Empirical Models of Respiration and Net Ecosystem Productivity and Their Applications in a Subtropical Coniferous Plantation in China

Abstract

Net ecosystem exchange (NEE), solar radiation (including photosynthetically active radiation PAR), and meteorological parameters were measured in a subtropical coniferous plantation in China during 2013–2016. Applying the PAR balance principle at a canopy level and analyzing the observation data, an empirical model of respiration (Re, EMRe) considering 3-factor and 2-factor situations was developed and tested for all sky conditions. Generally, the respiration simulations were in reasonable agreement with the observations for the hourly, monthly, and annual sums of respiration. For example, using 3-factor and 2-factor models, the estimated annual sums of daytime and nighttime respiration in 2013–2016 overestimated that which was observed by about 31% and 26%, respectively. Further applications of EMRe and an empirical model of gross primary production (GPP, EMGPP) developed previously at this site, and an empirical model of net ecosystem productivity (NEP, EMNEP) using 3-factor and 2-factor models were obtained (NEP = GPP-Re) and evaluated for all sky conditions. Generally, the simulations of the hourly, monthly, and annual sums of NEP showed reasonable performances. The estimated NEP values overestimated the observations by 22% and 27% for the hourly sums in 2013–2016 when using the 3-factor and 2-factor models, respectively, and 7% and 12% for annual sums in 2013–2015 (2016 data were not used as the CO2 flux measurements had some problems in the 2016 summer). The NEP estimations were evidently improved when more factors (e.g., dark respiration) influencing Re were considered in the daytime respiration compared to those without considering these factors. To simplify the numerous and complicated CO2 processes in the simulations of Re and NEP, the PAR energy method was applied to capture and describe its main processes and energy interactions. The PAR energy method was suitable for studying the energy relationships associated with CO2 processes and developing empirical models for the simulations of GPP, Re, and NEP. These models were useful tools to investigate the multiple interactions and mechanisms between CO2, other atmospheric compositions, and PAR. Thus, the energy method is suggested to be applied to carbon balance.