 2002; Alighalehbabakhani et al. 2017), which can affect the flow and the contributions of nutrients and organic matter, as we observed. After the significant increases in the nutrient and carbon concentrations and mass accumulated, there were increases in the phosphorus and carbon contents in the 6–8-cm layer (2003) (Fig. 6). These slight increases in C and P coincided with the period in which there was a significant increase in the copper concentration to around three times the background value, as reported by Cardoso-Silva et al. (2016a), which was attributed to applications of copper sulfate as an algicide in the Cantareira systemwatershed. This period also coincided with record water scarcity in the region, when the Cantareira system reached 1% of its storage capacity (Whately and Cunha 2007). Dry periods are generally associatedwith higher phytoplankton productivity (Nogueira et al. 1999; Santos et al. 2014), mainly because nutrients become concentrated in a smaller volume of water and because residence time of the water in the reservoir increases, leading to greater production Table 1 Descriptive statistics for sampling depth (m) and for sedi- ments of the Paiva Castro Reservoir: Total nitrogen (N) total phosphorus (P), organic carbon (C), molar ratios (C/N, N/P, and C/P), grain size, sulfate, dissolved oxygen, EH, pH, and Mn. SD, standard deviation; CV, coeffi- cient of variation Unit Minimum Site Maximum Site Mean SD CV N mg/g 0.61 1 3.4 9 2.2 1.1 49.8 C mg/g 14.1 5 64.3 6 28.7 15.8 55.2 P mg/g 0.6 1 11.5 7 6.5 3.1 58.4 C/N – 5.2 5 110.3 6 29.8 37.6 126.2 C/P – 3.3 7 162.5 1 38.1 60.2 157.8 N/P – 0.5 7 2.2 1 0.9 0.5 52.5 Silt % 30.5 1 85.7 3 59.1 18.0 30.2 Clay % 7.1 1 46.7 5 27.8 15.1 54.4 Sand % 0.2 4 62.4 1 12.6 23.3 185.2 Sulfate mg/kg 9.3 7 22.3 6 16.6 8.4 50.7 DO mg/L 6.3 8 10.0 3 8.2 1.4 16.7 EH 17 7 135 1 69.4 34.1 49.1 pH 6.5 4 8.0 1 6.9 0.5 7.1 Mn mg/kg 107.1 8 690.1 1 241.7 224.8 93.0 Flux – 4 8 and 5 64 6 17.8 18.1 101.9 Depth m 0.9 1 18.1 9 10.3 5.8 55.9 31782 Environ Sci Pollut Res (2018) 25:31776–31789 of organic matter and higher levels of P and C, as recorded in the sediments. On the other hand, during this same period with increased carbon and phosphorus contents, there was a slight decrease in the N content, which could have been a conse- quence of an increase in denitrification. Denitrification is the main mechanism leading to the loss of N from top sediments (Hou et al. 2014), and here it may have been associated with decomposition of the greater amounts of organic matter ob- served for this period. Subsequently, in the topmost layers, the P and C contents decreased, returning to values close to the mean, indicating that the loss of nitrogen by denitrification was also reduced. Therefore, the dry period recorded in the area leading to a higher productivity in the system is the pos- sible main factor driving changes in accumulation rates and nutrient concentrations in sediments. From 1989 to 1999, an increase of phosphorus in surface water was observed, caused by urban growth in the Paiva Castro Basin (Giatti 2000). This increase, however, was not Fig. 4 Surface sediment a phosphorus (P), b nitrogen (N), and c carbon (C) concentrations, and the molar ratios d C/N, e C/P, and f N/P, for the Paiva Castro Reservoir related to distance downstream from site 1. Only the sampling sites located in the longitudinal axis of the reservoir are shown (1 to 9) Environ Sci Pollut Res (2018) 25:31776–31789 31783 reflected in the sediment. Using the data available from IBGE (Brazilian Institute of Geography and Statistics) for popula- tion growth and agricultural activities from 1992 to 2010, no significant Pearson correlations (p > 0.05) were observed be- tween phosphorus concentration and population growth (r = 0.31), or between phosphorus content and temporary (r = 0.31) or permanent (r = 0.37) agricultural activities. Nevertheless, this