Untangling the effect of soil quality on rice productivity under a 16-years long-term fertilizer experiment using conditional random forest

•Most of the soil properties changed under different long-term fertilizer application treatments.•Conditional random forest was used to describe the rice yield variability from soil property.•Permutation based approach was used to select important soil properties.•Total K, urease, permanganate oxidizable carbon, and protease activities were most important for rice yield.•Accumulated local effect plots were used to get the effect of important soil properties.