Leaf wetness duration is one of the important input variables of plant disease model, which is related to the infection of many leaf pathogens and affects the infection and development rate of pathogens. The leaf wetness sensor can realize real-time and automated monitoring, and since the leaf wetness duration is affected by the interaction between the environment and plants, it needs to be calibrated in citrus orchards under irrigation. Citrus in growing season was used as experimental material to study the calibration method.The angle of the leaf wetness sensor was 30°, and two methods were used to determine the dry-wet threshold of the sensor: drip water to the sensor by pipetting gun and sprinkle irrigation facility to the sensor.The monitoring effects of sensors in different positions of the citrus canopy were compared, and the effects of rain and no rain conditions on the monitoring effects of the sensors were studied. Finally, the neural network model was used to verify the rationality of the threshold.The results showed that the leaf wetness sensor obtained a dry-wet threshold of 270mV in the irrigation environment. At this time, the monitoring effect of the sensor was the best, and the error was within 2h. By comparing with the prediction results of the neural network model, it was confirmed that the monitoring effect of the sensor was good under this threshold.The sensor located at the bottom of the citrus canopy had the highest monitoring accuracy, which can reach 0.95.The monitoring effect of the sensor was better in no rain condition than that in rainy condition.Overall, the calibration method of the leaf wetness sensor can be used to monitor the leaf wetness duration of irrigated citrus orchards, which met the requirements of the citrus disease early warning system.