| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Univ. Federal Rural de Pernambuco/Unidade, Acadêmica de Serra Talhada, Fazenda Saco c.p. 063, 56.900-000, Serra Talhada, Pernambuco, Brazil
b Depto. de Ciência do Solo, Escola Superior de Agricultura "Luiz de Queiroz"/Univ. de São Paulo, Av. Pádua Dias 11, 13.418-900, Piracicaba, São Paulo, Brazil
c Instituto de Ciencias Agrarias/CSIC, Calle Serrano, 115, Dup, 28006 Madrid, Spain
d Depto. de Edafologia e Química Agrícola, Faculdade de Bioloxia, Univ. de Santiago de Compostela, 15782 Santiago de Compostela, A Coruña, Spain
* Corresponding author (pablo{at}esalq.usp.br).
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: IGUA, Caranguejo Island • ITA, Itapanhaú • River • PM, Pai Matos Island • RE, Escuro River • RF, Fazenda River • XRD, x-ray diffraction
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Despite the importance of these environments on the shorelines of tropical zones, relatively few studies have been performed. The existing studies have been mainly environmental in nature and involve retention processes and the geochemical behavior of trace metals (Clark et al., 1998; Andrade and Patchineelam, 2000; Marchand et al., 2006), nutrient and organic matter cycling (Alongi et al., 1998, 2005; Bosire et al., 2005), sedimentation dynamics (Furukawa et al., 1997; Smoak and Patchineelam, 1999), and paleoenvironmental reconstruction (Behling and Costa, 2001, 2004; Cohen et al., 2005).
The minerals present in estuarine areas are of mainly detrital origin, being mixtures of terrestrial and marine sediments. Terrestrial sediments become suspended in river water and are washed to estuarine areas, while marine sediments were deposited during past transgressive events or are continually deposited by tidal currents, although authigenesis may also occur in these environments (Chamley, 1989).
Relatively few of the studies involving mangrove soils have considered their mineralogical constitution. Such studies can contribute to our comprehension of the processes of distribution of heavy metals in the environment (Caliani et al., 1997; Dong et al., 2000; Liaghati et al., 2003). Furthermore, knowledge about mineralogical assemblages allows us to understand the physicochemical conditions occurring in the environment, as investigated by Velde and Church (1999), Dillenburg et al. (2000), and Ku and Walter (2003), who evaluated the stability and authigenesis of soil minerals in tidal plains. The mineralogical composition has also been used as a tracer of the origin of sediments in estuarine and marine environments (Petschick et al., 1996; Heroy et al., 2003).
The most important studies of the mineralogy of mangrove soils are those of Marius and Lucas (1991) in mangrove ecosystems on the shoreline of western Africa, where the predominance of quartz, kaolinite, and smectite in the soils was observed, as well as the presence of pyrite, jarosite, and halite. Studies have also been performed on the mangrove soils on the Australian eastern coast (New South Wales), in which quartz, kaolinite, illite, and feldspar were identified in the clay fraction (Smith, 1996).
In a study of Brazilian mangrove soils in the State of Espírito Santo, Marius et al. (1987) identified an assemblage constituted mainly of kaolinite and gibbsite, as well as illite and interlayered smectite–illite. These researchers attributed the presence of gibbsite in the mangrove soils to the supply of coastal Tertiary sediments from the Barreiras Formation, which correspond to areas adjacent to the estuary. The presence of gibbsite as well as illite, illite–smectite, and kaolinite was also observed in sediments from Guanabara Bay in Rio de Janeiro (Faria and Sanchez, 2001). In mangrove ecosystems in the Amazon region, Behling and Costa (2004) and Costa et al. (2004) identified mainly smectite, kaolinite, and pyrite, as well as anatase, feldspar, and halite. Recently, in the State of São Paulo, Prada-Gamero et al. (2004) identified minerals of allochthonous origin (kaolinite and mica) and others of autochthonous origin (smectite and Fe-mica) in the clay fraction of a mangrove soil in the Baixada Santista region.
