Climate and soil pH control the reduction of cadmium in rice during 7 years of biannual liming

Background and aims Liming is widely known as a cost-efficient agricultural practice to reduce cadmium (Cd) concentration in rice ( Oryza sativa L.) grain (Cd R ) in the short term. However, the effect of long-term continuous liming on Cd R -reduction in the context of other environmental controls remains unknown. Methods We quantified the temporal stability of Cd R -reduction in response to biannually continuous liming and its correlations with dynamic soil properties and weather conditions based on a 7-year field experiment in a subtropical double-cropping rice ecosystem in a moderately Cd-contaminated acidic soil. Results The Cd R value declined by 51–90% with liming without compromising the rice yield. The degree of Cd R -reduction was similar between early and late rice seasons. However, the between-year Cd R -reduction was variable, remaining relatively stable in the first three years and fluctuating in the last four years. Soil pH and Olsen-P concentration at the ripening stage rose significantly, while the concentrations of diethylenetriaminepentaacetic acid (DTPA)-extractable Cd, Mn, and Zn in soil remained unaffected. The soil pH did not further increase after the second year of liming. The soil pH and Olsen-P concentration were positively correlated with cumulative liming, but the Cd R and Cd R -reduction exhibited no correlations with cumulative liming. The Cd R was correlated negatively with precipitation but positively with temperature and sunshine duration. Conclusions The effect of 7-year continuous liming on Cd R -reduction was temporally unstable and had no significant cumulative effect on Cd R . The between-year Cd R -reduction was not controlled by accumulative liming but primarily by dynamic soil properties and weather conditions.