Post-fire structure and decomposition dynamics of coarse woody material in the western Canadian continental boreal forest

In the western Canadian continental boreal forest the structural properties and coarse woody material (CWM) fuel loads of different stages of early stand development (defined as tree seedling, tree sapling and young tree) were compared in aspen (Populus tremuloides), jack pine (Pinus banksiana) and black spruce (Picea mariana) stands using a chronosequence approach. Stand ages ranged from 8 to 57 years. Decomposition of the fire-originated CWM was measured as changes in wood density over time. Aspen sites had the highest average CWM fuel load of 21.4 Mg/ha. Jack pine had the lowest average fuel load of 2.65 Mg/ha. Black spruce stands had an average fuel load of 3.01 Mg/ha.; Weighted, nonlinear regression was used to estimate parameter values for linear, exponential and sigmoidal decomposition functions of CWM wood density. The performances of the functions were compared using a secondary Akaike Information Criterion (AIC_c). Post-fire standing dead trees did not decompose significantly. For aspen CWM the sigmoidal function had the best fit as indicated by the lowest AIC_c (−8.395), followed by the linear and the exponential function. For jack pine CWM the AIC _c of the linear and sigmoidal function were very similar (−10.978 and 10.551) and the exponential function had the highest value and the poorest fit. For black spruce CWM all equations performed poorly; the linear and exponential function had similar values of AIC_c (−20.732 and −19.876) whereas the sigmoidal function had the highest value.; Variables influencing the moisture content of CWM like distance of the CWM from the ground and an estimated climate moisture index affected decomposition dynamics differently: CWM close to the ground decomposed faster than CWM elevated >5 cm off the ground, the effect of the climate moisture index was most pronounced in jack pine increasing the lag-term for decomposition. Proposed conceptual decomposition models identify the need for more quantitative data to improve breakdown functions of snags and the vertical distribution of CWM during the first decades after fire.