A non-point source (NPS) pollution research group from Innovation Academy for Precision Measurement Science and Technology proposed a real-time measurement of total nitrogen (TN) for agricultural runoff based on multiparameter sensors and intelligent algorithms.
NPS pollution is a severe global threat to the water environment. Monitoring is the basis for effective control and supervision of agricultural NPS pollution. However, conventional monitoring methods are insufficient to achieve these goals owing to the minute-scale change in runoff flow and concentration under rainfall condition. More efficient methods for real-time monitoring of NPS pollution are needed.
The proposed new method combines the timeliness of sensor detection and the accuracy of intelligent algorithms, based on the physical and chemical relationships between TN and sensor-measured indexes including pH, electrical conductivity, oxidation–reduction potential, water temperature, dissolved oxygen, ammonia nitrogen, and nitrate nitrogen. In addition, extra tree regression (ETR) was selected as the optimal algorithm for TN inversion with regard to the coefficient of determination (R2), accuracy (Acc), mean relative error (MRE), tolerance test to missing values and etc.
The researchers have found that the new method can achieve the monitoring frequency at the minute scale (<5 min), and has good applicability (R2>0.9, Acc >85%) under similar environmental backgrounds (fields or ditches). Furthermore, the method accuracy is acceptable in the case of partial variable missing (the number of missing variables (n)≤2, or the proportion of missing-value samples (P)≤75%), which makes up for the flaws of missing or abnormal data caused by typical sensor malfunctions. Compared with methods such as laboratory test, remote sensing inversion, and water quality automatic monitoring station, the proposed method has high precision and real-time performance, and can operate stably in small and micro water bodies under rainy, cloudy, and nighttime conditions.
The new method shows potential in high-frequency monitoring of NPS pollution and may be able to provide important technological support for pre-warning and emergency control of NPS pollution. This study was published in Water Research on December 2021 (https://doi.org/10.1016/j.watres.2021.117992).
It was supported by the Hubei Provincial Natural Science Foundation of China, the Strategic Priority Research Program of the Chinese Academy of Sciences. the Hubei Technological Innovation Special Fund of China (grant number 2018ACA148), and the Youth Innovation Promotion Association of the Chinese Academy of Sciences.
Principle of the proposed real-time measurement method of total nitrogen (TN) for runoff
Framework and application scenario of the proposed real-time measurement method of total nitrogen (TN) for runoff