Color profile: Disabled Composite Default screen
1742
Decomposition of broadleaf and needle litter in forests of British Columbia: influences of litter type, forest type, and litter mixtures1 C.E. Prescott, L.M. Zabek, C.L. Staley, and R. Kabzems
Abstract: We measured rates of decomposition at three sites representing the major mixedwood forest types of British Columbia: (i) boreal forests of white spruce (Picea glauca (Moench) Voss) and trembling aspen (Populus tremuloides Michx.); (ii) coastal forests of Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco) and red alder (Alnus rubra Bong.); and (iii) a wet interior forest of Douglas-fir, paper birch (Betula papyrifera Marsh.), and lodgepole pine (Pinus contorta Doug. ex Loud.). Mass loss of litter of each species (both pure and in combination with the other species) was measured for 2–5 years in forests of each species to determine (i) if broadleaf litter decomposed faster than needle litter, (ii) if litter decomposed faster in broadleaf or mixedwood forests than in coniferous forests, and (iii) if mixing with broadleaf hastened decomposition of needle litter. The broadleaf litters decomposed faster than needles during the first year but, thereafter, decomposed more slowly, so differences were small after 3 years. Litter tended to decompose faster in the broadleaf forests than in the coniferous forests. There was either no effect or a slight suppression of decomposition when litters were mixed; thus, there was no evidence that addition of broadleaf litter hastened decomposition of needle litter. The results clearly indicate that the mixing of needle litter with broadleaf litter is unlikely to hasten decomposition in mixedwood forests of British Columbia. The main influence of broadleaves was more rapid decomposition in broadleaf or mixedwood forest floors, which does not appear to be simply an effect of litter quality or litter mixing. Résumé : Les auteurs ont mesuré le taux de décomposition dans trois sites représentant les principaux types forestiers mixtes de la Colombie-Britannique : (i) forêt boréale d’épinette blanche (Picea glauca (Moench) Voss) et de peuplier faux-tremble (Populus tremuloides Michx.); (ii) forêt côtière de douglas de Menzies (Pseudotsuga menziesii (Mirb.) Franco) et d’aulne rouge (Alnus rubra Bong.); et (iii) forêt de l’intérieur humide composée de douglas de Menzies, de bouleau blanc (Betula papyrifera Marsh.) et de pin tordu (Pinus contorta Doug. ex Loud.). La perte de masse de la litière de chaque espèce (à la fois pure et combinée avec les autres espèces) a été mesurée durant 2–5 ans, dans les forêts de chaque espèce, en vue de déterminer (i) si la litière des feuillus se décompose plus rapidement que celle des conifères, (ii) si elle se décompose plus rapidement dans les forêts feuillues ou les forêts mixtes que dans les forêts conifériennes et (iii) si le mélange avec les feuilles des feuillus accélère la décomposition de la litière d’aiguilles. Durant la première année, la litière des feuillus se décomposait plus rapidement que celle des aiguilles, mais ensuite, elle se décomposait plus lentement, de sorte que la différence, après 3 ans, était faible. La litière avait tendance à se décomposer plus rapidement dans les forêts feuillues que dans les forêts conifériennes. Lorsque les litières étaient mélangées, il n’y avait pas d’effet, ou une légère diminution de la décomposition, donc pas d’évidence que l’addition de la litière des feuillus accélérait la décomposition de la litière des aiguilles. Les résultats indiquent clairement que le mélange de la litière des feuillus avec la litière des aiguilles n’accélère pas la décomposition dans les forêts mixtes de la Colombie-Britannique. La principale influence des feuillus réside dans une décomposition plus rapide sur le parterre de la forêt feuillue ou mixte, ce qui ne semble pas être simplement un effet de la qualité de la litière ou du mélange des litières. [Traduit par la Rédaction]
Prescott et al.