Due to the lack of detailed studies of the mineralogical composition of mangrove soils and in view of the relevance of the topic to scientists investigating coastal environments, the aim of the present study was to determine the representative mineralogical assemblages in the clay and silt size fraction, and to identify potential contrasts among them in the different sectors on the coast of the State of São Paulo (Brazil). For this, the mineral constituents were identified in 11 mangrove soils, which were selected according to their geomorphological position on the coast of the State of São Paulo.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Geomorphologically, the coast of São Paulo State is divided into two distinct regions. In the northern region, virtually the entire extension of the crystalline basement reaches the sea and mangrove forests are relatively scarce, small, and concentrated at the river mouth. The southern region comprises large plains, essentially formed by marine or fluvial-lacustrine deposits (Suguio and Martin, 1978). Mangrove forests are common there and are distributed along the several channels and rivers of the Cananéia–Iguape lagoon estuarine system. The Baixada Santista region is located in the central zone, linking the two distinct regions. It is slightly cut by the crystalline basement and also comprises a large coastal plain. There are also several mangrove forests associated with the Santos estuary (Lamparelli, 1998) in this region. The crystalline basement mainly consists of granites, gneisses, schists (mica schists and quartz schists), and migmatites, with some alkaline rocks present (Comissao de Solos, 1960).
The mangrove forests under study are located in the three areas of the coast of São Paulo State (Fig. 1 ). In the southern area, the following mangrove forests were sampled: Sítio Grande River (sampled at two points: SG1 and SG2), Ipaneminha River (IPA), and the transition between the Restinga and Ipaneminha rivers (TRA), Pai Matos Island (PM), Baguaçu Island (BAG), and Caranguejo Island (IGUA). Itapanhaú River (ITA) and Guaratuba River (GUA) mangrove forests in the Baixada Santista estuary were also sampled. In the northern area, Escuro River (RE) and Fazenda River (RF) mangrove forests were sampled.
|
To evaluate the mineralogy of the sediment that reaches the estuary, suspended sediments were collected from the main river in the southern area, the Ribeira do Iguape River, in Valo Grande next to an old dam. A pumping system was used to collect 200 L of water, which was stored in plastic containers and transported to the laboratory to obtain the sediment.
The soil samples were treated to remove organic matter and soluble salts. The sand and silt-plus-clay fractions were then separated by moist sieving and then the silt and clay were separated by decantation (Jackson, 1969).
X-ray diffraction (XRD) analysis was done on glass slides for oriented aggregates of the clay fraction saturated with Mg2+ and K+, and for random powder mounts of unsaturated clays and silt fractions. The K+–saturated samples were analyzed at 25°C (K25) and after heating to 300°C (K300) and 550°C (K550) for 3 h, whereas the samples saturated in Mg2+ were processed like oriented aggregates and were also solvated in ethylene glycol (Mg-E). In addition, the concentration of Fe oxides was determined by dissolution in NaOH (5 mol L–1), according to Kämpf and Schwertmann (1982), in the samples of the clay fraction of IGUA and PM, which had been processed as random powder. The analysis of the sediment sample was performed on bulk samples as oriented aggregates without any pretreatments.
The XRD patterns were obtained by a Philips PW 1130 diffractometer with a graphite monochromator, using Cu-K
radiation and applying 45 kV and a 40 mA current, with a scanning rate of 0.02° 2
s–1.
Transmission electron microscopy was used to analyze the clay fraction with the aim of identifying the morphology of the minerals present. For this, a transmission electron microscope Model STEM LEO 910 (Carl Zeiss, Oberkochen, Germany) was used. The samples were dispersed in dilute aqueous NH3 after drying in a copper net and, finally, they were recovered with a carbon pellicle.
The images of pyrite were obtained from the dense fraction of the fresh soil sample, which was separated with bromoform (density of 2.89 kg L–1). The sample was observed in a scanning electronic microscope, Model LEO 435 VP analyzer (Leica, Cambridge, UK), fitted with an energy dispersive spectrometry analysis system, Model Oxford ISIS 300 (Oxford Univ., Oxford, UK), at 20 kV and waiting time of 100 s.
