The measurement of ecosystem level respiration has taken on new importance with the recognition of global warming associated with increases in atmospheric concentrations of CO2. To evaluate temporal and small-scale spatial variability in soil CO2 efflux chamber measurements were taken in two Douglas fir stands and a clearcut in coastal British. Study sites were located near Campbell River on Vancouver Island and in Pacific Spirit Park in Vancouver. No significant relationships between temperature or soil moisture and were revealed, however, graphical analysis suggests weak trends in the data. Spatial variability was high at all sites. This prompted a more detailed study of micro-variation in efflux at the Pacific Spirit Park site. To characterize the spatial variability, 192 chamber measurements were taken over two days. Frequency histograms and variograms were used to determine the distribution and spatial interdependence of efflux values. The frequency histogram curve of efflux values roughly follows a log-normal distribution. Efflux values were independent at approximately 80cm distance. Ground topography influences the rate of efflux. Efflux was higher in hollows than on decayed logs or hummocks. The mean rates of efflux for two days in late September/ early October 2001 were compared to under-story eddy covariance data collected at the same location in 1992. Mean efflux rates differed, however, different climactic conditions make the two data sets difficult to compare. In an attempt to increase the rate at which chamber data could be collected, I designed a non-inserting collar attachment for use with portable soil chamber systems. The efflux measurements collected with the non-inserting collar did not differ significantly from measurements collected with conventional, inserted PVC collars. This non-inserting collar technique is only effective on flat forest floor surfaces without major undergrowth, however, it speeds the measurement process. The new non-inserting collar method of taking chamber measurements will help researchers to capture the shape of soil CO2 efflux distributions by collecting more data points. Models of forest carbon dynamics can incorporate these spatial distributions of efflux to further describe and enhance our understanding of the carbon balance in forest ecosystems.