does not imply that these activities did not influence the ecosystem. For the same period, Cardoso-Silva et al. (2016a) observed a significant correlation between the copper content of the sediment and population increase, which was attributed to the application of algicides. The absence of a significant correlation between phosphorus concentrations and population numbers may be explained by the fact that the Paiva Castro reservoir experiences periods of mesotrophy. In eutrophic and mesotrophic environments, the input of P over time is typically larger than the amount ultimately stored in the sediment (Rydin 2000). According to data from CETESB (Companhia de Tecnologia de Saneamento Ambiental - Environmental Sanitation Technology Company 2005–2015), between 2005 and 2010 conditions in the dam region ranged from mesotrophic to oligotrophic, with a meso- trophic average in 6 out of the 9 years evaluated (Table 3). Although higher levels of phosphorus in sediments can be an indicative of increased nutrient inputs to an ecosystem (Liu et al. 2010; Zan et al. 2012; Hou et al. 2014), the results showed the need for the levels of phosphorus in sediments to be care- fully evaluated, taking into account the changing concentra- tions of phosphorus in the water. There were significant correlations (p < 0.05) of nitrogen with population growth (r = 0.50) and temporary agricultural production (r = 0.55). For carbon, there was only a significant Fig. 5 Principal component analysis applied to the data for the superficial sediments in the limnetic area of the Paiva Castro Reservoir Table 2 Descriptive statistics for the concentrations (mg/g) of total nitrogen (N), total phosphorus (P), and organic carbon (C); the molar ratios C/N, N/P, C/P, Fe/P, and Fe/Mn; the dry bulk sediment density (BSD) (g/cm3); and the accumulation rates AR of N, P, and C, in the sediment core collected in the limnetic zone of the Paiva Castro Reservoir. SD, standard deviation; CV, coefficient of variation;Min, min- imum value; Max, maximum value (n = 18) Unit Min Sample Max Sample Mean SD CV N mg/g 1.6 31–32 cm 3.3 26–27 cm 2.2 0.4 19.4 C mg/g 12.0 24–26 cm 26.7 6–8 cm (2003) 20.6 4.4 21.2 P mg/g 0.7 26–27 cm 1.1 20–22 cm (1971) 0.9 0.1 12.5 Fe mg/g 40.5 31–32 cm 75.5 20–22 cm (1971) 60.1 9.7 16.0 C/N – 5.3 26–27 cm 15.2 20–22 cm (1971) 10.9 2.5 23.1 C/P – 39.8 26–27 cm 80.7 24–26 cm 60.2 11.2 18.5 N/P – 4.3 6–8 cm (2003) 7.6 0–4 (2010); 16–18 cm (1987) 5.7 1.0 17.9 Fe/P – 30.4 28–29 cm 48.2 20–22 cm (1971) 38.0 5.7 14.9 Fe/Mn – 0.1 22–24 cm 0.4 28–29 cm 0.2 0.4 40.0 N AR mg/cm2/year 1.2 20–22 cm (1971) 7.4 16–18 cm (1987) 3.6 2.0 55.0 C AR mg/cm2/year 0.2 26–27 cm 1.0 16–18 cm (1987) 0.4 0.2 58.6 PAR mg/cm2/year 0.1 24–26 cm 0.3 14–16 cm (1995) 0.2 0.1 52.6 BSD g/cm3 0.2 0–4 cm (2010) 0.5 31–32 cm 0.4 0.1 27.8 31784 Environ Sci Pollut Res (2018) 25:31776–31789 correlation with permanent agricultural production (r = 0.73). Despite the increase in nutrient contents and the significant correlations between nitrogen and carbon and the anthropic activities, the sediments in the region were not considered polluted in terms of the contents of nitrogen (2.2 ± 0.4 mg/g) and phosphorus (0.9 ± 0.1 mg/g) (Table 2), according to the Ontario sediment quality guidelines (SQGs). The Ontario SQGs constitute one of the few sediment quality guides for TN and TP in aquatic environments (Hou et al. 2014) and consider sediments to be contaminated when the concentra- tions exceed 4.8 mg/g for nitrogen and 2.0 mg/g for phospho- rus (Persaud et al. 1993). The Pearson correlation analysis identified positive corre- lations of the carbon content with the silt fraction (r = 0.53) and the iron content (r = 0.57). The variation as a function of the silt fraction was because organic carbon is progressively concentrated in finer-grained sediments (Meyers 1994; Bergamaschi et al. 1997) that provide more binding sites for