Temporal variability in soil CO2 emission from an orchard was measured using a dynamic open-chamber system for measuring soil CO2 efflux in Heshan Guangdong Province, in the lower subtropical area of China. Intensive measurements were conducted for a period of 12 months. Soil CO2 emissions were also modeled by multiple regression analysis from daily air temperature, dry-bulb saturated vapor pressure, relative humidity, atmospheric pressure, soil moisture, and soil temperature. Data was analyzed based on soil moisture levels and air temperature with annual data being grouped into either hot-humid season or relatively cool season based on the precipitation patterns. This was essential in order to acquire simplified exponential models for parameter estimation. Minimum and maximum daily mean soil CO2 efflux rates were observed in November and July, with respective rates of 1.98 ± 0.66 and 11.04 ± 0.96 μmol m-2 s-1 being recorded. Annual average soil CO2 emission (FCO2) was 5.92 μmol m-2 s-1. Including all the weather variables into the model helped to explain 73.9% of temporal variability in soil CO2 emission during the measurement period. Soil CO2 efflux increased with increasing soil temperature and soil moisture. Preliminary results showed that Q10, which is defined as the difference in respiration rates over a 10 -C interval, was partly explained by fine root biomass. Soil temperature and soil moisture were the dominant factors controlling soil CO2 efflux and were regarded as the driving variables for CO2 production in the soil. Including these two variables in regression models could provide a useful tool for predicting the variation of CO2 emission in the commercial forest soils of South China . Temporal variability in soil CO2 emission from an orchard was measured using a dynamic open-chamber system for measuring soil CO2 efflux in Heshan Guangdong Province, in the lower subtropical area of ​​China. Intensive measurements were conducted for a period of 12 months. Soil CO2 emissions were also modeled by multiple regression analysis from daily air temperature, dry-bulb saturated vapor pressure, relative humidity, atmospheric pressure, soil moisture, and soil temperature. Data was analyzed based on soil moisture levels and air temperature with annual data being grouped into either This was essential in the order of relatively cool season based on the precipitation patterns. This was essential in order to make the exponential models for parameter estimation. Minimum and maximum daily mean soil mean CO2 efflux rates were observed in November and July, with rates of 1.98 ± 0.66 and 11.04 ± 0.96 μmol m-2 s-1 being recorded. Annual average soil CO2 emission (FCO2) was 5.92 μmol m- 2 s-1. Including all the weather variables into the model helped to explain 73.9% of temporal variability in soil CO2 emission during the measurement period. Soil CO2 efflux increased with increasing soil temperature and soil moisture. Preliminary results showed that Q10, which is defined as the difference in respiration rates over a 10 -C interval, was partly explained by a fine root biomass. Soil temperature and soil moisture were the dominant factors controlling soil CO2 efflux and were regarded as the driving variables for CO2 production in the soil. Including these two variables in regression models could provide a useful tool for predicting the variation of CO2 emission in the commercial forest soils of South China.