Abstract: Soil contamination presents significant threats to surrounding terrestrial ecosystems, highlighting the pressing need for cost-effective and refined ecological risk assessment (ERA) methods for risk mitigation. This study proposes a tiered ERA framework that systematically integrates source apportionment, spatial regression, ecological risk indices, probabilistic risk assessment and field-based ecological surveys of native species. The framework provides a synthesized evidence base for establishing "pollution-effects" relationships at contaminated sites. Case studies conducted at two distinct mining sites demonstrated its practicality and effectiveness. Through the sequential steps of risk identification, risk screening, probabilistic risk quantification and causal validation, this hierarchical ERA framework enhances both the precision and ecological relevance of risk assessments, offering a localized and scientifically robust foundation for the management of ecological risks in contaminated sites and surrounding soils.