Hydrological control on the temporal variability of trace element concentration in the Amazon River and its main tributaries


Patrick Seyler, Laurence Maurice-Bourgoin, Jean Loup Guyot
ORSTOM, CP 7091, Lago Sul, CEP 71619-970 Brasília, DF, Brasil

Geraldo Boaventura
Instituto de Geociências, Universidade de Brasília, CEP 70910-900 Brasília, DF, Brasil

Corresponding author: seyler@unb.br



            The chemical composition and variability of the Amazon river are of interest for various reasons, including (i) as a major source of dissolved and particulate substances to the Atlantic ocean,  (ii) as a case study for furthering the understanding of trace element geochemistry in a major fluvial system and (iii) as an evaluation of the potential contamination of the river waters.

The Amazon mainstream is formed by the Solimões and the Madeira rivers, both draining the Andean part of the basin, and the Rio Negro river, volumetrically the largest tributary, which drains the inundaded forest of the Guyana Shield and Central Amazon.

During the 1965-1990 period the mean annual water discharge for the whole basin has been estimated to 209 000 m3.s-1 at the mouth (DNAEE, 1994) and 168 700  m3.s-1 at Óbidos gauging station situated upstream the marine influence and controlling 80% of the total discharge of Amazon river to the Ocean (Fig. 1). During the same period the contribution of Solimões at Manacapuru (confluence with the Rio Negro) has been of 103 000 m3.s-1, those of Negro and Madeira 28000 m3.s-1 and 31200 m3.s-1 respectively. 95% of the Amazon discharge at Óbidos comes from these three sources (Fig.2). The proportion of water originated from the Solimões, Negro and Madeira rivers varies with total discharge during the annual cycle (Molinier et al., 1996). Concerning sediment transport, the more recent results obtained by Callède et al., (1997) give a mean annual discharge of suspended sediment close to 600 106 tons at Óbidos station, where 97 % is due to the contribution of Andean tributaries (62% from the Solimões and 35% from the Madeira). The contributions of Negro, Trombetas, Tapajos and Xingú account for less than 3%. An examination of the suspended sediment concentrations (SSM) and discharge vs. time at Óbidos indicates that plots of the relations between sediment discharge and water discharge will form loops rather than straight line. During the hydrological period, SSM concentrations shows high frequency variations (10 days) and the sediment peak discharge precede of about three months the maximum water discharge (Fig. 3).

In order to assess the nature of trace element (V, Cr, Mn, Co, Cu, Zn, As, Rb, Sr, Mo, Cd, Sb, Cs, Ba, U) in the Amazon, a monthly time series covering a whole hydrological cycle was obtained at the Óbidos gauging station.



After filtration on 0.22 µm Millipore filters, in the laboratory of the research vessel, trace elements analysis were performed by Inductively Coupled Plasma-Mass Spectrometry (ICP-MS) using a VG-Elemental PQ2 instrument at the laboratoire de Géochimie (Montpellier University, France).

Ultra-clean sampling technique and analysis procedure validated by previous studies (Seyler & Elbaz-Poulichet, 1996; Elbaz-Poulichet et al., 1997) were used and the data reported here are consistent with those published by Gaillardet et al., 1997.




With regard to the temporal variation of the dissolved load, illustrated by the record of specific conductivity at Obidos, it shows that highest values are corresponding to the lowest flow discharge and lower values are corresponding to the falling stage, although the water contribution of Rio Negro is maximum. During the middle rising stage, conductivity is rather stable, due to the inputs of Madeira River (Fig.3)

            Concerning the trace element variation in time, several patterns are shown. Some metals concentrations (Mn, Co) show a high degree of variability and increase with increasing discharge. Other metals vary in a lesser extent and their concentrations decrease when the discharge increases (V, Mo, Cu, Cd, Sb, Cs, Rb, Sr, Ba, and U). The last set of elements shows relatively little variation in concentration (Cu, Ni, Co) and does not display consistent relationship with discharge. According to previous study (Konhauser et al., 1994, Gaillardet et al., 1997) there is little evidence of anthropogenic perturbation of dissolved metal concentrations in the Amazon, and these differences may be due to natural causes.

            Several explanations have been found:

* Variations of river chemistry may reflect variations of the sources. The Negro river has typically depleted in As, Mo, Sr, Ba, Cu, V as compared with Solimões concentration by a factor ranging from 26 (As, Sr) to 3 (V). Conversely, Negro waters are more enriched in Ni and Cs, and Madeira waters in Sb (by a factor of 3). The increased proportion of waters from the less solute Negro River during the high discharge period of Amazon contributes to the observed decrease of concentrations.

* Dissolved trace metals are not necessarily conservative upon mixing, since a large percentage of the reactive forms of some of those elements are adsorbed. The Negro River is about 2 pH unit more acid than the Solimões and has a very low suspended load. These differences lead to some desorption when the Solimões waters mix with the acidic waters of the Negro. A study of the mixing zone of these rivers already in progress shows for instance that 10 to 30 % of Cd and Cu will be complexed or adsorbed in this zone.

* Influence of the surrounding floodplain areas (called "varzea"): Following Richey et al., (1989), a large amount of the river water transit each year in the floodplain where anoxic conditions may occur. There is a direct exchange of suspended sediment between the varzea and the main river through the processes of entrainment and deposition. The deposition/resuspension cycle as well as the exchange rate between floodplain and mainstream channel may control at least partially the temporal variation of redox element concentrations such Mn and As. These elements show similar concentration in Solimões, Negro and Madeira rivers, and their variation in Obidos may be explained by the influence of remobilization processes occurring in the varzea.





Callède, J., Guyot, J.L., Molinier, M., Guimarães, V., Oliveira, E., and Filizola, N.P In: Sustainability of Water Resources under Increasing Uncertainty, edited by Rosbjerg, D., Boutayeb, N.E., Kundzewicz, Z.W., Gustard, A., and Rasmussen, P.F. IAHS. 240:163-172, 1997.

DNAEE. Contacto. 29:8-11, 1994

Elbaz-Poulichet F., Nagy G., Cserny T., Pomogyi A. 1997. Aquatic Chemistry, 6, 1-24.

Gaillardet J., Dupré B., Allègre C.J., Negrel P. 1997. Chemical Geology, 142, 141-173.

Konhauser K.O., Fyfe W.S., Kronberg B.I. Chemical Geology, 111, 115-175.

Molinier M., Guyot J.L., Oliveira E., Guimarães V. 1996. In L’hydrologie tropicale. IAHS Publ. 238, 209-222.

Richey J.E., Mertes L.A.K., Dunne T., Victoria R.L., Forsberg B.R., Tancredi A.C., Oliveira E. 1989. Global Biogeochemical cycles, 3(3), 191-204.

Seyler P. and Elbaz-Poulichet F. 1996, Journal of Hydrology, 180, 319-332.






Fig. 1: Map of the Amazon Basin with the situation of the Óbidos gauging station


Fig. 2: Monthly mean discharge decomposition, Amazon river at Obidos


Fig. 3 : Discharge, Suspended Sediment concentrations (SM), Conductivity
Amazon river at Obidos, 1995-1997








Fig. 4: Temporal patterns of some trace elements at the Óbidos Stations