Mapping Soil Organic Matter Content Based on Feature Band Selection with ZY1-02D Hyperspectral Satellite Data in the Agricultural Region

Abstract

Soil organic matter (SOM) is an essential nutrient for crop growth and development. Hyperspectral satellite images with comprehensive spectral band coverage and high spectral resolution can be used to estimate and draw a spatial distribution map of SOM content in the region, which can provide a scientific management basis for precision agriculture. This study takes Xinzheng City, Henan Province’s agricultural area, as the research object. Based on ZY1-02D hyperspectral satellite image data, the first derivative of reflectance (FDR) was processed on the original reflectance (OR). The SOM characteristic spectral bands were extracted using the correlation coefficient (CC) and least absolute shrinkage and selection operator (Lasso) methods. The prediction model of SOM content was established by multiple linear regression (MLR), partial least squares regression (PLSR), and random forest (RF) algorithms. The results showed that: (1) FDR processing can enhance SOM spectral features and reduce noise; (2) the Lasso feature band extraction method can reduce the model’s input variables and raise the estimation precision; (3) the SOM content prediction ability of the RF model was significantly better than that of the MLR and PLSR models. The FDR-Lasso-RF model was the best SOM content prediction model, and the validation set R2 = 0.921, MAEV = 0.512 g/kg, RMSEV = 0.645 g/kg; (4) compared with laboratory hyperspectral data-SOM prediction methods, hyperspectral satellite data can achieve accurate, rapid, and large-scale SOM content prediction and mapping. This study provides an efficient, accurate, and feasible method for predicting and mapping SOM content in an agricultural region.

Spectral Heterogeneity Analysis and Soil Organic Matter Inversion across Differences in Soil Types and Organic Matter Content in Dryland Farmland in China

“…However, the evaluation of model performance, as measured using R 2 and RMSE, varies significantly across different studies because of variations in sample collection methods, spectral processing techniques, and model construction approaches. These variations can significantly impact the accuracy and reliability of spectral modeling [52,53]. The PLSR model can identify system information and noise and permits regression modeling provided that the number of sample points is less than the number of variables.…”

Application of a Bi-Directional Gated Recurrent Unit Combined with a Recurrent Neural Network Model Based on Fusion Attention Mechanism in Estimating Soil Salinity

“…Especially for spectral estimation of soil salt, changes in water content will affect the content and movement of soil salt, which will have a greater impact on the spectrum [35]. Therefore, it is necessary to research the spectral impact of soil moisture and its removal methods [36]. In addition, if the estimation model based on the outdoor air-dried soil spectrum is extended to the outdoor wet soil spectrum [37], the impact of soil moisture on the spectrum must also be eliminated [38].…”

Estimation of the Total Soil Nitrogen Based on a Differential Evolution Algorithm from ZY1-02D Hyperspectral Satellite Imagery

“…Studies have shown that the d-factor can be used to evaluate the uncertainty of an estimation model [19,48]. The degree of uncertainty of an estimation model is proportional to the calculated value of the d-factor; that is, the degree of uncertainty of the estimated model will increase with the increase in the calculated value of the d-factor, and the d-factor calculation formula is as follows:…”