This study aims to reveal the characteristics of the changes in Net Ecosystem Exchange (NEE) in the rice paddy ecosystem in the Liaohe Plain and its correlation with meteorological factors. Utilizing the observational data from 2020 to 2022 collected at the Panjin Wetland Rice Paddy Agroecosystem Observatory, this study analyzes the changing characteristics of NEE, Gross Primary Productivity (GPP), and Ecosystem Respiration (Reco). The study analyzes the changing characteristics of temperature, precipitation, and wind speed and their correlations and interactions with NEE, GPP, and Reco. Additionally, it calculates the temperature sensitivity parameter for ecosystem respiration (Q10). The research results indicate that the annual total of NEE is negative, indicating a carbon sink. Specifically, NEE is negative during the growing season, representing a carbon sink, while it is positive during the non-growing season, indicating a carbon source. Among the three years, 2021 had the largest total NEE of −432.89 gCm-2•y-1. The combination of higher temperatures and more precipitation throughout the year contributed to the high NEE value. The daily variation of NEE first increases and then decreases, with the maximum NEE occurring in July, reaching a range of −20.6 to −26.0 μmol∙m- 2∙s-1. Temperature, precipitation, sunshine, and specific humidity all contribute to an increase in NEE values, among which temperature has the most significant impact on NEE. During the growing season, NEE significantly increases with rising temperature, reaching its maximum when the temperature is above 30℃, averaging −8.079 μmol•m-2•s-1. GPP values increased first and then decreased from June to September, with the maximum occurring in July, with a daily cumulative value of 10.69 to 13.55 gCm- 2•d-1. The annual variation of GPP peaks in summer, first increasing and then decreasing. The GPP in July and August accounts for 52.7 to 60.8% of the total annual GPP. There is a quadratic polynomial relationship of negative growth between NEE and net radiation, with the highest correlation value in July. During the daytime, NEE increases as net radiation increases but often lags behind the change in net radiation. Reco is the highest in the summer, up to 6.7 times higher than in other seasons. Reco exhibits an exponential relationship with soil temperature, with the highest correlation observed in autumn 2020. In summer, the variability of Reco with soil temperature is higher than in other seasons. The Q10 value in summer is the highest, ranging from 2.89 to 6.37, which is 2 to 3 times higher than in other seasons, followed by autumn. In 3a, the Q10 values in summer and autumn were the highest in 2021, and the corresponding annual NEE values were also the highest.