Atrazine is one of the most used herbicides in China. It is a persistent organic pollutant but has been widely used on Chinese farmlands for a long time. To assess its dietary and ecological risks to human and environment, in this study, atrazine residues were extracted with acetonitrile and then plant samples were detected with gas chromatography coupled with mass spectrometry (GC-MS) and soil samples were determined with gas chromatography coupled with nitrogen-phosphorus detector (GC-NPD). The limit of quantification (LOQ) of the method was 0.01 mg/kg for all matrices. The recoveries ranged from 82.0 to 105.4% for plant samples and 75.6 to 85.6% for soil samples. The final residues of atrazine in all plant samples were lower than LOQ. Dietary risk assessment suggested that under good agricultural practices (GAP) conditions, intake of atrazine from apples, grapes, and tea would exhibit an acceptably low health risk on consumers. However, the final residues of atrazine in soil samples were <0.01–9.2 mg/kg, and the half-lives were 2.0–9.1 days. Based on the species sensitivity distribution (SSD) model, the potential affected fraction (PAF) of atrazine in soil samples ranges from 0.01 to 65.8%. Atrazine residues in 43.1% soil samples were higher than 0.11 mg/kg, which was the hazardous concentration for 5% of species (HC₅) of atrazine in soil. These results suggested that the ecological risks of atrazine in apples, grapes, and tea garden soil would exhibit a high risk on environmental species even under the same GAP conditions. This study could provide guidance for comprehensive risk assessment of atrazine properly used in apple, grape, and tea gardens.