SEARCH

SEARCH BY CITATION

Keywords:

  • Black spruce;
  • boreal forest;
  • closed-crown forest;
  • disturbances;
  • fire;
  • lichen woodland;
  • logging;
  • resilience;
  • spruce budworm outbreak

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Study area
  5. Materials and methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. References
  10. Biosketches

Aim  Our two main goals are first to evaluate the resilience of the boreal forest according to latitude across the closed-crown forest zone using the post-disturbance distribution and cover of lichen woodlands and closed-crown forests as a metric, and second to identify the disturbance factors responsible for the regeneration and degradation of the closed-crown forest according to latitude since the 1950s.

Location  The study area extends between 70°00′ and 72°00′ W and throughout the closed-crown forest zone, from its southern limit near 47°30′ N to its northern limit at the contact with the lichen woodland zone at around 52°40′ N.

Methods  Recent (1972–2002) and old (1954–1956) aerial photos were used to map the distribution of lichen woodlands across the closed-crown forest zone. Forest disturbances such as fire, spruce budworm (Choristoneura fumiferana (Clemens)) outbreak, and logging were recorded on each set of aerial photos. Each lichen woodland and stand disturbance was validated by air-borne surveys and digitized using GIS software.

Results  Over the last 50 years, the area occupied by lichen woodlands has increased according to latitude; that is, 9% of the area that was occupied by closed-crown forests has shifted to lichen woodlands. Although logging activities have been concentrated in the same areas during the last 50 years, the area covered by logging has increased significantly. Outbreaks by the spruce budworm occurred predominantly in the southern (47°30′ N to 48°30′ N) and central (48°53′ N to 50°42′ N) parts of the study area, where balsam fir stands are extensive. In the northern part of the study area (51°–52°40′ N), extensive fires affected the distribution and cover of closed-crown forests and lichen woodlands.

Main conclusions  Over the last 50 years, the area occupied by closed-crown forests has decreased dramatically, and the ecological conditions that allow closed-crown forests to establish and develop are currently less prevalent. Fire is by far the main disturbance, reducing the ability of natural closed-crown forests to self-regenerate whatever the latitude. Given the current biogeographical shift from dense to open forests, the northern part of the closed-crown forest zone is in a process of dramatic change towards the dominance of northern woodlands.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Study area
  5. Materials and methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. References
  10. Biosketches

The boreal forest is the largest biome in North America and extends across the continent from Newfoundland to Alaska. The eastern part of this biome is divided into three distinct zones corresponding to the closed-crown forest, the lichen woodland (or taiga), and the forest-tundra (Rowe, 1972; Payette et al., 2001). The closed-crown forest zone includes a majority of dense black spruce (Picea mariana (Mill.) B.S.P.), jack pine (Pinus banksiana Lamb.) and balsam fir (Abies balsamea (L.) Mill.) stands on well-drained soils. However, the overall distribution of these stands is not homogenous, as there are low-density or open forest stands forming lichen woodlands. The distribution of lichen woodlands may be viewed as the opposite of the distribution of closed-crown forests: lichen–black spruce stands are distributed predominantly on well-drained soils, whereas dense black spruce stands grow on poorly to well-drained soils (Hare, 1959; Hare & Ritchie, 1972; Rowe, 1984). The northern part of the boreal forest, the forest-tundra, is a large ecotone between the lichen woodland zone and the treeless tundra. The forest-tundra is composed of lichen woodlands and krummholz (stunted forests), and treeless tundra-like communities on well-drained sites.

The decrease of the closed-crown forest cover is the result of a regression process of the boreal forest under the long-lasting influence of disturbance factors occurring during the Holocene (Jasinski & Payette, 2005). There are currently three major disturbances in the eastern North American boreal forest, namely wildfire, logging, and insect outbreaks. Fire is the main disturbance factor in several parts of the boreal forest that inhibits or terminates ecological succession. Post-fire tree regeneration depends on several factors, including climatic conditions, seed quality and fire severity (Johnson, 1992; Payette, 1992; Johnstone & Kasischke, 2005; Jayen et al., 2006). Fires ignited by natural causes can cover large areas in the closed-crown forest zone, the taiga zone, and the forest-tundra (Johnson & Miyanishi, 1999; Payette et al., 2001; Bergeron et al., 2004).

In the closed-crown forest zone, the distribution of several lichen woodlands is the consequence of reduced post-fire regeneration caused by successive disturbances (Payette et al., 2000) not necessarily associated with limited climatic conditions. In contrast to the case for the taiga zone, both anthropogenic and natural disturbances occur frequently in the area occupied by the southernmost lichen woodlands. Natural disturbances such as insect outbreaks followed by fire (Payette et al., 2000) or anthropogenic disturbance such as logging followed by fire (Payette & Delwaide, 2003) cause the regression of the closed-crown forest. Logging is a new disturbance in the closed-crown forest zone. Since the end of the 19th century, the southern boreal forest has been subjected to logging, particularly in eastern Canada. In the early 1900s, logging activity in southern Québec was confined to areas south of 49°N, whereas it currently extends to 51°N, which corresponds to a northward displacement of 220 km.

