Accurate chronologies of fluvial terraces are necessary to understand fluvial responses to climate and sea-level changes and uplift. Although feldspar post-infrared infrared stimulated luminescence (post-IR IRSL, or pIRIR) dating has been used to establish chronologies for Pleistocene fluvial terraces, the signals can be partially bleached in short, high-gradient river systems typical of Japanese rivers, resulting in overestimated ages. This study investigated aggradational fluvial terraces formed on the Tokachi Plain, northern Japan, during the last glacial period, and true depositional ages of the deposits were estimated by comparing multi-grain IRSL50/150, pIRIR50/150, and pIRIR50/225 ages, with single-grain pIRIR50/225 ages. IRSL50/225 and pIRIR50/225 ages of granite cobbles were also obtained and compared with the ages of multi-grains to evaluate the reliability of cobble ages. Fading-corrected IRSL 50/150 ages of multi-grains were consistent with pIRIR50/150 ages, indicating that the fading correction of IRSL50/150 signal was successful. Although equivalent dose distributions of single grains suggest that the pIRIR50/225 signal was completely bleached at deposition, pIRIR50/225 ages were slightly older than fading-corrected IRSL 50/150 and pIRIR50/150 ages possibly because of small residual doses. Fading-corrected IRSL50/225 and fading-corrected pIRIR50/225 ages of granite cobbles were consistent with the fading-corrected IRSL50/150 and pIRIR50/150 ages of multi-grains. However, comparison of cobble ages with climate records seems difficult due to their large errors. The fading-corrected IRSL50/150 and pIRIR50/150 ages of multi-grains indicate that fluvial aggradation occurred at 70–60 ka, followed by incision. Changes in the strength of East Asian summer monsoon precipitation are sufficient to explain fluvial evolution in the Tokachi Plain. Although some previous studies have argued that the ages of fluvial terraces might not be simply correlated to climate changes because of complex fluvial responses, significant changes in precipitation in the study area might have dominated the evolution of the entire fluvial system. Comparison with other fluvial systems in Japan suggests that slight differences in the responses of different fluvial systems to climate changes might be due to the spatial variability of those changes.