Carine Saison

Hyperaccumulation of Metals by Thlaspi caerulescens as Affected by Root Development and Cd–Zn/Ca–Mg Interactions

ABSTRACT The aim of this work was to study, in a rhizobox experiment, the phytoextraction of metals by the hyperaccumulator plant Thlaspi caerulescens in relation to the heterogeneity of metal pollution. Six treatments were designed with soils containing various levels of metals. Homogeneous soils and inclusions of soils in other soil matrices were prepared in order to vary metal concentration and localization. Growth parameters of the plant (rosette diameter and shoot biomass) and localization of roots and shoot uptake of Zn, Cd, Ca, and Mg were determined after 10 weeks of growth. The plants grown on the polluted industrial soils provided a larger biomass and had lower mortality rates than those grown on the agricultural soil. Moreover, these plants accumulated more Zn and Cd (up to 17,516 and 375 mg kg− 1 DM, respectively) than plants grown on the agricultural soil (up to 7300 mg Zn kg− 1 and 83 mg Cd kg− 1 DM). The roots preferentially explored metal-contaminated areas. The exploration of polluted soil inclusions by the roots was associated with a higher extraction of metals. Zinc and Cd in the shoots of Thlaspi caerulescens were negatively correlated with Ca and Mg concentrations; however, the soil supply for these two elements was identical. This suggests that there is competition for the uptake of these elements and that Zn is preferentially accumulated.

Chlordecone fate and mineralisation in a tropical soil (andosol) microcosm under aerobic conditions.

Chlordecone is a persistent organochlorine insecticide that, even decades after its ban, poses a threat to the environment and human health. Nevertheless, its environmental fate in soils has scarcely been investigated, and elementary data on its degradation and behaviour in soil are lacking. The mineralisation and sorption of chlordecone and the formation of possible metabolites were evaluated in a tropical agricultural andosol. Soil microcosms with two different soil horizons (S-A and S-B) were incubated for 215 days with 14C-chlordecone. At five different times (1, 33, 88, 150 and 215 days) the extractability of 14C-chlordecone was analysed. Mineralisation was monitored using 14CO2 traps of NaOH. The appearance of metabolites was studied using thin layer and gas chromatography techniques. At the end of the experiment, the water soluble 14C-activity was 2% of the remaining 14C-chlordecone for S-A and 8% for S-B. Only 12% of the remaining activity was non extractable and more than 80% remained extractable with organic solvents. For the first time to our knowledge, a significant mineralisation of chlordecone was measured in a microcosm under aerobic conditions (4.9% for S-A and 3.2% for S-B of the initial 14C-activity). The drastically lower emission of 14CO2 in sterilised microcosms indicated the biological origin of chlordecone mineralisation in the non-sterilised microcosms. No metabolites could be detected in the soil extracts. The mineralisation rate of chlordecone decreased by one order of magnitude throughout the incubation period. Thus, the chlordecone content in the soil remained large. This study confirms the existence of chlordecone degrading organisms in a tropical andosol. The reasons why their activity is restricted should be elucidated to allow the development of bioremediation approaches. Possible reasons are a heterogeneous distribution a chlordecone between sub-compartments with different microbial activities or a degradation of chlordecone by co-metabolic processes controlled by a limited supply of nutrients.

Effect of spatial heterogeneities of water fluxes and application pattern on cadusafos fate on banana-cultivated andosols.

In tropical humid environments under intensive banana production, pesticide transfer in waters can be of particular concern due to heavy rainfall, steep slopes, and soils with high infiltration capacities. The transfer in percolation and runoff waters of the nematicide cadusafos was investigated during a three month field experiment. The spatial heterogeneity of the banana plantation was taken into account by measuring percolation fluxes both under the banana plants and in the interrows with a specially designed lysimeter device installed at 60 cm depth. At the field scale, 0.34% of the pesticide applied was transferred in percolation, 0.13% in runoff. Forty-nine percent of cadusafos losses occurred by percolation under the banana plants, 23% by interrow percolation, and 28% by runoff. Losses were highest during the three weeks following cadusafos application, and this is also when dissipation in the soil was highest (calculated half-life in the soil: 7d). After this period, losses of cadusafos were low, both in soil and waters. Under the banana plant, saturated fluxes carried most of the pesticide, despite total percolation fluxes being at least five-times higher than saturated ones. Although overall pesticide transfer in water was low (0.5% of applied), it was not negligible due to the frequency of pesticide application in these areas.

Effects of thiobencarb in combinations with molinate and chlorpyrifos on selected soil microbial processes

The impact of pesticides, namely thiobencarb (TBC), molinate (MOL) and chlorpyrifos (CPF), on soil microbial processes was studied in two Australian soils. Substrate induced respiration (SIR), substrate induced nitrification (SIN) and phosphatases and chitinase enzymatic activities were assessed during a 30-day microcosm study. The pesticides were applied to soils at recommended rates either alone, or as binary mixtures with TBC. Soil samples were sampled at 5, 15 and 30 days after pesticide treatments. Substrate induced respiration was only transiently affected by pesticides in both soils. In contrast, the process of indigenous nitrification was affected by the presence of pesticides in both soils, especially when the pesticides were applied as binary mixtures. Substrate induced nitrification increased with pesticides in the Griffith soil (except with MOL+TBC after 5 days) whereas SIN values were non-significantly different to the control on the Coleambally soil. The binary mixtures of pesticides with TBC resulted in a decrease in SIN in both soils, but the effects disappeared within 30 days. The enzymatic activities were not consistently affected by pesticides, and varied with the soil and pesticides studied. This study showed that, when applied at recommended application rates, TBC, MOL, and CPF (individually or as binary mixtures), had little or only transitory effects on the functional endpoints studied. However, further investigations are needed to assess the effect on microbial densities and community structure despite the low disturbance to the functions noted in this work.