the adsorption of organic matter (Froehner and Martins 2008). The grain size variation is shown in Fig. 3. The correlation between the carbon and iron contents could be explained by the organic matter decomposition process, during which iron oxides are consumed (Wetzel 2001) and because organic matter also acts as a metal bind- ing phase (Cardoso-Silva et al. 2016b). C/N, C/P, and N/P molar ratios Stratigraphic fluctuations of C/N in sediment cores are interpreted as temporal shifts in the relative contributions of terrestrial and aquatic OM to the sediment (Brenner et al. 2006; Piwińska et al. 2018). The C/N molar ratios (10.2 ± 2.5) indicated a predominance of autochthonous OM over time in the dam region (Table 2; Fig. 7).When all the sediment Fig. 6 Concentrations (a) and accumulation (b) rates of P, C, and N in the sediment core from the limnetic zone of the Paiva Castro Reservoir Environ Sci Pollut Res (2018) 25:31776–31789 31785 organic matter is derived from the phytoplankton community, the C:N:P ratio should be close to the Redfield ratio (108:16:1) and the C/N ratio will be close to 6.6, as observed in Table 2. This pattern is expected in reservoirs where the allochthonous contribution generally predominates only in the fluvial areas and decreases downstream (Kimmel et al. 1990). The C/P ratios presented a mean of 60.2 ± 11.2, with a maximum of 80.7 and a minimum of 39.8. Considering the Redfield ratio (108:1), it was found an average of 54:1, there- fore a low C/P ratio. This might indicate anthropogenic source of P, or diagenetic changes in the sediment, involving C loss. Current local legislation (CONAMA no. 344/04) and the Ontario SQGs only consider the sediment to be polluted when TP > 2 mg/g, at which an alert is issued. Considering the classification of Zhang et al. (2008) for lacustrine environ- ments, the Paiva Castro sediments are moderately polluted with phosphorus. According to the latter classification, sedi- ments with total phosphorus contents < 0.5 mg/g are not con- sidered polluted by this element, while values between 0.5 and 1.3mg/g indicate moderate pollution environments and values > 1.3 mg/g indicate highly polluted sediments. The sediment quality guidelines provide useful tools, although there is a need to establish regional reference values for nutrients. A change from the reference condition and the present status is a crucial consideration in any ecological research and assess- ment program (Bennion et al. 2004). The N/P ratio obtained here (2.9 ± 0.5) suggests that nitrogen has been the limiting element over time (Table 2). Retention and release of phosphorus over time Several factors suggested that over time, the phosphorus was largely retained in the sediment, as indicated by the Fe/Mn and Fe/P molar ratios and the correlations between P and the con- servative elements Al and Fe. The Fe/Mn molar ratio remained relatively constant during the reservoir operational period, with a low coefficient of variation (22.01%) suggest- ing that oxic conditions persisted in the environment. Small changes were observed in the Mn content, together with peaks Table 3 Mean values for Carlson’s (1977) Trophic State Index (TSI) adapted for tropical environments by Lamparelli (2004): (1) at a point monitored by CETESB in the Paiva Castro reservoir dam area (1 TSI); (2) at the dam area (2 TSI); (3) mean values in two seasons in three points at the reservoir (fluvial, central, and dam areas) (3-TSI). Ultraoligotrophic (≤ 47), oligotrophic (47 < IET ≤ 52), mesotrophic (52 < IET ≤ 59), eutro- phic (59 < IET ≤ 63), hypereutrophic (63 < IET ≤ 67), supereutrophic (> 67) 1-TSI CETESB 2-TSI (Pires 2015) 3-TSI (Matta 2016) Dry season Rainy season 2005 52 – – – 2006 53 – – – 2007 55 – – – 2008 51.3 – – – 2009 55.75 – – – 2010 53 – – – 2011 53 50.2 2012 49 – – – 2013 51 – – – 2014 51 – – – 2015 54 – – – 2016 – – 46.6 ± 1.8 53.8 ± 0.8 Entries in