Objective  Heavy metal contamination of agricultural soils threatens food security. It’s imperative to seek effective methods to assure the safe production of crops. Understanding the long-term accumulation/loss patterns of heavy metals is crucial for the sustainable use of agricultural soils. Although many successful attempts have been made regarding the safe utilization of contaminated agricultural soils using either risk control or remediation approaches, the dynamic evolving nature of soil heavy metals in the long run had been neglected. Therefore it’s crucial to develop an effective method to predict metal accumulation patterns.

  Method  This study developed a Long-term Influx Versus Efflux (LIVE) model to assess the long-term evolving features of heavy metals in agricultural soils on the basis of mass balance calculations. Typical values of various inputs and outputs were selected, after which the accumulation behaviors of representative contaminants, namely, Cd and As were assessed.

  Result  Representative evolving patterns of soil metal concentrations are put forward, namely, unidirectional, fluctuating, and suddenly changing. These patterns are further verified with previously reported field monitoring data.

  Conclusion  Constant influx and efflux result in an unidirectional pattern, fluctuation in water flux leads to fluctuation in soil metal concentration, while source control of metal input can mitigate or even reverse the trend of metal accumulation immediately. The results will shed lights on the long-term safe utilization for contaminated agricultural soils.