Desorption of pesticides (fenobucarb, endosulfan, and dichlorodiphenyltrichloroethane (DDT)) from soil to aqueous solution with the simultaneous presence of dissolved organic carbon (DOC), sodium dodecyl sulfate (SDS), and sodium oxalate (Oxa) was investigated in batch test by applying a full factorial design and the Box-Behnken response surface methodology (RSM). Five concentration levels of DOC (8 to 92 mg L(-1)), SDS (0 to 6.4 critical micelle concentration (CMC)), and Oxa (0 to 0.15 M) were used for the experiments with a rice field topsoil. The results of RSM analysis and analysis of variance (ANOVA) have shown that the experimental data could be well described by quadratic regression equations with determination coefficients (R (2)) of 0.990, 0.976, and 0.984 for desorption of fenobucarb, endosulfan, and DDT, respectively. The individual effects and interaction of DOC, SDS, and Oxa were evaluated through quadratic regression equations. When the aqueous solution includes 50 mg L(-1) DOC, 3.75 CMC SDS, and 0.1 M Oxa, the maximum desorption concentrations of fenobucarb, endosulfan, and DDT were 96, 80, and 75 μg L(-1), respectively. The lowest concentration of SDS, DOC, and Oxa caused the minimum desorption. This point at conditions of concern for flooding water is high content of organic compounds causing potentially high contamination by desorption, and the remarkably lower desorption at organic matter-free conditions. The suspended organic matter is one of the common characteristics of flooding and irrigation water in rice fields, and surfactants from pollution increase the problem with desorption of legacy pesticides in the rice fields.