Characterizing Soil Profile Salinization in Cotton Fields Using Landsat 8 Time-Series Data in Southern Xinjiang,...

Area of Science:

• Agricultural Science
• Remote Sensing
• Environmental Science

Background:

• Soil salinization severely impacts agricultural productivity and food security in arid regions.
• Accurate soil profile salt data are essential for effective irrigation management.

Purpose of the Study:

• To assess the feasibility of using Landsat 8 time-series data for monitoring soil salinization.
• To develop remote sensing models for estimating soil conductivity at various depths.

Main Methods:

• Acquired 172 Landsat 8 images (2013-2019) from Xinjiang, China.
• Synthesized multiyear extreme vegetation index values and collected 540 soil conductivity samples.
• Developed three Cubist models to monitor soil conductivity at depths of 0-0.375m, 0-0.75m, and 0-1.00m.

Main Results:

• The Cubist models demonstrated high prediction accuracy for soil profile electrical conductivity.
• Determination coefficients for the prediction set were 0.80 (0-0.375m), 0.74 (0-0.75m), and 0.72 (0-1.00m).
• Digital mapping of soil salinization was successfully performed.

Conclusions:

• It is feasible to use multiyear extreme vegetation indices combined with Cubist modeling for regional soil profile salinization monitoring.
• This approach offers a viable method for managing land productivity and food security in arid areas affected by soil salinity.