italic mean oligotrophic 47 < IET ≤ 52; entries in bold mean mesotrophic 52 < IET ≤ 59 Fig. 7 Distribution of N/P, C/N, and C/P molar ratios for the sedi- ment core from the limnetic area of the Paiva Castro Reservoir 31786 Environ Sci Pollut Res (2018) 25:31776–31789 in Fe levels and minimum values for the Fe/Mn ratio during the period corresponding to filling of the reservoir, suggesting the existence of reducing conditions in the hypolimnion (Fig. 8). The initial phase of operation of a reservoir is usually characterized by increases in productivity, caused by the de- composition of organic matter, which can cause depletion of oxygen and a shift in the redox potential. The Fe/P molar ratio (38.0 ± 5.7) corroborated the predom- inance of oxic conditions over time. It was previously reported that freshwater sediments with a total Fe/P molar ratio > 8.5 could retain phosphate in the oxidized surface layer, but that phosphate is released when the ratio is < 8.5 (Jensen et al. 1992). When the sediment surface is oxic, strong adsorption of dissolved phosphate onto solid iron oxyhydroxides limits the efflux of phosphorus by preventing diffusion of phosphate into the water column (Katsev et al. 2006). On the other hand, under anoxic conditions, iron oxyhydroxides reductively dis- solve phosphate, which is released into the water column (Katsev et al. 2006). Significant positive correlations between total phosphorus and the contents of Fe (r = 0.51) and Al (r = 0.53) suggest favorable conditions for phosphorus immobili- zation by adsorption of inorganic forms of P onto sediment metallic oxides (Fonseca et al. 2011; Dittrich et al. 2013; Hayakawa et al. 2015) favoring the retention of P in the sediment. Conclusions and final considerations The spatial heterogeneity observed in the Paiva Castro sedi- ments is associated with both natural variations in the water body and variations induced by human impacts. The marked spatial heterogeneity, seen as a downstream decrease in the allochthonous contribution to the sediments of organic mate- rial and increases in nitrogen and phosphorus concentrations, is mainly associated with different water flow velocities and OM sources in the different compartments of the reservoir. Considering temporal variability, although there was not a significant change over time in C or N concentration, evi- denced by the coefficients of variation < 21%, these variables showed significant correlations with the population increase and the agricultural activities in the drainage basin. The sedi- ment C/P molar ratios were indicative of anthropic impacts in the region, because they were low compared to the Redfield ratio. Although there are signs of environmental degradation in the composition of the Paiva Castro Reservoir sediments, predominance of oxic conditions over space and time and particle sizes < 63 μm provided conditions favorable to the retention of phosphorus in the sediments, hence controlling the eutrophication process of the water column. We suggest future works focusing on sequential extraction procedures for the determination of phosphorus forms in sediment to better understand the burial and diagenesis of P in sediments. Anyway, the geochemical data were effective for detection of impacts in the Paiva Castro drainage basin. We recom- mended that human activities in the river basin should be managed carefully, in addition to improving sewage treat- ment. The approaches in this research could be extended to other locations, especially in mesotrophic and oligotrophic reservoirs, to provide information on historical human impacts in those ecosystems. Acknowledgements We are grateful to the Postgraduate Program in Environmental Sciences, at the Sorocaba campus of UNESP, to the Ecology Department at the Biosciences Institute and the Chemistry Department at the Oceanographic Institute of the University of São Paulo for technical