[Hyperspectral Detection Model for Soil Dispersion in Zhouqu Debris Flow Source Region]

Area of Science:

• Geoscience
• Remote Sensing Technology
• Soil Science

Context:

• Soil dispersion is a critical issue affecting soil structure and agricultural productivity.
• Hyperspectral remote sensing offers a non-destructive method for analyzing soil properties.
• Understanding the spectral signatures of dispersive soils is essential for effective management.

Purpose:

• To investigate sensitive spectral bands for detecting soil dispersion using hyperspectral remote sensing.
• To develop and validate a soil dispersive hyperspectral remote sensing model.
• To identify the primary soil minerals and elements contributing to soil dispersion.

Summary:

• Fourier transformation effectively separated signals from noise, enabling the development of a mineral identification system for accurate spectral data.
• A multiple linear regression model demonstrated a high correlation between sensitive bands and soil dispersion, proving effective for forecasting.
• Analysis of mineral spectra revealed that sodium, calcite, and illite are strongly correlated with soil dispersion, with sodium being the most significant factor.

Impact:

• Provides a novel hyperspectral remote sensing approach for assessing soil dispersion.
• Identifies key mineralogical indicators (sodium, calcite, illite) responsible for soil dispersion.
• Enables accurate soil dispersion forecasting, aiding in land management and agricultural planning.