Long-term relationships among pesticide applications, mobility, and soil erosion in a vineyard watershed

Abstract

Agricultural pesticide use has increased worldwide during the last several decades, but the long-term fate, storage, and transfer dynamics of pesticides in a changing environment are poorly understood. Many pesticides have been progressively banned, but in numerous cases, these molecules are stable and may persist in soils, sediments, and ice. Many studies have addressed the question of their possible remobilization as a result of global change. In this article, we present a retro-observation approach based on lake sediment records to monitor micropollutants and to evaluate the long-term succession and diffuse transfer of herbicides, fungicides, and insecticide treatments in a vineyard catchment in France. The sediment allows for a reliable reconstruction of past pesticide use through time, validated by the historical introduction, use, and banning of these organic and inorganic pesticides in local vineyards. Our results also revealed how changes in these practices affect storage conditions and, consequently, the pesticides' transfer dynamics. For example, the use of postemergence herbicides (glyphosate), which induce an increase in soil erosion, led to a release of a banned remnant pesticide (dichlorodiphenyltrichloroethane, DDT), which had been previously stored in vineyard soil, back into the environment. Management strategies of ecotoxicological risk would be well served by recognition of the diversity of compounds stored in various environmental sinks, such as agriculture soil, and their capability to become sources when environmental conditions change.

Keywords: DDT; glyphosate; lake sediment; pesticides; soil erosion.

Fig. 1.

The Lake Saint André watershed and the vineyards in 2009 (interpreted from aerial photographs), as well as the bathymetric map with the location of core SAN11P2 retrieved from the deeper part of the lake.

Fig. 2.

Data from core SANP2. From left to right: (A) photograph and grain size contour plot with two primary populations centered at 0.3 and 14 µm, (B) LOI at 550 °C (organic matter) and 950 °C (carbonates), (C) ²¹⁰Pb_ex activity, (D) ¹³⁷Cs activity, and (E) the age model.

Fig. 3.

Chronological variations in pesticide fluxes. (A) Herbicides: AMPA, deisopropyl atrazine, and diflufenican; (B) fungicides: Bordeaux mixture (Cu/Rb), captan, dimethomorph, mancozeb (Zn/Rb), and pyrimethanil; (C) insecticides: dicofol, bromopropylate, and bifenthrin; (D) DDT and metabolites: DDT, DDE/DDT, DDD/DDE, and ƩDDT. The gray area in the AMPA profile denotes the lower detection limit for this compound.

Fig. 4.

Chronological variation in levels of AMPA, deisopropyl atrazine, subaerial flux, and sum of DDT and DDE/DDT. The horizontal dotted lines denote the two primary changes in the sedimentation rate.