assistance. The authors thank three anonymous re- viewers for their constructive comments which have improved the quality of our manuscript and Dr. Diego Javier Perez Ortega for clarifying some questions about geoprocessing. Funding information Financial support for this work was provided by FAPESP (Fundação de Amparo à Pesquisa do Estado de São Paulo, grant no. 470443/2008) and CAPES-PNPD (Coordenação deAperfeiçoamento de Pessoal de Nível Superior). Compliance with ethical standards Conflict of interest The authors declare that they have no conflict of interest. References Alighalehbabakhani F, Miller CJ, Selegean JP, Barkach J, Abkenar SMS, Dahl T, Baskaran M (2017) Estimates of sediment trapping rates for two reservoirs in the Lake Erie watershed: past and present scenar- ios. J Hydrol 544:147–155 Allen HE, Boothma W, Di Toro DM, Mahony JD (1991) Determination of acid volatile sulfide and selected simultaneously extractable metals in sediment. EPA 821-R-91-100, USEPA, Office of Water, Office of Science and Technology, Health and Ecological Criteria, Washington Ambühl H, Bührer H (1975) Technik der Entnahme ungestörter Grossproblen von Seesedimenten: ein verbessertes Boohrlot. Schweiz Z Hydrol 37:175–186 0.40 0.35 0.30 0.25 0.20 0.15 0.10 0.05 0.00 0 4 6 8 10 12 14 16 18 20 22 24 26 27 28 29 30 31 20 10 20 06 20 03 20 01 1999 19 97 19 95 19 87 19 79 19 71 19 63 80.0 70.0 60.0 50.0 40.0 30.0 20.0 10.0 0.0 Fe m g/gFe /M n Depth cm Fe/Mn ratio Fe mg/g Fig. 8 Fe/Mn molar ratio and Fe content for the sediment core from the limnetic area of the Paiva Castro Reservoir Environ Sci Pollut Res (2018) 25:31776–31789 31787 Andersen JM (1976) An ignition method for determination of total phos- phorus in lake sediments. Water Res 10:329–331 Angelini R, Bini LM, Starling FLRM (2008) Efeitos de diferentes intervenções no processo de eutrofização do lago Paranoá (Brasília - DF). Oecologia Brasiliensia 12(3):564–571 APHA - American Public Health Association (2005) Standard methods for the examination of water and wastewater. American Public Health Association, Washington Bartoszek L, Tomaszek JA (2007) Phosphorus distribution in the bottom sediments of the Solina–Myczkowce reservoirs. Environ Prot Eng 33(2):25–33 Bennion H, Fluin J, Simpson GL (2004) Assessing eutrophication and reference conditions for Scottish freshwater lochs using subfossil diatoms. J Appl Ecol 41:124–138 Bergamaschi BA, Tsamakis E, Ceil RG, Eglinton TI, Montlucon DB, Hedges JI (1997) The effect of grain size and surface area on organic matter, lignin and carbohydrate concentration, and molecular com- positions in Peru Margin sediments. Geochim Cosmochim Acta 61(6):1247–1260 Boyle JF (2001) Inorganic geochemical methods in palaeolimnology. In: Last WM, Smol JP (eds) Tracking Environmental Change Using Lake Sediments. Volume 2: Physical and Geochemical Methods. Kluwer Academic, New York, pp 83–142 Brenner M, Hodell DA, Leyden BW, Curtis JH, Kenney WF, Binhe G, Newman JM (2006)Mechanisms for organic matter and phosphorus burial in sediments of a shallow, subtropical, macrophyte-dominated lake. J Paleolimnol 35:129–148 Cardoso-Silva S, Ferreira PAL, Moschini-Carlos V, Figueira RCL, PompeoMLM (2016a) Temporal and spatial accumulation of heavy metals in the sediments at Paiva Castro Reservoir (São Paulo, Brazil). Environ Earth Sci 75:1–16 Cardoso-Silva S, Silva DCVR, Lage F, Rosa AH, Moschini-Carlos V, Pompeo MLM (2016b) Metals in sediments: bioavailability and toxicity in a tropical reservoir used for public water supply. Environ Monit Asses 188:310 Carlson RE (1977) A trophic state index for lakes. Limnol Oceanogr 22(2):361–369 Cetesb - Companhia de Tecnologia de Saneamento Ambiental (2002 to 2015) Qualidade das águas superficiais no Estado de São Paulo. CETESB: São Paulo. http://www.cetesb.sp.gov.br/agua/aguas- superficiais/35-publicacoes-/-relatorios. Accessed 30 June 2016 Cochran JK, Hirschberg