The present study aimed to evaluate the short-term biomass accumulation of forest trees exposed to wet acidic depositions. A hierarchical Bayesian model of tree growth was developed based on the data of a short-term experiment in which 2-year-old Japanese red pine (Pinus densiflora Sieb. et Zucc.) seedlings were exposed to aqueous phase OH radicals generated by an iron-oxalate-H2O2 mist (a pseudo-polluted dew) over two growing periods. We conducted a statistical comparison of tree growth between the control and pollution treatment groups by using the growth model incorporated the random effects due to the unknown characteristics of each seedling. The variability among seedlings is expressed in this model by the posterior probabilistic distributions of unobserved dry weight of a stem cohort before exposure treatment. The analysis of the effects of pollution treatment on the stem growth revealed that this treatment decreases the biomass allocation in the current year stems. However, the effects on the relative growth rate of pre-existing stems were unclear. Based on these results, we can speculate that in a polluted environment, the short-term growth of the young stems in the seedlings inhibited by pollution treatment, thereby resulting in the slowdown of long-term biomass accumulation. This can explain the patterns observed in the declining Japanese red pine forests that are subjected to (OH)-O-center dot-generating dews in the polluted area of western Japan. (c) 2006 Elsevier B.V. All rights reserved.