Fallen wood constitutes a critical component of the forest carbon pool,and the CO₂ emitted through its respiration is of great significance to global climate change and carbon cycle. In-depth study of the changing characteristics of respiration rate of fallen wood and its influencing factors can help to improve the understanding of the carbon dynamics of forest ecosystems. Karst forest is a natural forest with strong originality and more fallen woods in China,but the understanding of the respiratory model and driving factors of karst forest fallen woods is still unclear. Fallen woods are vital part of forest carbon pool,and the CO₂ released by their respiration is relevant to global climate change and carbon cycle. The dynamic closed-air chamber analysis method,utilizing the EGM-4 portable environmental gas detection system linked to a self-made chamber,was employed to monitor the daily and seasonal respiration dynamics of fallen woods from four prevalent tree species (Machilus rehderi,Cyclobalanopsis gracilis,Platycarya strobilacea,and Liquidambar formosana) at varying decay stages in karst evergreen and deciduous broad-leaved mixed forests. We assessed the impacts of time,tree species,decay degree,and physicochemical properties on the respiration rate of fallen wood,identified the principal factors influencing this rate,and developed a linear mixed-effects model incorporating these factors,with individual pieces of wood treated as a random effect. The results indicated: 1) The annual respiration variation of fallen wood from different tree species in karst forests exhibited a uniform pattern of single-peaked curves,peaking in summer (average 1.14 g CO₂ m^-2 h^-2) and bottoming out in winter (average 0.13 g CO₂ m^-2 h^-2). 2) The respiration rate of fallen wood was profoundly influenced by seasonal factors (P<0.001),with a notable interactive effect observed among seasons,species,and decay degree (P<0.01). The respiration rate of fallen woods was positively correlated with the temperature and initial elemental nitrogen content of theirselves,and negatively correlated with the water content. 3) The linear mixed-effects model's fit accuracy (R²=0.90,AIC=125.241,BIC=146.082),accounting for the random effect of individual pieces of wood,outperforms the extended model (R²=0.83,AIC=139.062,BIC=152.881) and the basic linear model (R²=0.73,AIC=151.660,BIC=159.476). This study revealed the annual dynamic characteristics of the inverted respiration rate and its main factors,emphasizing the important role of the degree of decay and environmental factors on the release of inverted CO₂. These results provide an important basis for the further study of the carbon cycle in karst forests,and also provide data support for the improvement of the forest carbon model.