Construction and Evaluation of Prediction Model of Main Soil Nutrients Based on Spectral Information

Abstract

The rapid and accurate detection of soil nutrient content through spectral technology is one of the requisite technologies for precision fertilization, which, however, is an unsolved issue. In order to achieve this purpose, a more robust and accurate model is established in this study. The regression algorithm is integrated with effective wavelength selection to construct the prediction model for total nitrogen, available phosphorus, and available potassium (N, P, and K), which removes the need for complex pretreatment and algorithm constraints. According to the research results, with regard to the prediction of soil nitrogen, phosphorus, and potassium contents, the joint interval support vector regression (Si-SVR) model performed best in modeling, with the root mean square error of prediction (RMSEP) limited to 0.0231%, 1.0554%, and 3.4225%, respectively. In addition, the relative percentage deviation (RPD) values were restricted to 2.68, 2.12, and 2.37, respectively. As indicated by the prediction results obtained for the above three nutrient contents, the RPD values of the Si-SVR model prediction accuracy evaluation indicators exceeded 2.0, which evidences a high level of prediction accuracy. This method makes it possible for spectral data to be applied in practical production, and these results provide a valuable reference for the effective detection of major soil nutrients.

Hyperspectral Inversion of Soil Carbon and Nutrient Contents in the Yellow River Delta Wetland

“…The principles of reflectance spectroscopy in soil science are related to the variability of material surfaces and their optically active components [15]. For example, soil elements, such as carbon, nitrogen, and phosphorus, have a significant impact on the form and nature of soil reflectance spectra and can be estimated quickly [16][17][18][19]. Recently, the use of hyperspectral techniques to obtain information on soil elemental content has gained popularity and become a reliable method for exploring soil-related issues [20][21][22].…”

Spatial Prediction of Total Nitrogen in Soil Surface Layer Based on Machine Learning

“…Presently, most of the models used for spatial prediction of TN are divided into two categories. The first type is linear models, which are constructed by simulating the linear relationship between the reflectance of remote sensing image bands and the TN content, and thus inverse models, including partial least squares regression (PLSR) [8][9][10], multiple linear regression (MLR) [11,12], and other models. However, due to the multiple and complex relationships between the reflectance of multispectral image bands and soil nutrient content, the constructed linear models are not sufficient to reflect the spatial distribution of nutrients well and are lacking in prediction accuracy.…”