DJ, Wang J, Dere C (1998) Atmospheric depo- sition of metals to coastal waters (Long Island Sound, New York U.S.A.): evidence from saltmarsh deposits. Estuar Coast Shelf Sci 46:503–522 Cunha DGF, Casali SP, Falco PB, Thornhill I, Loiselle SA (2017) The contribution of volunteer-based monitoring data to the assessment of harmful phytoplankton blooms in Brazilian urban streams. Sci Tot Environ 584-585:586–594 Dadi T,Wendt-Potthoff K, KoschorreckM (2017) Sediment resuspension effects on dissolved organic carbon fluxes and microbial metabolic potentials in reservoirs. Aquat Sci 79:749 Dittrich A, Chesnyuk A, Gudimov J, McCulloch S, Quazi J, Young J, Winter E, Stainsby GA (2013) Phosphorus retention in a mesotro- phic lake under transient loading conditions: insights from a sedi- ment phosphorus binding form study. Water Res 47:1433–1447 Eillers JM, Kann J, Cornett J, Moser K, St. Amand A (2004) Paleolimnological evidence of change in a shallow, hypereutrophic lake: Upper Klamath Lake, Oregon, USA. Hydrobiol 520:7–18 Elser JJ, Andersen T, Baron JS, Bergstrom AK, Jansson M, Kyle M (2009) Shifts in Lake N:P stoichiometry and nutrient limitation driv- en by atmospheric nitrogen deposition. Science 326(5954):835–837 EMPLASA. Empresa Metropolitana de Planejamento da Grande São Paulo (2006) Metrópoles em Dados. EMPLASA, São Paulo Esteves FA (2011) Limnologia. INEP: Interciência, Rio de Janeiro Fonseca TR, Canário MM, Barriga FJAS (2011) Phosphorus sequestra- tion in Fe-rich sediments from two Brazilian tropical reservoirs. Appl Geochem 26:1607–1622 Frascareli D, Cardoso-Silva S, Mizael JO, Lopez-Doval J, Pompêo MLM, Moschini-Carlos V (2018) Spatial distribution, bioavailabil- ity, and toxicity of metals in surface sediments of tropical reservoirs, Brazil. Environ Monit Asses 190:1–15 Froehner S, Martins RF (2008) Avaliação da composição química de sedimentos do Rio Barigüi na Região Metropolitana de Curitiba. Quim Nova 31(8):2020–2026 Giatti LL (2000) Reservatório Paiva Castro – Mairiporã – SP. Avaliação da qualidade da água sobre alguns parâmetros físicos, químicos e biológicos (1987–1998). FSP, USP, São Paulo, p. 87 Hammer Ø, Harper DAT, Ryan PD (2001) PAST: paleontological statis- tics software package for education and data analysis. Palaeontol Electron 4(1):9 Hautier Y, Niklaus PA, Hector A (2009) Competition for light causes plant biodiversity loss after eutrophication. Science 324(5927): 636–638 Hayakawa A, Ikeda S, Tsushima R, Ishikawa Y, Hidaka S (2015) Spatial and temporal variations in nutrients in water and riverbed sediments at the mouths of rivers that enter Lake Hachiro, a shallow eutrophic lake in Japan. Catena 133:486–494 He J, Lu C, Fan Q, Xue H, Bao J (2011) Distribution of AVS-SEM, transformation mechanism and risk assessment of heavy metals in the Nanhai Lake in China. Environ Earth Sci 64:2025–2037 Hedges JI, Oades JM (1997) Comparative organic geochemistry of soils and marine sediments. Org Geochem 27(7/8):319–361 Holligan SG, Montoya JP, Nevins JL, McCarthy JJ (1984) Vertical dis- tribution and partitioning of organic carbon in mixed, frontal and stratified waters of the English Channel. Mar Ecol Prog Ser 14:111– 127 Hou D, He J, Lu C, Dong S, Wang J, Xie Z, Zhang F (2014) Spatial variations and distributions of phosphorus and nitrogen in bottom sediments from a typical north-temperate lake, China. Environ Earth Sci 71:3063–3079. https://doi.org/10.1007/s12665-013-2683-6 Jensen HS, Kristensen P, Jeppesen E, Skytthe A (1992) Iron: phosphorus ratio in surface sediments as an indicator of phosphorus release from aerobic sediments in shallow lakes. Hydrobiologia 235(236):731– 743 Jia X, Luo W, Wu X, Wei H, Wang B, Phyoe W, Wang F (2017) Historical record of nutrients inputs into the Xin’an Reservoir and its potential environmental implication. Environ Sci Pollut Res 24: 20330–20341 Katsev S, Tsandev I, Heureux I, Rancourt DG (2006) Factors controlling