1750
Introduction There is increasing interest in Canada in management of mixedwood forests, particularly in boreal forests (Andison and Kimmins 1999). In British Columbia, the Forest Prac-
tices Code recommends maintenance of a broadleaf component on sites that naturally had broadleaves. Among the perceived benefits of a broadleaf component are more rapid decomposition and mineralization, which result in faster nutrient cycling and enhanced productivity (Perry et al. 1987;
Received November 18, 1999. Accepted August 1, 2000. C.E. Prescott,2 L.M. Zabek, and C.L. Staley. Department of Forest Sciences, University of British Columbia, 2424 Main Mall, Vancouver, BC V6T 1Z4, Canada. R. Kabzems. B.C. Ministry of Forests, Prince George Region, 8808 72 Street, Fort St. John, BC V1J 6M2, Canada. 1 2
Editorial decision on acceptance of this paper was made by W. Jan A. Volney. Corresponding author. e-mail:
[email protected]
Can. J. For. Res. 30: 1742–1750 (2000)
I:\cjfr\cjfr30\cjfr-11\X00-097.vp Monday, October 23, 2000 2:57:35 PM
© 2000 NRC Canada
Color profile: Disabled Composite Default screen
Prescott et al.
Comeau 1996; Simard 1996). Paper birch (Betula papyrifera Marsh.) has an “almost legendary reputation as a soil improver” (Binkley and Giardini 1998); Dimbleby (1952) suggested that birch could convert a raw humus into a fertile mull with high pH in 60–100 years. Clumps of aspen (Populus tremuloides Michx.) in boreal clearcuts had greater productivity, nutrient content, nutrient return in litter, and soil N availability, prompting Paré and Van Cleve (1993) to suggest that development of aspen after clear-cutting accelerates nutrient cycling because of higher nutrient concentrations in aspen foliar litter and subsequent faster decomposition. Binkley (1983) found greater soil C and N content and N availability and greater total net primary productivity where red alder (Alnus rubra Bong.) was present in a Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco) stand. There are several reasons why a broadleaf component might increase the rate of litter decomposition. First, broadleaf litter generally has higher nutrient concentrations and lower lignin and polyphenol concentrations than needle litter so would be expected to decompose faster (Perry et al. 1987; Peterson et al. 1997). Cornelissen (1996) found that leaves of deciduous species decomposed twice as fast as those of evergreens under controlled conditions. In the International Biological Program studies (Cole and Rapp 1981, cited in Perry et al. 1987), turnover of forest floor organic matter in temperate deciduous stands was more than four times faster than in temperate coniferous stands. Fried et al. (1990) found faster turnover of forest floor biomass under bigleaf maple (Acer macrophyllum Pursh) trees than under Douglasfir trees on the same sites. Klemmedson (1992) reported faster decomposition of Gambel oak (Quercus gambelii Nutt.) compared with ponderosa pine (Pinus ponderosa Dougl. ex. Laws). Flanagan and Van Cleve (1983) found that paper birch litter decomposed six times faster than black spruce (Picea mariana (Mill.) BSP) litter. However, other studies have not consistently reported faster decomposition or N mineralization of broadleaves compared with conifer needles (McClaugherty et al. 1985; Gower and Son 1992). Taylor and Parkinson (1988) found that trembling aspen leaves decomposed more rapidly than pine needles, except under very dry and cold conditions, when pine decomposed faster. The forest floor that develops under broadleaf or mixed forests might be more conducive to decomposition than that under conifers. Broadleaves are generally thought to produce mull forest floors that are richer in nutrients and promote rapid decomposition. This is largely a consequence of the greater abundance of soil macrofauna, particularly earthworms, in broadleaf forests (Killham 1994), whose comminution and mixing activities stimulate microbial activity, thereby hastening decomposition (Anderson 1988; Visser 1985). There is little evidence however that the nature of the forest floor greatly influences decomposition rates. McClaugherty et al. (1985) did not find faster decomposition of a standard litter in deciduous forests than in coniferous forests. In a microcosm experiment under identical climatic conditions, Prescott (1996) reported faster decomposition of birch leaf litter in a forest floor of red alder compared with lodgepole pine (Pinus contorta Dougl. var. latifolia Engelm.) or Douglas-fir forest floors; however, this difference was much smaller if the greater mass of faunal faeces in the alder for-
