Use of a regression method to partition sources of ecosystem respiration in an alpine meadow

The partitioning of ecosystem respiration (ER) into plant respiration (PR), aboveground-part respiration (AGR), root respiration (RR), and microbial respiration (MR) components is crucial for understanding the responses of carbon cycling in terrestrial ecosystems to climate change. Here, we present the ER-PB/AGB regression method, which is a modification of the SR-BGB method (PB, plant biomass; AGB, aboveground biomass; SR, soil respiration; BGB, belowground biomass) and is based on the assumption of a linear relationship between biomass and respiration rate for the partitioning of ER into PR, AGR, RR, and MR. Diurnal measurements of CO2 flux and biomass analysis were conducted in three Kobresia (Kobresia pygmaea, Kobresia humilis, and Kobresia tibetica) meadows on the Qinghai-Tibetan Plateau. We found significant linear relationships between ER and PB/AGB in the three meadows. However, the relationships between measured SR and BGB were either not significant or lower than those between ER and PB/AGB. The relative contributions of respiration components (AGR, RR, MR) to ER decreased consistently in the order AGR > MR > RR in the three Kobresia meadows. The contributions of RR and MR to SR calculated by the proposed ER-PB/AGB method differed widely among the three meadows and were consistently higher (RR) and lower (MR) than those by the SR-BGB method in all three meadows. Compared with the SR-BGB technique, our ER-PB/AGB regression method proved capable of determining more accurately the temporal changes in a larger number of respiration components.