long-term phosphorus efflux from lake sediments: exploratory reactive-transport modeling. Chem Geol 234:127–147 Kimmel BL, Lind OT, Paulson LJ (1990) Reservoir primary production. In: Thorton KW, Kimmel BL, Payne FE (eds) Reservoir limnology: ecological perspectives. John Wiley, New York, pp 133–193 Koerselman W, Meuleman AFM (1996) The vegetation N:P ratio: a new tool to detect the nature of nutrient limitation. J Appl Ecol 33(6): 1441–1450 Ladwig R, Heinrich L, Singer G, Hupfer M (2017) Sediment core data reconstruct the management history and usage of a heavily modified urban lake in Berlin, Germany. Environ Sci Pollut Res 24(32): 25166–25178 Lamparelli MC (2004) Graus de trofia em corpos d’água do Estado de São Paulo: avaliação dos métodos de monitoramento. Doctoral the- sis - Universidade de São Paulo Liu JY, Wang H, Yang HJ, Ma YJ, Cai OC (2009) Detection of phospho- rus species in sediments of artificial landscape lakes in China by fractionation and phosphorus-31 nuclear magnetic resonance spec- troscopy. Environ Pollut 157(1):49–56 Liu SM, De Zhu B, Zhang J, Wu Y, Liu GS, Deng B, Zhao MX, Liu GQ, Du JZ, Ren JL, ZhangGL (2010) Environmental change in Jiaozhou 31788 Environ Sci Pollut Res (2018) 25:31776–31789 http://www.cetesb.sp.gov.br/agua/aguas-superficiais/35-publicacoes-/-relatorios http://www.cetesb.sp.gov.br/agua/aguas-superficiais/35-publicacoes-/-relatorios https://doi.org/10.1007/s12665-013-2683-6 Bay recorded by nutrient components in sediments. Mar Pollut Bull 60:1591–1599 Mackereth FJH (1966) Some chemical observations on post-glacial lake sediments. Phil Trans R Soc Lond B250:165–213 Matta AL (2016) Dinâmica do plâncton na represa Paiva Castro (Sistema Cantateira, Mairiporã, São Paulo, Brasil. Master thesis - Universidade de São Paulo Meguro M (2000) Métodos em Ecologia. São Paulo. Apostila de Metodologias para a disciplina BIE - 321 Ecologia Vegetal. Instituto de Biociências, USP, São Paulo Meyers PA (1994) Preservation of elemental and isotopic source identi- fication of sedimentary organic matter. Chem Geol 144:289–302 Molinaroli E, Guerzoni S, De Falco G, Sarretta A, Cucco A, Como S, Simeone S, Perilli A, Magni P (2009) Relationships between hydro- dynamic parameters and grain size in two contrasting transitional environments: the Lagoons of Venice and Cabras, Italy. Sediment Geol 219(1):196–207 Moreira SRD, Fávaro DIT, Campagnoli F, Mazzili BP (2002) Sedimentations from the reservoir Rio Grande (São Paulo/ Brasil). In: Warwick P (ed) Environmental radiochemical analysis II. Royal Society of Chemistry, Maidstone, pp 383–390 Moschini-Carlos V, Bortoli S, Pinto E, Nishimura PY, Pompêo M (2009) Cyanobacteria and cyanotoxin in the Billings Reservoir (São Paulo, SP, Brazil). Limnética 31:227–236 Nogueira MG, Henry R, Maricatto FE (1999) Spatial and temporal het- erogeneity in the Jurumirim reservoir, São Paulo, Brazil. Lakes Reserv Res Manag 4(3–4):107–120 Oluyedun OA, Ajayi SO, Van Loon GW (1991) Methods for fraction- ation of organic phosphorus in sediments. Sci Tot Environ 106:243– 252 Ortega DJP, Pérez DA, Américo JHP, Carvalho SL, Segovia JA (2016) Development of index of resilience for surface water in watersheds. JUEE 10(1):72–82 Perkins RG, Underwood GJ (2001) The potential for phosphorus release across the sediment-water interface in an eutrophic reservoir dosed with ferric sulphate. Water Res 35(6):1399–1406 Persaud D, Jaagumagi R, Hayton A (1993) Guidelines for the protection and management of aquatic sediment quality in Ontario. Water Resources Branch, Ontario Ministry of the Environment, Toronto Piper CS (1947) Soil and plant analysis. Interscience Publishers, New York Pires DA, Tucci A, Carmo-Carvalho M, Lamparelli MC (2015) Water quality in four reservoirs of the metropolitan region of São Paulo, Brazil. Acta Limnol Bras 27(4):370–380 Piwińska D, Gruca-Rokosz R, Bartoszek L, Czarnota J (2018) Spatial diversity