Abstract

In order to accurately acquire soil total nitrogen (STN) content and soil moisture (SM) content, an in-situ STN-SM content detector was developed, and the stability and accuracy of the detector were tested. The in-situ detector consisted of an optical unit, a control unit and a mechanical unit. The optical unit comprised eight single band Near-infrared (NIR) laser sources at 1260, 1330, 1360, 1430, 1530, 1580, 1660 and 1450 nm, collimator and an InGaAs photoelectric sensor, a laser source holder with “Eight faces” umbrella structure and sapphire glass. The control unit was composed of hardware part and software part. The hardware part included I-U conversion circuit, amplifier and filter circuit, and JN5139 module responsible for A/D conversion and data transmission. The software part of the program was written in the JN5139 microcontroller unit. The mechanical unit comprised a cylindrical detection darkroom, an electrical component mounting tray and a cylindrical shading and dust cover. The stability test results showed that the repeatability error at eight sensitive bands were 0.7758%, 0.8091%, 0.7958%, 0.5189%, 0.6405%, 0.8556%, 0.7391% and 0.5294%, respectively. The correlation coefficients between the absorbance of in-situ detector and MATRIX-I Fourier transform near-infrared spectrophotometer were all above 0.88. A portable LED STN content detector was taken as the reference in the performance test. The results showed that the calibration coefficient Rc2 and the validation coefficient Rv2of in-situ detector model were 0.8322 and 0.7832, respectively. While those of the portable LED STN content detector model were 0.7205 and 0.6851, respectively. Two regression models were established for the SM content by respectively using the portable LED STN content detector and the in-situ detector, and the results showed that both detectors had good detection effects in SM. The experimental results showed that compared with the portable LED STN content detector, in-situ detector greatly improved the STN content detection precision, and could realize STN and SM content field detection in real-time.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call