The semiquantitative estimations of quartz, gibbsite, goethite, and total phyllosilicates was performed by examination of the XRD patterns of random powder mounts by integration of the area of maximum diffraction at 0.426, 0.483, 0.418, and 0.444 nm, respectively, for those minerals, and by using intensity factors, according to Schultz (1964) and Biscaye (1965). The relative abundance of kaolinite, smectite, vermiculite, and mica was estimated similarly, by examination of the XRD patterns of samples saturated with Mg2+, in peak areas at 0.72 nm for kaolinite, 1.00 nm for mica, 1.41 nm for vermiculite, and 1.64 nm for smectite from samples in Mg-E. Due to the applied method, the presented values do not have an absolute meaning, but they present internal values relative to the samples.
|
RESULTS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
Mica was identified by peaks at 1.00, 0.5, and 0.334 nm. In the samples from the northern region (RE and RF), however, the 0.5-nm peak was observed either at very low intensity or was absent. Kaolinite was identified by the presence of peaks at 0.712 and 0.356 nm in the silt fraction of these soils (Fig. 2).
The presence of vermiculite was observed in the RF mangrove soil, mainly at a depth of 60 to 80 cm. The XRD patterns obtained with random powder mounted silt, silt on oriented aggregates, solvated in ethylene glycol, and heated to 550°C are shown in Fig. 3a . According to the patterns shown, there is a peak at 1.4 nm, which was not displaced after treatment with ethylene glycol, but collapsed after heating to 550°C.
|
According to the XRD patterns obtained (Fig. 4–7
), the mineralogical assemblage of the clay fraction of soils consisted of smectite, vermiculite, illite, kaolinite, gibbsite, goethite, quartz, and feldspars. As in the silt fraction, the accessory minerals found and identified in all the samples were quartz, feldspars, and mainly gibbsite.
|
|
|
|
Mica and kaolinite were found at all sampling points. Mica was identified by the peaks at 1.00, 0.50, and 0.334 nm, and it must be pointed out that in the northern mangrove forest soils (Fig. 7), the 0.50-nm peak occurred at a very low intensity when compared with the 1.00-nm peak. Kaolinite was demonstrated by the presence of peaks at 0.712 and 0.356 nm, which disappeared after heating to 550°C. Kaolinite presented generally intense peaks at similar levels in almost all samples, of low intensity only in the subsuperficial samples of ITA and RF.
Except in the RF mangrove soils, smectite was identified in all the samples by the presence of a peak at 1.41 nm, which changed to 1.68 nm in Mg-E. The highest intensity of the peak corresponded to the PM mangrove soil, and the lowest intensity to the IGUA soil, mainly in the subsuperficial layer. The ITA and RE samples showed similar patterns to each other, with the level of peak intensity intermediate between the levels of PM and IGUA. The 1.41-nm peak did not change position after treatment in ethylene glycol and collapsed to 1.00 nm in K550, proving the presence of vermiculite in the RF soils (Fig. 7).
The semiquantitative data for phyllosilicates and some accessory minerals are shown in Table 1 . The results show that the most abundant phyllosilicate in PM and IGUA was kaolinite (38–55%), except in the subsurface of PM, in which smectite prevailed (36%). Traces of gibbsite and quartz were found in these mangrove soils, with the latter mainly occurring (11%) in the subsurface of PM. Goethite was present in small amounts in the 0- to 20- and 60- to 80-cm layers of the IGUA mangrove soil (1 and 3%, respectively). In the Baixada Santista, smectite was the most abundant phyllosilicate in the ITA mangrove forest soil (36 and 66% in the surface and subsurface, respectively), with gibbsite (1%) and traces of quartz also occurring. The northern mangrove soils presented a significantly higher amount of gibbsite than the other mangrove soils, with values varying between 2 and 24% and the highest concentrations in RF (14 and 24%), highlighting an increasing gradient of concentration moving northward along the shoreline for this mineral. Trace amounts of quartz were found in these mangrove soils and there was a predominance of smectite in RE (39 and 38%), whereas RF presented a greater abundance of kaolinite in the surface (54%) and mica in the subsurface (60%). Transmission electron microscopy enabled observation of the morphology of some minerals identified by XRD, such as kaolinite, which presented a plain hexagonal shape (Fig. 8a ). The morphology observed in the images in Fig. 8c and 8d can be attributed to the shapes of vermiculite and smectite, respectively. The structures resembling long tubes (Fig. 8a and 8b) suggest the presence of the tubular form of halloysite in the clay fraction in soils of the RE mangrove forest.
|
|
|
|
|
DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The different peaks corresponding to feldspar may indicate the presence of either potassic feldspars or plagioclases. The presence of these minerals in soils generally is related to the sand and silt fractions, although they are also reported to occur in the coarse clay fraction and rarely in the fine clay fraction (Brown, 1980). Quartz follows a pattern similar to the feldspars regarding occurrence, according to the granulometry presented in a review by Drees et al. (1989). These minerals originate from the weathering of rocks of the crystalline complex of the Serra do Mar, the primary source of minerals to the sediments of the São Paulo coastal plain, but may also originate from sandy coastal sediments transported during the last Holocenic transgressive events.