Although the origin and dynamics of lichen woodlands at their southern range limit in eastern Canada have recently been documented (Payette et al., 2000; Payette & Delwaide, 2003; Jasinski & Payette, 2005), there are no data on their distribution and abundance across the closed-crown forest zone. Because lichen woodlands have been found to derive from closed-crown, spruce–moss forests unable to re-establish after disturbance, an evaluation of the proportion of both of these forest types across the closed-crown forest zone is necessary to document the current status and resilience of the closed-crown forest. In a boreal forest context, the resilience (sensuHolling, 1973) of the closed-crown forest corresponds to its ability to self-regenerate after a disturbance without any change in species composition and structure.

The distribution and abundance of the main forest types across the closed-crown forest zone may be represented by a model in which woodlands and forests coexist but in changing proportion according to latitude (Fig. 1) (see also Timoney et al., 1993). In the closed-crown forest zone, lichen woodland stands are sparsely distributed and their abundance increases to reach a maximum of occupation on well-drained soils in the taiga zone (Fig. 1). The maximum of the curve corresponds to a zone of about 2° of latitude in which most (> 95%) well-drained sites are occupied by lichen woodland stands. North of the area of maximum woodland cover in the taiga zone, tundra communities expand with latitude across the forest-tundra zone (Fig. 1 and Payette et al., 2001). Thus the overall distribution and coverage of the three main vegetation types, the closed-crown forest, the lichen woodland and the tundra, according to latitude are associated with a variable regeneration success closely linked with the most frequent disturbance factors. In this perspective, field surveys and historical data may be used to determine the degree of resilience of the closed-crown forest.

image

Figure 1.  Potential distribution of lichen woodland (percentage of the total well-drained soil surface occupied by lichen woodland) across the boreal forest in eastern Canada.

Download figure to PowerPoint

The main objectives of this study are to evaluate the spatial distribution of the lichen–spruce woodland and the resilience of the coniferous forest according to latitude across the closed-crown forest zone. It is hypothesized that the latitudinal distribution of the lichen–spruce woodland on well-drained soils follows a unimodal curve from a minimum cover at its southern limit of distribution to a maximum cover near the northern limit of the spruce–moss forest. It is also hypothesized that the coniferous forest is more resilient to disturbances in the southern part than it is in the northern part of the closed-crown forest zone. To meet our two objectives, we documented the distribution of the lichen woodland and of the closed-crown forest as well as the main stand disturbances over the last 50 years using aerial photos of a large area covering the latitudinal distribution of the closed-crown forest zone in central and southern Québec.

Study area

  1. Top of page
  2. Abstract
  3. Introduction
  4. Study area
  5. Materials and methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. References
  10. Biosketches

Lichen woodlands of the closed-crown forest zone are dominated by black spruce and/or jack pine with a cover generally less than 40% (Johnson & Miyanishi, 1999; Payette et al., 2000). The ground layer is dominated by Cladina rangiferina (L.) Nyl, Cladina stellaris (Opiz) Brodo, Cladina mitis (Sandst.) Hustich, and several species of the genus Cladonia. The shrub layer is composed of small shrubs such as Rhododendron groenlandicum Retz., Kalmia angustifolia L., Vaccinium angustifolium and Betula glandulosa Michx. (Morneau & Payette, 1989; Riverin & Gagnon, 1996; Johnson & Miyanishi, 1999; Payette et al., 2000; Simard & Payette, 2001).

