Researchers from the Institute of Applied Ecology, Chinese Academy of Sciences (CAS) have developed a new sensor that enables rapid, in-situ monitoring of soil nitrate nitrogen (NO₃⁻-N), a key nutrient for crop growth.

The study was published in the journal ACS Sensors (American Chemical Society's journal).

Accurate measurement of soil NO₃⁻-N is essential for precision fertilization and stable crop yields. In agriculture, monitoring NO₃⁻-N in real time has long been a challenge. Conventional soil nutrient testing relies on laboratory-based chemical analysis, which is time-consuming. Existing nitrogen monitoring technologies also face limitations in in-situ deployment, temporal resolution, and continuous tracking; furthermore, they are prone to interference from soil moisture, salinity, and complex field conditions, making them inadequate for the high-frequency and large-scale monitoring demanded by modern precision agriculture.

To address these challenges, Associate Researcher Jian Gu’s team, in collaboration with Shenyang WITU Agricultural Science and Technology Co., Ltd., developed an in-situ NO₃⁻-N sensor based on Dielectric Spectroscopy Micro Domain Mediation Analysis and Component Response Algorithm Construction Theory (DIS M A R T). This approach enables the separation of overlapping signals from different soil components. The sensor operates through a dual-band frequency-splitting mechanism. In the low-frequency range of 1–50 MHz, stable impedance matching helps suppress interference from soil moisture and salinity. In the high-frequency range of 100–500 MHz, enhanced electromagnetic coupling allows the sensor to capture the characteristic relaxation signals of NO₃⁻-N with high sensitivity. According to the researchers, this coordinated use of two frequency bands enables more reliable identification of NO₃⁻-N under complex field conditions. Field tests were carried out across five soil types, including brown, black, red, saline-alkali, and loess soils. The results showed strong agreement between sensor readings and laboratory-measured values, with determination coefficient (R²) values of 0.943–0.987, indicating high accuracy.

The researchers also tested the sensor under typical agricultural scenarios, including before and after critical events: fertilization, precipitation events and autumn harvest activities. In these dynamic conditions, the monitoring results closely tracked actual changes in soil NO₃⁻-N, demonstrating the sensor’s suitability for continuous field use.

The sensor achieved millisecond-level response times, while maintaining low measurement errors. Even under extreme conditions such as subzero temperatures and high humidity, signal drift remained minimal, indicating stable performance. By supplying high-resolution, real-time data on soil NO₃⁻-N, the technology is expected to support more precise fertilizer application, and contribute to the development of intelligent and sustainable agricultural systems.

Figure 1. Sensor structure and principle of operation (Image by GU Jian).

Figure 2. Soil NO₃⁻-N concentrations detected by various sensors (SN) and measured spectrophotometrically (MS) before and after autumn harvest (Image by GU Jian).

Figure 3. Relationship between detected (PN) and measured (EN) NO3–-N concentrations (Image by GU Jian).