High-frequency dynamics of CO 2 emission flux and its influencing factors in a subtropical karst groundwater-fed reservoir, south China.

Revealing the magnitude, dynamics, and influencing factors of CO2 emissions across the water-air interface in karst water with high resolution is crucial for accurately assessing the carbon budget in a karst environment. Due to the limitations of observation methods, the current research is still very insufficient. To solve the above problems and clarify the main influencing factors of CO2 emission in karst water, this study selected Dalongdong (DLD) Reservoir, located in the typical karst peak and valley area in southwest China, to carry out a multi-parameter high-frequency monitoring study from January to December 2021, and used the thin boundary model method to estimate the CO2 flux across the water-air interface (CF). The average annual flux of DLD reservoir is 84.48 mmol·(m2 ·h)-1 , which represents a CO2 source overall. However, during the stratification period in August, there is a transient carbon sink due to negative CO2 emission. The alteration of thermal stratification in water is crucial in regulating the seasonal variation of CF. Meanwhile, the diurnal variation is significantly influenced by changes in hydrochemical parameters during the thermal stratification stage. Compared to low wind speeds (<3 m/s), high wind speeds (≥3 m/s) have a greater impact on the CO2 flux. Furthermore, high-frequency continuous data revealed that the reservoir triggered a CO2 pulse emission during the turnover process, primarily at night, leading to unusually high CO2 flux values. It is of great significance to monitor and reveal the process, flux, and control factors of CO2 flux in land water at a high-frequency strategy. They will help improve the accuracy of regional or watershed carbon budgets and clarify the role of global land water in the global carbon budget.