Mitigation of Two Insecticides by Wetland Plants: Feasibility Study for the Treatment of Agricultural Runoff in Suriname (South America)

Abstract

In agricultural areas, pesticides can enter receiving waterbodies by means of agricultural runoff and pollute these systems. Constructed wetlands are capable of removing several pollutants including pesticides. Unfortunately, few studies are from South America, and therefore, information is urgently needed on pesticide mitigation in agricultural runoff by native plants. To this aim, an experimental setup of polypropylene tubs was used, which were planted with two types of native plants (Nymphaea amazonum and Eleocharis mutata). Mesocosms were exposed to low (10 μg/l) and high (30 μg/l) target concentrations of lambda-cyhalothrin, while for imidacloprid, a low (60 μg/l), high (180 μg/l), and an extra high (1,000 μg/l) dose, were applied using batch experiments of 2 weeks each. Removal efficiencies for lambda-cyhalothrin from the water phase showed 100 % removal at 72 h for both low and high target concentrations for N. amazonum mesocosms, while for E. mutata mesocosms, a 100 % removal was observed at 48 h for mesocosms exposed to low target concentrations and for high target concentrations at 72 h. For imidacloprid, a 100 % removal was observed for E. mutata and 86 % for N. amazonum mesocosms exposed to low target concentrations (60 μg/l) at 216 h. For the highest dose (1,000 μg/l), the removal efficiency was on average 72 % at 216 h for both types of mesocosms. Statistical two-way ANOVA analysis (α = 0.05) showed that the removal of lambda-cyhalothrin was independent of the dose applied and the plant type, while for imidacloprid, removal was dependent on the dose applied and independent of the plant type. After the experimental period, analyses of the plants and sediment showed that 48.5 % of the applied amount of lambda-cyhalothrin was detected in the sediment and 0.4 % in plant material (shoots and leaves), while the amount in roots was below the limit of detection for N. amazonum mesocosms. For E. mutata mesocosms, 44.6 % of lambda-cyhalothrin was detected in sediment and 0.5 % in roots. For N. amazonum mesocosms, 78.9 % of the applied amount of imidacloprid was retained in plants (plant material and roots) and 17.31 % in sediment, while for E. mutata mesocosms only 0.5 % of imidacloprid was detected in plant material and roots. In this experiment, the DT50 of lambda-cyhalothrin in the water phase of both types of mesocosms was on average 1 day, while for imidacloprid, this was calculated to be around 1–10 days. The results obtained provide necessary information for the construction of a field scale wetland capable of efficient removal of pesticides in agricultural runoff.