Leaching patterns of organic pollutants in agricultural fields

Organic pollutants, including pesticides and pharmaceuticals from irrigation with treated effluents, drift to the environment, risking habitats and organisms. Where nutrient behavior is predictable, organic pollutants include hundreds of molecules, and predicting their environmental fate is not obvious. Some research investigated tenth or hundreds of pesticides; however, the authors' discussion focused on the most frequently detected compounds, neglecting the latent information in the distribution of uncommon and low-concentration compounds.  Other research has focused on specific compound leaching or environmental fate. This research aims to develop fundamental principles for understanding and organizing knowledge related to the fate and transport of a large number of organic compounds in the environment, supported by field observations.

The current study was conducted in two fields on the Kishon Stream banks, a coastal stream in Israel. Field areas are characterized by heavy soils and high groundwater tables. A subsurface tile drainage system was installed to reduce water levels. This system provided easy access to the subsurface. Together with piezometers, it provided easy investigation of surface-subsurface-groundwater continuum interactions. 

Groundwater time series were collected before, during, and after the storm from the shallow piezometers (5 m). The time interval between samples was 2-3 days to closely track the pollutants leaching. Subsurface and surface water were collected during the storm. Visual classification of time series together with clustering methods could distinguish different leaching processes and governing factors involved. A linear fit was applied to obtain correlated processes and concentrations regarding all detected compounds in any two samples.

Groundwater time series displayed four patterns for most compounds. Very mobile or low-mobility compounds exhibited decreasing concentrations at the storm start. Low-intermediate mobility compounds and legacy pollutants exhibited a concentration rise. All compounds were diluted in the storm peak. Post-storm peak concentrations in the groundwater were correlated with the subsurface.

The linear fit of groundwater on the second day to the subsurface water was insignificant. However, the best fit was detected on the fifth day, after the storm peak, demonstrating the subsurface pollutants retardation after the storm peak. Dendrogram distinguished pre-storm samples and post-storm samples, relating post-storm concentrations to storm peak. We propose to attribute the pre-storm water to old soil water leaching resulting in low-intermediate pollutants leaching involved in adsorption-desorption processes in soil water. After the storm peak, we expect the leaching of pollutants washed from the upper soil layer; in a case, their mobility is significant enough. Very immobile compounds did not emerge after the storm, nor very mobile, which are not expected to occupy the soil column. The current study takes the advantage of many compounds to define patterns and rules that can explain the transport processes regarding the governing factors: mobility, environmental concentration, and timing in the storm.