Seasonal patterns in soil surface CO2 flux under snow cover in 50 and 300 year old subalpine forests

Soil CO2 flux can contribute as much as 60–80% of total ecosystem respiration in forests. Although considerable research has focused on quantifying this flux during the growing season, comparatively little effort has focused on non-growing season fluxes. We measured soil CO2 efflux through snow in 50 and 300 year old subalpine forest stands near Fraser CO. Our objectives were to quantify seasonal patterns in wintertime soil CO2 flux; determine if differences in soil CO2 flux between the two forest ages during the growing season persist during winter; and to quantify the sample size necessary to discern treatment differences. Soil CO2 flux during the 2002–2003 and 2003–2004 snow season averaged 0.31 and 0.35 lmols m 2 s 1 for the young and old forests respectively; similar to the relative difference observed during summer. There was a significant seasonal pattern of soil CO2 flux during the winter with fluxes averaging 0.22 lmols m 2 s 1 in December and January and increasing to an average of 0.61 lmols m 2 s 1 in May. Within-plot variability for measurements used in calculating flux was low. The coefficients of variation (CV) for CO2 concentration, snowpack density, and snow depth were 17, 8 and 14%, respectively, yielding a CV for flux measurements within-plot of 29%. A within plot CV of 29% requires 8 sub-samples per plot to estimate the mean flux with a standard error of ±10% of the mean. Variability in CO2 flux estimates among plots (size = 400 m) was similar to that within plot and was also low (CV = 28%). With a CV of 28% among plots, ten plots per treatment would have a 50% probability of detecting a 25% difference in treatment means for a = 0.05.