Purpose <p>Soil spectroscopic technology has proven to be a real-time alternative for estimating soil organic carbon (SOC); however, soil water often affects spectra and limits the accuracy of SOC estimation. This study aimed to decrease the effect of soil water on SOC spectroscopic modeling.</p> Materials and methods <p>A separate correction dewatering machine (SCDM) method was used to estimate SOC. This method can effectively mitigate the effect of soil water by performing correction processes for each spectral band individually. The soil spectra and soil water contents used for conducting this study were obtained using an ASD FieldSpec 3 spectroradiometer.</p> Results and discussion <p>After a series of correction processes, the optimal SCDM model (root mean square error (RMSE), 2.26&#xa0;g/kg; ratio of performance to deviation (RPD), 2.51; ratio of performance to interquartile range (RPIQ), 3.90) was acquired to determine the optimal model, which provided better modeling performance than a multiple linear regression (MLR) model.</p> Conclusions <p>The SCDM method is a valuable reference for dewatering and represents a currently available methodology for estimating soil properties spectroscopically.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Separate correction dewatering machine for the spectroscopic determination of organic carbon in wet soil

  • Lixin Lin,
  • Xixi Liu,
  • Yuan Sun,
  • Tao Xie

摘要

Purpose

Soil spectroscopic technology has proven to be a real-time alternative for estimating soil organic carbon (SOC); however, soil water often affects spectra and limits the accuracy of SOC estimation. This study aimed to decrease the effect of soil water on SOC spectroscopic modeling.

Materials and methods

A separate correction dewatering machine (SCDM) method was used to estimate SOC. This method can effectively mitigate the effect of soil water by performing correction processes for each spectral band individually. The soil spectra and soil water contents used for conducting this study were obtained using an ASD FieldSpec 3 spectroradiometer.

Results and discussion

After a series of correction processes, the optimal SCDM model (root mean square error (RMSE), 2.26 g/kg; ratio of performance to deviation (RPD), 2.51; ratio of performance to interquartile range (RPIQ), 3.90) was acquired to determine the optimal model, which provided better modeling performance than a multiple linear regression (MLR) model.

Conclusions

The SCDM method is a valuable reference for dewatering and represents a currently available methodology for estimating soil properties spectroscopically.