Particle abrasion, which is typical in the environment studied, explains the fractionation and maintenance of primary minerals in the clay fraction of mangrove soils, which in turn is related to the stability ensured by the geochemical environment rich in basic cations provided by the sea water and the high concentration of biogenic Si, as described by Michalopoulos and Aller (2004). The presence of anatase in mangrove soils is related to the terrestrial origin of the sediments and to the strong resistance of this mineral to weathering (Milnes and Fitzpatrick, 1989).
Crystalline Fe oxyhydroxides, such as goethite, are unstable in anoxic SO4–reducing conditions and can be readily reduced by microbial activity and transformed into pyrite in a very short period (days) (Canfield et al., 1993). The main end product of SO4 reduction in these types of environments should be pyrite (Howarth and Merkel, 1984). In the present study, the pH–Eh conditions are close to the stability field of pyrite (see, e.g., Fanning et al., 2002), which is consistent with the presence of this mineral in these soils (Table 2 , Fig. 10). In a study involving sequential chemical extraction, Ferreira et al. (2007b,c) also found that pyrite is the most abundant form in mangrove soils from the southern shoreline and Baixada Santista.
|
Goethite was detected by XRD in the IGUA soil because of the particular conditions that occur in this area. The IGUA mangrove forest is located close to the Valo Grande, i.e., in the mouth of a branch of the river Ribeira do Iguape. The suspended sediments, which contain goethite, are deposited directly on the IGUA soil (Fig. 3b). Further study is required, however, regarding the presence of goethite at depth. Bioturbation by crabs may explain the presence of this mineral at depth. The effects of this kind of bioturbation on the geochemical behavior of Fe in mangrove soils in Brazil have been discussed by Ferreira et al. (2007b).
Pyrite is formed by the reduction of Fe3+ supplied by terrestrial sediments and SO42+ supplied by sea water, and the reaction is mainly promoted by SO4–reducing bacteria (Desulfovibrio spp.) (Doner and Lynn, 1989). This mineral occurs as individual cubes in the clay fraction and as framboids in the silt fraction (Fanning et al., 2002).
The soils in areas adjacent to the coastal plain in São Paulo State present a mineralogical assemblage that is mainly comprised of kaolinite and gibbsite (Rossi, 1999; Furian et al., 2002; Ishida, 2005). This explains the presence of gibbsite in the mangrove soils under study. The decrease in the gibbsite levels moving southward along the shoreline (Fig. 4–7, Table 1) may be related to the longer distance between the mangrove forests and the Serra do Mar. Thus, northern mangrove soils, which are located closer to the source, presented higher levels of gibbsite. Southern mangrove soils contained lower levels of gibbsite, but they are farther away from the source and located in large estuarine systems, where the material has been subject to different processes and has probably been present for a longer period. The presence of significant quantities of gibbsite in mangrove soils has been reported by Marius et al. (1987), who studied mangrove soils in the Brazilian State of Espírito Santo, relating the presence of gibbsite with soils of adjacent areas, originating from Tertiary sediments of the Barreiras Formation.
Gibbsite is not stable in mangrove soils because, according to Michalopoulos and Aller (2004), the existence of high concentrations of biogenic silica and basic cations (Ca and Mg), mainly in estuarine environments, favors reverse weathering processes, i.e., the neoformation of 1:1 phyllosilicates from Al oxyhydroxides. This process would involve readsorption of Si in the gibbsite structure to form kaolinitic minerals, as suggested by Macías and Ojea (1980) for salt marsh environments. According to the consideration above, gibbsite would not be expected to be found in large amounts in mangrove soils.
It can therefore be inferred that the diminishing levels of gibbsite moving southward along the shoreline are related to the sedimentation dynamics. Since Baixada Santista and the south shoreline are more closed environments and the sediments are remobilized within these systems, the particles therefore reside longer in the system, favoring the process of transformation from gibbsite to kaolinite as described above.
