Quantitative heterogeneous source apportionment of toxic metals through a hybrid method in spatial random fields

Area of Science:

• Environmental Science
• Geochemistry
• Pollution Control

Background:

• Toxic metals in soils present significant risks to food security and human health.
• Accurate source apportionment is essential for effective pollution prevention strategies.

Purpose of the Study:

• To develop and validate a novel hybrid method for exploring the heterogeneity of toxic metal pollution sources in spatial random fields.
• To improve the accuracy of toxic metal source apportionment and prediction.

Main Methods:

• A hybrid method combining positive matrix factorization (PMF), Bayesian maximum entropy (BME), and integrative predictability criterion (IPC).
• Application of the method to identify and quantify toxic metal sources in soil samples.

Main Results:

• Cadmium (Cd), Arsenic (As), and Copper (Cu) were identified as predominant soil pollutants, with exceedance rates of 27%, 12%, and 11%, respectively.
• The hybrid method showed superior prediction accuracy for toxic metals (44% and 46% improvement) when incorporating major and all sources.
• Major sources include coal combustion (Hg), traffic emission (Pb, Zn), industrial activities (As), agricultural activities (Cd, Cu), and natural sources (Cr, Ni).
• Non-major sources, such as industrial activities (Ni, Pb, Zn) in the north and natural sources (Cd, Cu, As, Pb, Zn) in the south, significantly impact regional pollution.

Conclusions:

• The novel hybrid method offers a superior approach for toxic metal source apportionment in spatial random fields.
• While major sources are primary contributors, the influence of non-major and local sources on regional toxic metal accumulation must be considered for targeted pollution control.