Rice fertilizer application is often disturbed by the moisture or mud and water in the field, particularly, in the rainy and humid southern region of China. The complicated working environment can result in fertilizer clogging or leakage in the fertilizer application machinery. The photoelectric and capacitive sensors are often developed to monitor the fertilizer application blockage at present. However, the traditional sensors are susceptible to fertilizer dust during operation, resulting in reduced sensitivity and even monitoring failure, which cannot meet the actual production needs. In this study, an airflow anti-clogging and monitoring system was proposed for fertilizer application in the rice direct seeding machine. The agronomic requirements were combined with the fertilizer application in the direct seeding of southern rice. The smoothness of the fertilizer application pipeline was effectively improved to prevent fertilizer clogging. Once the fertilizer clogging occurred, the alarm alerts were run in the system simultaneously. Firstly, the working principle of the fertilizer application system was determined to design the key components, such as the airflow diversion duct and the airflow change feedback device. At the same time, the structural parameters were optimized for the key components of devices. A parameter calibration test was carried out to determine the air pressure value (P1) of less than 1 000 Pa at the inlet of the airflow diversion duct. Secondly, a coupled simulation was carried out using Fluent and Rocky software. An experiment was then conducted to calculate the motion state of fertilizer particles with or without applied airflow in the airflow diversion duct of the key component. The test results showed that the velocity of fertilizer granules increased significantly under the action of airflow. The kinetic energy of the fertilizer granules increased by about 39.7%, compared with no airflow. It infers that the airflow effectively improved the smoothness of the fertilizer application pipeline. Finally, the prototype test was performed on the blockage monitoring and alarm performance of the device. The air pressure pre-test demonstrated that the air pressure of the air inlet should be taken as 600-800 Pa in the static test of the prototype. A combination of inlet air pressure, pipe inner diameter, and device fertilization rate was then tested to determine the optimal operating parameters of the device. The test protocol was designed by the Box-Behnken response surface method. After that, the regression model and the variation of the response surface were used to determine the inner diameter of the airflow diversion pipe of 28 mm, the air pressure value of 700 Pa at the air inlet, and the fertilizer application rate of 20 g/s, as the better working parameters of the device. The optimal combination of parameters was selected for the test in the field trials. Consequently, the accuracy of alarm monitoring was above 90.0%, and the highest accuracy was 96.7%. The better performance of the device was achieved in the stability, which fully met the requirement of clogging monitoring accuracy of the rice fertilizer device. The finding can provide an important reference for fertilizer application in the anti-blocking and blockage monitoring for rice production.