Kaolinite is commonly found in estuarine environments, in both tropical and temperate climates (Behling and Costa, 2001, 2004; Belzunce-Segarra et al., 2002; Prada-Gamero et al., 2004). In the present study, kaolinite occurred in the silt and clay fractions in all studied mangrove soils. Like gibbsite, kaolinite is an allochthonous mineral that originates from kaolinitic soils adjacent to the coastal plain. Although generally widespread, kaolinite was most abundant in the IGUA mangrove soil, which can be explained by the constant discharge of kaolinitic sediment via the Ribeira do Iguape River (Fig. 9).
Neogenesis of kaolinite via the reverse weathering process, as described above for gibbsite, must also be considered in mangrove soils. The kaolinite present in the silt fraction is possibly a pseudomorphous kaolinite resulting from the weathering of biotite (Kretzschmar et al., 1997; Jolicoeur et al., 2000).
Furthermore, at least part of the smectite in these soils is of detrital origin, and related to the last transgressive and regressive events, i.e., the Cananéia Transgression occurred approximately 120,000 yr before present (BP), correlated to the Sangamonian in North America (Chappell, 1983), and the Santos Transgression, which occurred 5100 yr BP. In the former transgressive event, at least part of the smectites may have been formed in the continental platform, since marine smectites are primarily formed in this environment (Hillier, 1995). Moreover, the platform sediments would have been exposed during the marine regression of the last glacial maximum (about 17,000 yr BP) and, finally, they could have been remobilized to the current coastline during the Santos Transgression (5100 yr BP).
Authigenesis of smectite in mangrove soils should also be considered. Chamley (1989) suggested that the precipitation of Fe and Si ions dissolved in pore water may occur, which forms neogenic nontronite. Velde and Church (1999) reported that cyclic changes in the redox potential at the soil surface of a salt marsh led to the formation of smectitic minerals from the transformation of illite and chlorite. Smectite may also be formed in this way in mangrove soils as these environments are the tropical equivalent of salt marshes. Smectites are unlikely to be of terrestrial origin in mangrove soils in humid tropical regions. Terrestrial soils around the studied coastal plain rarely contain smectites (Sakai and Lepsch, 1984; Rossi, 1999; Furian et al., 2002; Ishida, 2005).
The mica present in mangroves of the southern coast and the Baixada Santista is possibly muscovite in the silt fraction and illite in the clay fraction. Because of the absence or low intensity of a 0.5-nm peak, however, the mica in the silt and clay fractions in the mangrove soils from the northern coast must be biotite or even a glauconitic mineral (Fanning et al., 1989; Moore and Reynolds, 1989). In the present study, biotite probably occurred in the clay fraction in northern mangrove soils (RE and RF) because of the close proximity between the complex of crystalline rocks of the Serra do Mar and the mangrove areas. Moreover, the rocks in the crystalline complex around the mangrove forests of the northern shore contain higher levels of maphic minerals (Instituto de Pesquisas Tecnológicas, 1981).
The vermiculite identified in the silt fraction as well as in the clay fraction in RF probably originated from the Serra do Mar soils, since this mineral can occur secondarily in these soils (Furian et al., 2002; Ishida, 2005). It is probably trioctahedral vermiculite originating from the direct transformation of biotite (Douglas, 1989), as suggested by the presence of this mica in the silt and clay fractions in RF.
The transmission electron micrographs demonstrated the presence of tubular halloysite in the RE soil. Halloysite may be formed by the weathering of biotite present in granite and gneiss (Kretzschmar et al., 1997) and also of feldspars (Dixon, 1989). In wet, warm climates under well-drained conditions, generally biotite or feldspars transform to kaolinite (Kretzschmar et al., 1997; Dixon, 1989), so the presence of halloysite in mangrove soils probably would be conserved only if it was kept constantly under poorly drained conditions, as these soils are.
|
CONCLUSIONS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Received for publication May 31, 2007.
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  F. P. Ferreira, P. Vidal-Torrado, P. Buurman, F. Macias, X. L. Otero, and R. Boluda Pyrolysis-Gas Chromatography/Mass Spectrometry of Soil Organic Matter Extracted from a Brazilian Mangrove and Spanish Salt Marshes Soil Sci. Soc. Am. J., May 1, 2009; 73(3): 841 - 851. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Vadose Zone Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