1743
est floor was included in the estimate of litter remaining. As Aber and Melillo (1991) point out, by moving or ingesting litter soil macrofauna make litter disappear more quickly, but the effect of soil fauna on decomposition is largely unknown. The mixing of needle litter with broadleaf litter in mixedwood forests may also hasten the decomposition of the needle litter. Such an effect has been reported by Fyles and Fyles (1993), when ground litter of red alder was mixed with Douglas-fir litter. Taylor et al. (1989) reported faster decomposition (after the initial leaching phase) of aspen litter when mixed with green alder (Alnus crispa (Ait.) Pursch) litter. Rustad and Cronan (1988) found that mixtures of white pine (Pinus strobus L.), red spruce (Picea rubens Sarg.), and red maple (Acer rubrum L.) decomposed faster than pure litter of any of the species. McTiernan et al. (1997) mixed litter of seven tree species (four broadleaf and three conifers) in all combinations and found significant stimulation of CO2 release in 8 of the 21 combinations and significant suppression in one combination. However, Klemmedson (1992) found no effect of mixing litters of Gambel oak and ponderosa pine nor did Rustad and Cronan (1998) find an effect of mixing litters of red spruce, white pine, and red maple on decomposition rate. Thomas (1968) found no effect of mixing loblolly pine (Pinus taeda L.) needles with dogwood (Cornus florida L.) leaves on first-year decomposition rate. In this study we measured rates of decomposition at three sites representing the major mixedwood forest types of British Columbia: (i) boreal forests of white spruce (Picea glauca (Moench) Voss) and trembling aspen; (ii) coastal forests of Douglas-fir and red alder; and (iii) a wet interior forest of Douglas-fir, paper birch, and lodgepole pine. In all three experiments, litter of each species was incubated in forests of each species and decomposition rates of both pure and mixed litter of each species were measured. By so doing we were able to address three questions pertinent to understanding decomposition in mixedwood forests. (1) Does broadleaf litter decompose faster than needle litter? (2) Does litter decompose faster in broadleaf or mixedwood forests than in coniferous forests? (3) Does mixing broadleaf and needle litter hasten decomposition of needle litter? We were also able to address a related question of whether or not litter of a given species decomposes fastest in forests of the same species. Field incubations of foliar litter enclosed in mesh bags over 2–5 years were used to address these questions.
Study sites and methods Experiment 1: spruce–aspen Four plots were established near Dawson Creek, B.C., in northeastern British Columbia (55°46′N, 120°14′W). Three of the plots (aspen, mixedwood, and clearcut) were in the Bear Mountain Community Forest, 10 km southwest of Dawson Creek. The white spruce plot was in Woodlot License 233, located 35 km northwest of Dawson Creek. There were no pure spruce stands closer to the other plots. All plots were in the moist–warm subzone of the Boreal White and Black Spruce biogeoclimatic zone (BWBSmw1; DeLong et al. 1990). Annual average precipitation is 467 mm of which mean rainfall is 250 mm from May to September, and an© 2000 NRC Canada
I:\cjfr\cjfr30\cjfr-11\X00-097.vp Monday, October 23, 2000 2:57:35 PM
Color profile: Disabled Composite Default screen
1744 nual mean snowfall is 189 cm; mean daily temperatures range from 15.4°C in July to –15.1°C in January (Lord and Green 1986). The sites are classified as Orthic Grey Luvisols, with loam to silt–loam textures. The aspen stand was 96 years old and had a density of 800 stems/ha. The mixedwood stand was 110 years old with 450 stems/ha, comprised of 56% spruce, 33% aspen, and 11% lodgepole pine. The clear-cut plot was part of the mixedwood stand that was harvested once, 7 years before this study began and brushed 1 year before. The spruce stand was 95 years old with a density of 1075 stems/ha. Foliar litters of white spruce and trembling aspen were collected in September 1992 from mature aspen and spruce stands near Dawson Creek. Litter bags (10 × 10 cm) were constructed of a double layer of 1.5 mm mesh fiber glass screening and filled with 2.0 g (dry mass) of litter. Three types of bags were made, containing either 2.0 g of spruce needles, 2.0 g of aspen leaves, or a mixture of 1.0 g of each species. Double bags were used to reduce spillage of spruce needles while allowing movement of soil fauna. Spillage from the litter bags during transport was always
