Effect of three partial cutting practices on stand structure and growth of residual black spruce trees in north-eastern Quebec

Partial cutting practices are increasingly used in boreal forests for two major reasons: (1) maintaining age structure and tree diameter distribution according to the principles of ecosystem-based management and (2) increasing tree growth by decreasing competition. This study evaluated the effects of three different partial cutting treatments applied to even- and uneven-aged black spruce (Picea mariana (Mill.) B.S.P.) stands in north-eastern Quebec, Canada. The effect of partial cutting was assessed by comparing treated and control plots in terms of age structure, diameter and spatial distribution, amount of deadwood and tree radial growth. Age structure and diameter distribution were not different from control plots after partial cuttings applied in both uneven-aged and evenaged stands, but lower deadwood basal area was observed. Tree radial growth generally increased following treatments in uneven-aged stands but can be limited by tree age and inter-tree competition. In even-aged stands, tree removal was more uniformly distributed and the overall reduction in inter-tree competition resulted in an increased tree radial  . Overall, these results suggest that the studied partial cuttings were adequate for maintaining structural attributes and increasing tree growth, but adjustments should be made to treatments to increase the amount of deadwood to a level observed in natural forests and to lower inter-tree competition.

Black spruce (Picea mariana (Mill.) B.S.P.) is a widespread species across North America, and its abundance and wood properties make it highly valued by the forest industry in north-eastern Canada (Viereck and Johnston, 1990; Saucier, 1998; Zhang and Koubaa, 2008). Natural structures of black spruce forest in eastern Quebec are heavily influenced by long fire return intervals that can reach over 500 hundred years because of high precipitation and cold temperatures (Foster, 1983; Bouchard et al., 2008). This long fire cycle is also associated with secondary disturbances, such as spruce budworm (Choristoneura fumiferana Clem.) outbreaks, which create uneven aged forest. Consequently, 60 – 70 per cent of black spruce-dominated stands are uneven-aged (Boucher et al., 2003; Côté et al., 2010). Compared with even-aged stands, uneven-aged stands have a wider range of age distribution (Rossi et al., 2009), which is generally associated with a higher proportion of small-diameter trees (10-14 cm; inverse J-shaped diameter distribution) or an irregular diameter distribution (Boucher et al., 2003) and a large amount of deadwood (McGee et al., 1999). The complexity of uneven-aged structures is generally accompanied by a distinct biodiversity as high heterogeneity offers habitats for a wide variety of birds, insects and plants (Bergeron and Noël, 2008; Lowe et al., 2011).

However, black spruce stands are traditionally harvested through clear-cutting (Youngblood and Titus, 1996), which promotes a regular, even-aged stand structure. In addition, clear-cutting can also decrease the long-term abundance and recruitment of deadwood (Bergeron et al., 1999) and thus alter the biodiversity compared with irregular, uneven-aged forests (Hansen et al., 1991; McComb et al., 1993). In recent years, partial cuttings have increasingly been used in black spruce stands of eastern Canada. The objective is to hasten the return to their initial structural attributes (age and diameter distributions). According to the principle of ecosystem-based management, they aim at harvest forests within the observed limits of occurrence of natural disturbances (Hunter, 1990; Bergeron et al., 1999). As they remove a portion of the stems, partial cuttings applied in uneven-aged stands emulate secondary disturbances which might accelerate the return to their initial structure. Nevertheless, in the short term (3 years or less), some partial harvesting treatments could simplify the stand structure (i.e. transform uneven-aged structure into even-aged) if the removal intensity is too high (Cimon-Morin et al., 2010; Ruel et al., 2013). Given that stand structure changes over time due to tree growth and recruitment, the negative effect of high intensity partial cutting may decrease with time.

Tree mortality after silvicultural treatments is increasingly evaluated due to its importance on both biodiversity and productivity. By decreasing stand density, partial cutting can reduce the density-dependent mortality (self-thinning) by decreasing the relative density, which is the ratio between the observed stand density and the maximum density attainable in a stand with the same mean volume (Drew and Flewelling, 1979), which is beneficial from a productivity point of view. However, deadwood is important in the ecosystem as it creates habitats for a wide variety of species (Bergeron and Noël, 2008; Jetté et al., 2008). For example, cavity- and snag-dependent species could be reduced (e.g. Vanderwel et al., 2009) with decreasing mortality of large-diameter trees after partial harvesting (Fridman and Walheim, 2000; Fraver et al., 2002). Dead large-diameter trees provide more advantages for biodiversity since they persist for a longer period of time and a wider range of vertebrates can use them when compared with smaller trees (Conner et al., 1975; Cline et al., 1980; DeGraaf and Shigo, 1985). In contrast, increased wind penetration in the stand following a partial cutting can increase the risk of windthrow. After intense tree removal in an irregular stand, 11 per cent (Groot, 2002) to 32 per cent (Riopel et al., 2010) of residual trees can be lost due to windthrow during the first 5 years. In natural forests, the risk of mortality generally increases with increasing tree size (Canham et al., 2001; Rich et al., 2007) although partial cuttings removing a large number of codominant and dominant trees, and thus leaving smaller trees with low vigour, could be associated with a greater risk of mortality (Caspersen, 2006; Powers et al., 2010). Mortality is therefore highly dependent upon the type of partial cutting method used (Powers et al., 2010). However, to our knowledge, only a few studies have compared the amount of deadwood in post harvested black spruce stands with unmanaged stands as well as compared different partial cutting methods (e.g. Cimon-Morin et al., 2010).

Tree spatial distribution, or horizontal structure, is another key characteristic of forest stands because it can affect productivity, stand dynamics, and structural diversity (Homyack et al., 2004). The horizontal structure is closely related to stand dynamics through processes such as competition, growth, mortality and tree recruitment (Cale et al., 1989; Nathan and Muller-Landau, 2000). Parameters such as soil, climate, topography and disturbance regimes can also influence the horizontal structure of a stand (Bonan, 1989). However, very little information is available on stand spatial structure following intermediate silvicultural treatments as most emphasis has been placed on stand composition and vertical structure (Gauthier et al., 2008) .

After partial cutting, residual tree growth is generally stimulated through increases in soil temperature, nutrient cycling (Thibodeau et al., 2000), moisture availability (Fayle, 1983) and solar radiation. Because stem value is largely dependent on its diameter, improved radial growth of residual trees may thus lead to higher financial returns at the stand level despite the reduced tree density (Liu et al., 2007). While tree growth responses to partial cutting of different intensities have been studied across the Canadian boreal forest (e.g. Abies balsamea (L.) Mill: Raulier et al., 2003; Bourgeois et al., 2004), little has been done on black spruce stands, except after commercial thinning (Vincent et al., 2009; Krause et al., 2011). While single-tree growth response was highly variable within a stand, a better growth response was observed on trees with the lower growth before thinning (Krause et al., 2011) and with less individual tree competition afterwards (Vincent et al., 2009). Therefore, tree growth response is expected to vary depending on the partial cutting method and the residual stand structural attributes (e.g. stand diameter distribution and spatial distribution). Although black spruce response to commercial thinning and fertilization has been quantified (e.g. Weetman 1975), knowledge on its growth response after partial cutting treatments, particularly in uneven-aged stands, is limited.