Comment on "Aging discrepancies of white spruce affect the interpretation of static age structure in boreal mixedwoods"

Abstract

There have been a recent series of papers that have documented underestimation of the age of boreal forest conifers. DesRochers and Gagnon (1997) demonstrated that excavation, serial sectioning to locate the root collar, and detailed dendrochronological analysis of aboveand below-ground sections (cross-dating) of the lower stem of black spruce (Picea mariana (Mill.) BSP) is necessary for exact age estimates. This has been further demonstrated on balsam fir (Abies balsamea (L.) Mill.) by Parent et al. (2000, 2002) and Parent and Morin (2002), and for a variety of boreal mixedwood species by Gutsell and Johnson (2002). Peters et al. (2002) have demonstrated the same difficulties in aging white spruce (Picea glauca (Moench) Voss). We have two objectives in writing this comment on the Peters et al. (2002) paper: (1) To clarify and discuss the issue of the timing of white spruce recruitment in boreal forests. We are concerned that Peters et al. (2002) will lead readers to believe that recruitment of white spruce in boreal mixedwood forests occurs only for a short period immediately after disturbance. (2) We also wish to discuss the place of very accurate but expensive techniques versus less accurate and inexpensive techniques in stand dynamics research. Peters et al. (2002) report the age structure and aging error for white spruce from nine 7-year-old and three older boreal mixedwood stands. They found a 2.4-year mean underestimate of time of establishment in a 20-year-old stand and a 6.4-year underestimate in two 38-year-old stands if tree sections were cut at ground level and rings counted, compared with ages determined by excavation to the root collar and careful cross-dating. Their conclusions were that “previous assumptions that age of white spruce can be accurately determined by ring counts at ground level (Youngblood 1995; Lieffers et al. 1996; Galipeau et al. 1997) are not valid in mixedwood stands in Alberta”, and “the mean number of missing rings found for trees in 38-year-old stands (6.4) nearly accounts for the regeneration lags and extended regeneration duration typically observed in non-cross-dated older stands (Youngblood 1995; Lieffers et al. 1996)”, and “aging errors have led to inaccurate depiction of regeneration patterns during early mixedwood stand development”. We are concerned that readers of Peters et al. (2002) will conclude that recruitment of spruce occurs almost exclusively in a short period after disturbance and that previous studies indicating that recruitment is sometimes delayed or extended were in error. From our reading of these three older papers cited in Peters et al. (2002), we believe that the authors would still have come to the same conclusions that they state in their papers even if they had aged their stands by excavation and cross-dating (assuming similar errors in aging). These papers’ general conclusions were that white spruce recruitment often begins immediately or shortly after disturbance, but in some stands or conditions the recruitment is delayed or extended over a long time period. The data from 46to 86-year-old Alberta stands in Lieffers et al. (1996) show two of nine stands with at least a 20-year delay before spruce recruitment began, and all sites with continuing spruce recruitment for 15–65 years. Youngblood (1995) found white spruce recruitment periods extending for more than 100 years based on ground-level age, although he did not sample all the spruce in his stands. Galipeau et al. (1997) found an initial 20-year white spruce recruitment pulse was followed by a second pulse beginning 40 years after fire in their 70-year-old stand. Even if the ground-level aging error increases with stand age, as Peters et al. (2002) suggest, the magnitude of the errors they found is certainly not large enough to eliminate these recruitment lags and prolonged recruitment periods. Peters et al. (2002) reported data on only three older stands with a narrow range of conditions; in all of their stands the stand-replacing fire was synchronous with a mast year, and there were ample mature spruce seed trees at time of disturbance. In boreal mixedwoods, however, there is considerable spatial and temporal variation in the availability of a seed source, substrates for seed germination, and in the light resource available for seedling growth and survival (Lieffers and Stadt 1994; Wright et al. 1998; Canham et al. 1999). Thus, the older stands studied by Peters et al. (2002) represented a small fraction of the conditions for white