Spectral behavior modeling of soil texture over dust center of Khuzestan province using Hyperspectral images and Random Forest model

Authors

1 Department of Soil Science and Engineering, Faculty of Agriculture, Shahid Chamran University of Ahvaz, Ahvaz, Iran

2 Soil Conservation and Watershed Management Research Institute, Tehran, Iran

3 Department of Soil Science, Faculty of Agriculture, Tarbiat Modares University, Tehran, Iran

Abstract

Soil texture had an essential role in soil resistance of wind erosion. Due to useful Features of Hyperspectral images including low cost and high speed to the evaluation of soil properties, it can be used to determination of soil texture. The aim of this study was to evaluate the spectral behavior of clay, sand and silt content in susceptible soils of Khuzestan province implementing the PLS-RF model. Initially, the main factors were determined, using partial least squares regression (PLSR). Accordingly, the Random Forest model was implemented on main factors. Subsequently, the comparing of performance related to main spectral and five preprocessing methods, i.e. the Savitzky-Golay filter, the first derivative with the Savitzky-Golay filter (FD-SG), the second derivative with the Savitzky-Golay filter (SD-SG), the normalization of standard method (SNV), and the continuum removed method (CR), applied to eliminate the noise and also to increase the accuracy of the PLS-RF model. The Results showed that the best method with high performance was the continuum removal for clay content (, PRDcal=1.93) and silt content, (PRDcal=1.65). Additionally, the second derivative method was the best performance to the evaluation of sand content (PRDcal=1.98). In this study, the key wavelength for clay content was determined in the range of 1200-1210, 1800 and 2200 nm and the results for sand was 1400-1450, 1910-1930, 2200 and 2220 nm, and finally for silt was 1320, 1615, 2200 nm.

Keywords

References

  Hillel, D., 1980. Applications of soil physics. Academic Press, San Diego.
  Ben-Dor, E., Chabrillat, S., Demattê, J., Taylor, G., Hill, J., Whiting, M., Sommer, S., 2009. Using imaging spectroscopy to study soil properties. Remote Sensing of Environment 113, 38-55.
  Curcio, D., Ciraolo, G., D’Asaro, F., Minacapilli, M., 2013. Prediction of soil texture distributions using VNIR-SWIR reflectance spectroscopy. Procedia Environmental Sciences 19, 494-503.
  Mohamed, E., Saleh, A., Belal, A., Gad, A.A., 2017. Application of near-infrared reflectance for quantitative assessment of soil properties. The Egyptian Journal of Remote Sensing and Space Science 21, 1-14.
  Xuemei, L., Jianshe, L., 2013. Measurement of soil properties using visible and short wave-near infrared spectroscopy and multivariate calibration. Measurement 46(10), 3808-3814.
  Silva, E.B., Ten Caten, A., Dalmolin, R.S.D., Dotto, A.C., Silva, W.C., Giasson, E., 2016. Estimating Soil Texture from a Limited Region of the Visible/Near-Infrared Spectrum. Digital Soil Morphometrics, Springer, pp. 73-87.
  Wang, J., Ding, J., Abulimiti, A., Cai, L., 2018. Quantitative estimation of soil salinity by means of different modeling methods and visible-near infrared (VIS–NIR) spectroscopy, Ebinur Lake Wetland, Northwest China. Peerj 6, 4703.
 
Advanced Applied Geology
Volume 9, Issue 4
January 2020
Pages 466-479

Files

Share

How to cite

Statistics