The seismoacoustic analysis method has broad potential applications to source parameter estimation for near-surface explosion events such as industrial explosions and terrorist attacks. In this study, current models were improved by modifying the acoustic model and adopting the Bayesian Markov-chain-Monte-Carlo inversion method. The source parameters of near-surface small-yield chemical explosions were analyzed via the improved seismoacoustic analysis model and by the estimation accuracy of seismoacoustic joint inversion. Estimation and analysis results showed that the improved seismoacoustic analysis model considered ground shock coupling and the impact of explosion products ejecting from the surface so that the improved acoustic impulse relation was more consistent with the measured data than the Ford impulse relation. It is suitable for deep-burial, shallow-burial, and near-surface aerial explosions. Furthermore, trade-offrelationships were declined through the application of the improved model to source parameter inversion for near-surface small-yield chemical explosions, and source parameter estimation accuracy was improved.