ABSTRACT An accurate source term estimation of radionuclides released during the Fukushima accident is essential for understanding the accident progression of reactors and the environmental impact assessment. The hypothetical amount emission-regression estimation method was used, in which the deposition distribution is weighted based on the hourly deposition obtained from mesoscale meteorological model calculations assuming hypothetical amount emissions. The weighted emission rate is determined to minimize the difference between the sum of these distributions and a soil contamination map. The preceding study focused on obtaining tentative results for the source terms. Subsequent examination revealed that if any part of the time when a release is thought to have occurred is missing from the estimated release period, the entire calculation will be distorted, leading to the conclusion that a recalculation is necessary. Therefore, present study performed the recalculation by extending the estimation period to cover all major releases. Consequently, unspecified release events were clarified, and their correspondence to in-core events was confirmed. Based on the present study’s results, the possibility of increased iodine and Cs releases in the evening of 11 March 2011, and early mornings of March 13 and 14, 2011, which were not predicted by the WSPEEDI reverse calculation, was highlighted.