Mobility of 2,4-Dichlorophenoxyacetic Acid, Glyphosate, and Metribuzine Herbicides in Terra Rossa-Amended Soil: Multiple Approaches with Experimental and Mathematical Modeling Studies

Abstract

2,4-Dichlorophenoxyl acetic acid (2,4-D), glyphosate, and metribuzine are used extensively as an herbicide in the vicinity of Antalya, Turkey. The objective of this study was to investigate the mobility of such herbicides in terra rossa-amended soil by multiple approaches with batch, miscible-displacement, and mathematical modeling studies. The results indicated that the sorption isotherms exhibited predominantly linear behavior over a wide range (0–100 mg/L) of selected herbicides. The average distribution coefficient (Kd) obtained from linear sorption isotherm was 0.95, 1.20, and 44.15 L/kg for 2,4-D, metribuzine, and glyphosate, respectively. After hydrogen peroxide treatment for eliminating organic carbon content in the soil, the average distribution coefficient (Kd) obtained from the linear sorption isotherm was 0.096, 0.118, and 4.37 L/kg for 2,4-D, metribuzine, and glyphosate, respectively. Miscible-displacement experiments also showed that the sorption of 2,4-D, metribuzine, and glyphosate was described by linear and rate-limited processes in the soil. Transport of metribuzine and 2,4-D was slightly but glyphosate was extensively delayed compared to non-reactive tracer experiments in the soil. Hydrogen peroxide treatment experiments also indicated that the organic carbon content was responsible for the non-ideal sorption behavior of such herbicides. The rapid desorption can be attributed to soft carbon fraction (humic/fluvic acid and lipids), whereas slower desorption can be responsible by hard carbon fraction (black carbon, kerogen) of soils that led to chemically non-ideal behavior. Two-site transport modeling analyses, including linear sorption and rate-limited sorption–desorption, provided a reasonably good match to the herbicide breakthrough curves indicating mathematical modeling results were generally consistent with the results obtained from the batch and column experiments. As a result, sorption of 2,4-D and metribuzine onto terra rossa-amended soil was low which could lead to contamination risk of karstic groundwater resources overlying the Antalya travertine plateau in the region. On the other hand, glyphosate exhibits much more sorption affinity in the soil and leaching risk is limited as with more delay. As a result, investigating the mobility of such herbicides in top soils overlying the karstic aquifer systems in Antalya would be very important to identify the contamination risk of groundwater resources. The results from the present study would also help in the design of effective herbicide management strategies in Antalya, Turkey.