ABSTRACT
Nitrogen nitrates play a significant role in the soil's nutrient cycle, and near-infrared spectroscopy can efficiently and accurately detect the content of nitrate-nitrogen in the soil. Accordingly, it can provide a scientific basis for soil improvement and agricultural productivity by deeply examining the cycle and transformation pattern of nutrients in the soil. To investigate the impact of drying temperature on NIR soil nitrogen detection, soil samples with different N concentrations were dried at temperatures of 50 °C, 65 °C, 80 °C, and 95 °C, respectively. Additionally, soil samples naturally air-dried at room temperature (25 °C) were used as a control group. Different drying times were modified based on the drying temperature to completely eliminate the impact of moisture. Following data collection with an NIR spectrometer, the best preprocessing method was chosen to handle the raw data. Based on the feature bands chosen by the RFFS, CARS, and SPA methods, two linear models, PLSR and SVM, and a nonlinear ANN model were then established for analysis and comparison. It was found that the drying temperature had a great effect on the detection of soil nitrogen by near-infrared spectroscopy. In the meantime, the SPA-ANN model simultaneously yielded the best and most stable accuracy, with Rc2 = 0.998, Rp2 = 0.989, RMSEC = 0.178 g/kg, and RMSEP = 0.257 g/kg. The results showed that NIR spectroscopy had the least effect and the highest accuracy in detecting nitrogen at 80 °C soil drying temperature. This work provides a theoretical foundation for agricultural production in the future.
ABSTRACT
Microplastic accumulation in the soil may have a negative impact on environmental security and human health. The lack of standardized method to identify and quantify microplastics in the soil is an obstacle to research. Existing techniques are time-consuming and cumbersome. We took the mixture of soil and low-density polyethylene (LDPE)∖polyvinyl chloride (PVC) as the research object to explore its spectral characteristics in the 0.6-1.8 THz (THz) band. We have proposed the new method to establish the Least Squares Support Vector Machine (LS-SVM) model using THz spectral data to quickly detect the microplastic pollution level in the soil from different regions. The local model is based on local soil training data set to predict local microplastic pollution, for LDPE, the average correlation coefficient (R) is 0.9833, and the average root mean square error (RMSE) is 0.0050, whereas for PVC, the average R is 0.9686, and the average RMSE is 0.0071. However, it seems to be useful only for local regions. The multisource model is that nine training sets are combined into one training set to simultaneously predict the degree of microplastic pollution in each area, for LDPE, the average R is 0.9895, and the average RMSE is 0.0007, for PVC, the average R is 0.9831, and the average RMSE is 0.0009. The results indicated that terahertz combined with LS-SVM model have a good effect on predicting the degree of soil microplastic pollution.