Predicting the spatial pollution of soil heavy metals by using the distance determination coefficient method

Area of Science:

• Environmental Science
• Geochemistry
• Spatial Analysis

Background:

• Soil heavy metal contamination poses risks to ecosystems and human health.
• Accurate spatial distribution assessment is crucial for effective contamination control.
• Arid regions present unique challenges for soil contamination management.

Purpose of the Study:

• To simulate the spatial distribution of seven heavy metal concentrations in Wuwei, China.
• To compare the effectiveness of Land Use Regression (LUR) models against ordinary kriging (OK) interpolation.
• To identify key environmental factors influencing heavy metal concentrations in arid zone soils.

Main Methods:

• Soil surface samples (0-20 cm) were collected and analyzed for heavy metal concentrations.
• Land Use Regression (LUR) models were employed to predict spatial heavy metal distribution.
• A Distance Decay REgression Selection Strategy (ADDRESS) and distance-coefficient of determination (DCD) were used for covariate selection.

Main Results:

• LUR models demonstrated high accuracy with adjusted R-squared values above 0.6 for most heavy metals.
• LUR models outperformed ordinary kriging in predicting spatial heavy metal concentrations.
• Road proximity (motorways, primary, secondary roads) and building presence were significant factors, as was the normalized difference vegetation index (NDVI) and soil nutrients.

Conclusions:

• Land Use Regression (LUR) modeling is an effective tool for assessing heavy metal soil contamination in arid regions.
• Road networks and land use significantly influence heavy metal distribution within specific buffer zones.
• The study provides critical data for managing heavy metal contamination and associated health risks in arid environments.