Abstract: Intermittent irrigation is one common practice of water management in paddy field of hilly areas, and the intense alternating wet and dry process may have a significant effect on the environmental behavior of various pollutants. Based on the laboratory batch equilibrium adsorption experiment, isotherms of the insecticide chlorpyrifos and its major degradation product 3,5,6-trichloro-2-pyridinol (TCP) in the cultivated layer (0-20 cm) and underlying uncultivated layers (20-50 cm) of the paddy fields of calcareous purple soil were obtained and fitted by the linear and Freundlich models. During the rice-growing season, field application of chlorpyrifos in paddy soils and continuous on-site monitoring were carried out for the observation of temporal changes in the concentration of both chlorpyrifos and TCP in the floodwater and soil pore water at different depths. The results obtained under the conditions of intermittent irrigation and continuous flooding were compared. The results showed that the adsorption isothermal data for both chlorpyrifos and TCP fitted well both models. The calculated values of Freundlich sorption capacity and linear distribution coefficient for chlorpyrifos and TCP were found higher for the cultivated soil layer than those for the uncultivated layers. For all the tested soil samples, chlorpyrifos had much higher values of Freundlich sorption capacity (in the range of 34 to 170) and linear distribution coefficient (in the range of 44 to 171) than those for TCP, which had Freundlich sorption capacity and linear distribution coefficient in the range of 0.62 to 0.67 and 0.47 to 0.78, respectively. This indicated that chlorpyrifos could be easily adsorbed to the soil and maintained in the surface soil, while TCP could easily migrate and disperse in the environment. Following the pesticide application, concentrations of chlorpyrifos and TCP in the floodwater decreased rapidly with time and reached to stable low levels (i.e., 10% of initial concentrations) within the first 3 and 6 days, respectively. Chlorpyrifos was mostly adsorbed in the cultivated soil layer, while TCP was mainly found in the aqueous phase of both cultivated and uncultivated layers. Both chlorpyrifos and TCP could reach the depth of 50 cm with infiltrating water via various soil macropores (e.g., cracks, worm burrows, and root channels). Irrigation method had shown an effect on their concentrations in soil pore water, with concentrations generally found lower under intermittent irrigation as compared to continuous flooding. Apparently, water movement in paddy field exerted a greater impact on the TCP concentration in soil pore water of the cultivated soil layer under intermittent irrigation. After draining out the floodwater, the TCP concentration of soil pore water at the 10 cm depth decreased rapidly and remained stable after re-irrigation; in the contrast, TCP increased steadily during the first 3 weeks following chlorpyrifos application under continuous flooding condition. In addition, rain events during the floodwater draining periods and irrigation events had resulted in the transient releases of both chlorpyrifos and TCP from the soil solid phase to the aqueous phase, followed by subsequent leaching at elevated concentrations. Such effect was found more apparent for TCP, which had a lower sorptivity than its parent compound chlorpyrifos. The marked decreases in the concentrations of chlorpyrifos and TCP with time in soil pore water at the depth of 10 cm during the floodwater draining periods may be attributed to the enhanced degradation of both compounds under the improved oxidative conditions. The results above have suggested that, in future research, due attention should be paid to the environmental behavior of TCP in paddy fields at lowlands of hilly areas that may pose a risk to the shallow groundwater used as drinking water source for surrounding rural residents.