C. Grimaldi

Nitrate depletion during within-stream transport: effects of exchange processes between streamwater, the hyporheic and riparian zones.

In regions with intensive agriculture and shallow hydrological systems, headstreams are often polluted with nitrate even at the springs. In North-West France, nitrate concentration was seen to decrease downstream during baseflow conditions when the stream flows on granite, but this does not occur on schist. In order to explain this difference in behaviour, we analysed the groundwaters and surveyed the redox conditions (using a field test for ferrous iron) in near-bank wet meadows as well as in the hyporheic zone. We show that the wet meadow groundwater was denitrified and that oxygen and nitrate were presentaround the stream channel in a wide zone on granite,compared with a very restricted zone on schist. Ongranite, exchanges between the stream and the hyporheic zone are favoured by sandy or peaty material having high hydraulic conductivity. This gives rise to two processes (1) lateral inflow of denitrified water from wet meadows, (2) in the opposite direction, supply of stream nitrate to denitrification sites in the hyporheic zone. In the second case, a high hydraulic conductivity also reduces the water residence time and limits denitrification, resulting in high levels of oxygen and nitrate. On schist, the low hydraulic conductivity prevents an efficientconnection between surface and subsurface waters.

Distribution and dynamics of gibbsite and kaolinite in an oxisol of Serra do Mar, southeastern Brazil

Abstract In the Serra do Mar region, in southeastern Brazil, the soil mantle is mainly characterised by (i) a gibbsitic saprolite, (ii) various kaolinitic horizons within the gibbsitic material, (iii) kaolinito-gibbsitic topsoil horizons. This organisation does not match with the thermodynamic stability of gibbsite and kaolinite accompanying the solution percolation through soil profiles. A study of the micromorphological, mineralogical and chemical properties of the soil mantle reveals that this organisation arises from the in situ development of the soil from the crystalline bedrock. The bauxitic weathering of the bedrock, even if it is rich in quartz, can be explained by a fast renewal of the solutions and/or a high solubility of the kaolinite. Recycling of Si and Al by the forest can maintain a dynamic equilibrium of kaolinite in the topsoil horizons, as observed in Amazonia. The kaolinitic compact horizons evolve upslope at the expense of the gibbsitic material. At the contact between kaolinitic and gibbsitic material, dissolution patterns of quartz and gibbsite are observed, indicating that this evolution is in process. These observations and the organisation of the soil mantle set the problem of the apparent stability of gibbsite and kaolinite in this environment. Various assumptions that could explain this organisation of the soil mantle are discussed. Changes in the activity of water due to the pore size diminution and displacement of the gibbsite–kaolinite equilibrium appear insufficient to explain the stability of kaolinite. However, it could be allotted to the slow down of water flows in the soil mantel. Lastly, the eventual role of the complexing organic matter is presented. More investigations on the biogeochemical cycle of Si and Al and on the physico-chemical processes at the soil solution–mineral interface are necessary to explain the stability and dynamics of gibbsite and kaolinite in this environment.

Variations of plant and soil 87Sr/86Sr along the slope of a tropical inselberg.

Abstract• From the summit downslope a granitic inselberg in French Guiana, soils and vegetation evolve from bare granite covered by cyanobacteria, to a savannah-type vegetation on thin patchy sandy accumulations, then to a low forest on shallow young soils and to a high forest on deep highly weathered ultisols.• We have used element budgets and Sr isotopic variations in soils and plants to investigate the mineral nutrient supply sources of the different plant communities.• Granite and atmospheric deposition have 87Sr/86Sr ratios of 1.3 and 0.71, respectively. The 87Sr/86Sr ratio of cyanobacteria (0.72) suggests granite weathering by cyanobacteria crusts. The 87Sr/86Sr ratio of the savannah-type vegetation is 0.73 and varies between 0.75 and 0.76 in the low and high forest leaf litter regardless of soil depth, age and degree of impoverishment.• These almost constant ratio suggest that forest Sr comes from rainwater and from the summit of this inselberg, where it is released and redistributed along the slope, by surface flow, lateral redistribution of litter, and mineral particles. However, because of its very low content in the rock and soils, Ca is supplied to plants by atmospheric deposition.Résumé• Du sommet vers la base d’un inselberg granitique (Nouragues, Guyane Française), les sols et la végétation évoluent depuis des savanes sur des ilots sableux entre les affleurements rocheux couverts de cyanobactéries, vers une forêt basse sur sols peu épais, riches en minéraux altérables puis une forêt haute sur sols très profonds et altérés.• Les variations isotopiques du strontium des sols et des plantes ont été mesurées pour rechercher les sources de nutriments des différentes communautés végétales.• Les rapports 87Sr/86Sr du granite et des dépôts atmosphériques sont respectivement de 1,3 et 0,71. Le rapport 87Sr/86Sr des cyanobacteries (0,72) suggère une libération de Sr par altération du granite. Le rapport 87Sr/86Sr de la savanne est de 0,73 et varie entre 0,75 et 0,76 dans les litières de forêt basse et haute, quelle que soit la profondeur, et la richesse en minéraux altérables des sols.• La faiblesse et l’homogénéité surprenante de ces rapports suggèrent une alimentation en Sr des forêts essentiellement à partir de dépôts atmosphériques et des sols de la partie haute de l’inselberg, via des écoulements de surface, des redistributions latérales de litière et de particules minerales lors de crises érosives. Cependant, en raison de l’extrême pauvreté de la roche et des sols en calcium, le Ca des communautés végétales provient de la pluie.