Dissipation and leaching of pesticides in planted and unplanted soil mesocosms: insights from compound-specific isotope analysis (CSIA)

In contaminated agroecosystems, the rhizosphere at the soil/plant interface may act as a hotspot of pesticide transformation. However, results based on pesticide concentrations only may not be conclusive because degradative and non-degradative (e.g. sorption) processes simultaneously dissipate pesticides. Compound-specific stable isotope analysis (CSIA) allows to identify degradation processes by analyzing changes of natural stable isotope abundances of isotopic elements (e.g., ¹³C, ¹⁵N) during (bio)degradation of pesticides.

The purpose of this study was to evaluate and compare the contribution of processes responsible for pesticide dissipation in planted and unplanted soil mesocosms using CSIA. Five widely used aniline herbicides and fungicides (i.e., acetochlor, alachlor, S-metolachlor, metalaxyl and butachlor) were spiked at 25 µg g^-1 in planted (with Lolilum multiflorum sp.) and unplanted mesocosms with contrasted forest and vineyard soils. The mesocosms were incubated in a climate chamber for 75 days under controlled humidity, light and temperature. Three successive rainfalls (15 mm) were applied at on day 45, 60 and 75 days to collect soil leachate. Bulk soil and rhizosphere samples were collected at the end of the experiment. Pesticides were extracted from leachate and soil for quantification using GC-MS and pesticide CSIA (δ¹³C and δ¹⁵N) using GC-IRMS. Pesticide uptake by plants was negligible.

Pesticides concentrations in leachate ranged from 0.9 (acetochlor in planted mesocosms) up to 7.1 mg L^-1 (metalaxyl in unplanted vineyard soil). Pesticide concentrations were higher in unplanted microcosms, in particular in the vineyard soil microcosms. This suggests that plants and higher organic content in forest soil limited pesticide leachate. In total, 3, 3, 13, 8 and 4% of acetochlor, alachlor, metalaxyl, S-metolachlor and butachlor were exported in leaching, respectively. In addition, the first rainfall accounted for 70 to 99% of total leaching losses, indicating large export in both soil types during the first event following application. CSIA of pesticides in leachate confirmed that in situ degradation occurred in both planted and unplanted microcosms, although degradation extent may be larger in planted than in unplanted mesocosms. Isotope fractionation in alachlor, acetochlor, S-metolachlor and butachlor was more pronounced in planted mesocosms for both soils (Δδ¹³C up to 9.5‰), while it was only observed in planted forest microcosms for metalaxyl (Δδ¹³C = 5.9‰). This suggests that soil physico-chemical properties and/or the rhizosphere (i.e., root exudates and uptake) influenced pesticide degradation.

Up to 6, 8, 29, 20 and 5% of acetochlor, alachlor, metalaxyl, S-metolachlor and butachlor, respectively, persisted in soil at the end of the experiment. This corresponds to pesticide dissipation ranging from 58 to 91% due to both sorption and degradation in soil. Accordingly, the isotopic fractionation in acetochlor and alachlor in soil and rhizosphere was only observed after 75 days in both soils (Δδ¹³C = 8.2 and 3.2‰, respectively), and remained low or insignificant for the other pesticides. Our results emphasize variability in the extent of pesticide dissipation in soils, and CSIA of pesticides indicates that pesticide degradation mainly occurs in the soil solution leached during rainfall events rather than in bulk soil or the rhizosphere.