ABSTRACT Antecedent soil moisture before freezing can affect greenhouse gases (GHG) fluxes from soils during thaw, but their critical threshold values for GHG fluxes and the underlying mechanisms are still not clear. By using packed soil-core incubation experiments, we have studied nitrous oxide (N2O), carbon dioxide (CO2) and methane (CH4) fluxes from a mature broadleaf and Korean pine-mixed forest soil and an adjacent white birch forest soil with nine levels of soil moisture ranging from 10 to 90% water-filled pore space (WFPS) during a 2-month freezing at −8°C and the following 10-day thaw at 10°C. The threshold values of soil moisture ranged from 50 to 70% WFPS for CH4 uptake and from 70 to 90% WFPS for N2O and CO2 emissions from the two soils during the freeze-thaw period. Under the optimum soil moisture condition, fulvic-like compounds with high bioavailability contributed more than 60% of dissolved organic matter (DOM) in the soil. Cumulative N2O emissions from forest soils during the freeze-thaw period were greatest when the concentration ratio of nitrate-N to dissolved organic carbon (DOC) was 0.04 g N g−1 C. Cumulative soil CO2 emissions and CH4 uptake during the freeze-thaw period were both regulated by the interaction between soil DOC and net N mineralization. The activities of β-1,4-glucosidase and β-1,4-N-acetyl-glucosaminidase, microbial biomass C and N, and the microbial biomass C-to-N ratios, were all significantly correlated to the soil N2O, CO2, and CH4 fluxes. Overall, upon a freeze-thaw period with different soil moistures, GHG fluxes from forest soils were jointly regulated by inorganic N and DOC concentrations, and related to the labile components of DOM released into the soil, which could be strictly controlled by the related microbial properties.