Carbon dioxide exchange fluxes of a boreal peatland over a complete growing season, Komi Republic, NW Russia

Abstract

The carbon pool of peatlands has been considered as potentially unstable in a changing climate. This study is the first presenting carbon dioxide (CO₂) net ecosystem exchange, CO₂ efflux due to ecosystem respiration and CO₂ uptake by gross primary production over a complete growing season for different microforms of a boreal peatland in Russia (61°56′N, 50°13′E). CO₂ fluxes were measured using the closed chamber technique from the 25th April in the period of snow melt until the end of the vegetation period and the first frost on the 20th October 2008 at seven different microform types: minerogenous and ombrogenous hollows, lawns and hummocks, respectively, and Carex lawns situated in a transition zone between minerogenous and ombrogenous mire parts. The total number of chamber flux measurements was 5,517. Ombrogenous hummocks and lawns were sources of CO₂ over the investigation period whereas hollows and minerogenous lawns were CO₂ sinks. Some plots of Carex lawns and minerogenous hummocks were sinks while other plots of these microform types were sources. The CO₂ fluxes were characterised by large variability not only between the microform types but also within the respective microform types. Of all microform types, the Carex, ombrogenous, and minerogenous lawns showed the highest variability in CO₂ fluxes, which is probably related to a stronger within-microform heterogeneity in vegetation composition and coverage as well as in the water table level. Air temperature was one of the dominant controls on the CO₂ flux dynamics. Water table and green area index were found to have strong influence on CO₂ fluxes both within different patches of the same microform type as well as between different microforms.

Appendix

Description of measurement plots, model parameters, coefficient of determination, root mean square error, F values and levels of significance.

Table 4 Predominant vegetation (with coverage, %), microform type, mean water table depth (with range in parenthesis, the minus sign was used when the water table was above the surface) for the investigation period and maximum GAI of all measurement plots in the Ust-Pojeg peatland, 2008

Table 5 NEE model (Eqs. 2 and 3) parameter values (P ₁ if modelled based on GAI in μg CO₂ m⁻²s⁻¹°C⁻¹, if modelled based on ETI in μg CO₂ m⁻²s⁻¹°C⁻²; k ₁ in μmol CO₂ m⁻²s⁻¹; a ₁ in μg CO₂ m⁻²s⁻¹; b ₁ in °C⁻¹; c ₁ in cm⁻¹) and their standard errors in parentheses, coefficient of determination (R²), root mean square error (RMSE in μg CO₂ m⁻²s⁻¹), F value and number of measurements (N) for all plots

Table 6 Respiration model (Eqs. 4 and 5) parameter values (a ₂ in μg CO₂ m⁻²s⁻¹; b ₂ in °C⁻¹; c ₂ in cm⁻¹; u in cm; s in cm) and their standard errors in parentheses, coefficient of determination (R²), root mean square error (RMSE in μg CO₂ m⁻²s⁻¹), F value and number of measurements (N) for all plots

Table 7 GPP model (Eqs. 6 and 7) parameter values (P ₂ if modelled based on GAI in μg CO₂ m⁻²s⁻¹°C⁻¹, if modelled based on ETI in μg CO₂ m⁻²s⁻¹°C⁻²; k₂ in μmol CO₂ m⁻²s⁻¹) and their standard errors in parentheses, coefficient of determination (R²), root mean square error (RMSE in μg CO₂ m⁻²s⁻¹), F value and number of measurements (N) for all plots