characterising certain chemical substances in sediments of Besko Reservoir. JEE 19(1):104–112 PompêoMLM,Moschini-Carlos V, Lopez-Doval JC, Abdalla-Martins N, Cardoso-Silva S, Freire RHF, Beghelli FGS, Brandimarte AL, Rosa AH, López P (2017) Nitrogen and phosphorus in cascade multi- system tropical reservoirs: water and sediment. Limnol Rev 17(3): 133–150 Ribeiro DC, Martins G, Nogueira R, Cruz JV, Brito AG (2008) Phosphorus fractionation in volcanic lake sediments (Azores, Portugal). Chemosphere 70(7):1256–1263 Rippey B, Anderson NJ (2008) The accuracy of methods used to estimate the whole-lake accumulation rate of organic carbon, major cations, phosphorus and heavy metals in sediment. J Paleolimnol 39:83–99 Robbins JA, Edgington DN (1975) Determination of recent sedimenta- tion rates in Lake Michigan using Pb-210 and Cs-137. Geochim Cosmochim Acta 39:285–304 Ruttenberg KC, Gofii MA (1997) Phosphorus distribution, C:N:P ratios, and 613C, in arctic, temperate, and tropical coastal sediments: tools for characterizing bulk sedimentary organic matter. Mar Geol 139: 123–145 Rydin E (2000) Potentially mobile phosphorus in Lake Erken sediment. Water Res 34(34):2037–2042 SABESP- Companhia de Saneamento Básico do Estado de São Paulo (2018) SABESP, São Paulo Santos JCN, de Andrade EM, de Araújo Neto JR, Meireles ACM, de Queiroz Palácio HA (2014) Land use and trophic state dynamics in a tropical semi-arid reservoir. Rev Ciên Agron 45(1):35–44 Silva EAS (2002) Eutrofização no Reservatório Paiva Castro do Sistema Cantareira na Região Metropolitana de São Paulo (1987-1997). São Paulo, FSP- USP, p 135 Smal H, Ligęza S, Baran S, Ojcikowska-KapustaA OR (2013) Nitrogen and phosphorus in bottom sediments of two small dam reservoirs. Pol J Environ Stud 22(5):1479–1489 Smith VH, Schindler DW (2009) Eutrophication science: where dowe go from here? Trend Ecol Evol 24(4):201–207 Smol JP (2008) Pollution of lakes and rivers- a paleoenvironmental per- spective. Blackwell, Oxford, 382 pp Strickland JD, Parsons TR (1960) A manual of seawater analysis. Bull Fish Res Board Can (125):1–185 US EPA United States Environmental Protection Agency (1996) Method 3050B. Acid digestion of sediments, sludges and soil. Revision 2. Washington DC Wang J, Chen J, Ding S, Guo J, Christopher D, Dai Z, Yang H (2016) Effects of seasonal hypoxia on the release of phosphorus from sed- iments in deep-water ecosystem: a case study in Hongfeng Reservoir, Southwest China. Environ Pollut 219:858–865 Wetzel RG (2001) Limnology- lake and river ecosystems. Academic Press, California, 1066 pp Whately M, Cunha PM (2007) Cantareira 2006: Um olhar sobre o maior manancial de água da Região Metropolitana de São Paulo - Resultados do diagnóstico socioambiental participativo do Sistema Cantareira. Instituto Sócio Ambiental, São Paulo Zan F, Huo S, Xi B, Zhu C, Liao H, Zhang J, Yeager KM (2012) A 100- year sedimentary record of natural and anthropogenic impacts on a shallow eutrophic lake, Lake Chaohu, China. J Environ Monit 14: 804–816 Zhang RY, Wu F, Fu P, Li W, Wang L, Liao H, Guo J (2008) Characteristics of organic phosphorus fractions in different trophic sediments of lakes from the middle and lower reaches of Yangtze River region and Southwestern Plateau, China. Environ Pollut 152: 366–372 Environ Sci Pollut Res (2018) 25:31776–31789 31789 Factors... Abstract Introduction Material and methods Study area Sampling Laboratory analyses Sedimentation rate and 210Pb dating Nutrients, organic carbon, grain size, sulfate, and metals Data analysis Results and discussion General characteristics of the sediments Spatial heterogeneity of the superficial sediments Geochemistry of nutrients and carbon C/N, C/P, and N/P molar ratios Temporal changes in the sediment core Geochemistry of nutrients and carbon C/N, C/P, and N/P molar ratios Retention and release of phosphorus over time Conclusions and final considerations References