The study area extends between 70°00′ W and 72°00′ W throughout the closed-crown boreal forest zone from its southern limit near 47°30′ N to its northern limit at the contact with the lichen woodland zone around 52°40′ N. The study area is dominated by black spruce–moss forest stands. Balsam fir stands with paper birch (Betula papyrifera Marsh.) and white spruce (Picea glauca (Moench) Voss) and jack pine stands are also present. Most of the studied forest stands are on well-drained, podzolic soils developed in glacial and fluvio-glacial deposits. For logistical and sampling purposes, the study area was subdivided into three parts. The southern part is located 120 km north-east of Québec City (47°30′ N, 70°–72° W) in the Parc des Grands-Jardins (PGJ) (see no. 1 in Fig. 2) and the Réserve faunique des Laurentides (RFL) (see no. 2 in Fig. 2). The area is dominated by balsam fir–paper birch stands and corresponds to the southernmost boreal forest zone (Bergeron, 1996). The mean altitude ranges between 600 m and 800 m above sea level (a.s.l.) in the PGJ, with scattered hills above 950 m a.s.l., whereas the mean altitude of the RFL is 900 m a.s.l. The mean annual temperature is about 0°C and −0.5°C in the PGJ and the RFL areas, respectively (Boisclair, 1990). There is a strong precipitation gradient from east to west; that is, from the PGJ (< 800 mm year−1) to the RFL (1500 mm year−1). Balsam fir and paper birch stands dominate at low and medium elevations in the RFL area, whereas mixed black spruce and balsam fir stands predominate at higher elevations. In the PGJ area, dense black spruce stands are found in wet and mesic sites, and lichen woodlands in mesic and dry sites. Clear-cutting and spruce budworm (Choristoneura fumiferana (Clemens)) outbreaks are the main disturbances in the RFL area. During the 20th century, there were three major outbreaks in the study area: 1914–1919, 1944–1951 and 1975–1985 (Blais, 1983). The spruce budworm outbreaks recorded during the 20th century extended across the closed-crown forest zone (Morin & Laprise, 1990). In the PGJ area, fire and spruce budworm outbreaks are the principal disturbances, but clear-cutting has been prohibited since the creation of the PGJ in 1981 (Dussart & Payette, 2002). The middle part of the study area (50° N, 70°–72° W) is located about 400 km north of Québec City and corresponds to the main core of the black spruce–moss forest within the closed-crown forest zone. The mean annual temperature is about 1.1°C, and there is more than 850 mm of precipitation (including 3.50 m of snow – 350 mm of precipitation) (Environnement Canada, 2003). The northern part of study area extends to the limit of the closed-crown forest zone and the beginning of the lichen woodland zone (52°41′ N, 70°–72° W). The mean annual temperature is about −2°C, and annual precipitation totals 800 mm of rain (including 3 m of snow) (Environnement Canada, 2003).

image

Figure 2.  Location of the study area (rectangle) in southern and central Québec. Vegetation zones according to Payette (1992) are distributed as follows from south: the mixed forest zone, the closed-crown forest zone, the lichen woodland zone (taiga), the forest-tundra zone, the shrub tundra zone, and the herb tundra zone.

Download figure to PowerPoint

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Study area
  5. Materials and methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. References
  10. Biosketches

The distribution of lichen woodlands and on-site disturbances (fire, insect epidemics and logging) were evaluated based on a comparative analysis of aerial photos taken at an interval of about 50 years. Two sets of aerial photos at a scale of 1/40,000 were considered, namely a recent collection (1972–2002) from the Québec photo service and an old collection (1954–1956) from the Canadian photo service. The only aerial photos available north of the logging zone date back to 1972, representing a 30-year difference from the 2002 survey. Lichen woodland cover and site disturbances are spatially dynamic, so, in order to have representative data, air-borne surveys and field verification were used to update the distribution of lichen woodlands and site disturbances to the year 2002. A total of 19 transects were used to evaluate the spatial distribution of the main vegetation types on well-drained soils and site disturbances over the 50-year period, from the southern limit of the lichen woodland at 47°30′ N to the northern limit of the closed-crown forest zone near 51°30′ N and the southern part of the lichen woodland zone (52°40′ N). Each transect was 1 km wide and 140 km long and was located at exactly the same geographic coordinates on each set of aerial photos. The first transect was randomly selected at the limit of the southern lichen woodlands in the PGJ area, and then the subsequent transects were placed systematically at every 15 min of latitude between 70° and 72° W.

The boundaries of each lichen woodland were delineated and validated in each transect so that the total area occupied by lichen woodlands could be calculated, and an identification of lichen woodlands was made on each set of aerial photos so that the difference in cover over the 50-year period could be evaluated. Aerial photos were corrected and interpreted by the authors. Then, air-borne surveys with field checks in each of the 19 transects were carried out for validation of every lichen woodland identified on the aerial photos according to methods in Payette et al. (2001). Multiple but small-scale disturbances were sometimes difficult to detect using photo interpretation (e.g. spruce budworm outbreak prior to wildfire), given that fire erases signs of previous disturbances. Only fires occurring during short intervals are generally detected on aerial photos, but no such fires were recorded in the study area. Logging areas were easy to identify on aerial photos because of their somewhat polygonal shapes and the presence of skid trails. Burned areas were more extensive than logged areas and showed uniform grey patterns. The density of dead stems lying on the ground was used to evaluate stand density prior to fire. The spatial pattern of a spruce budworm outbreak in a mature forest resembles the pattern of a Swiss cheese; that is, circular or crescent-shape hollows within a continuous and homogenous forest cover.

All lichen woodlands identified and validated on the two sets of aerial photos were digitized. Ground control points were determined with ground surveys (global positioning system) and 1/50,000 topographic maps provided by the Québec Department of Natural Resources. These points were taken on visible physical features on the landscape (water bodies, etc.). On the corresponding image, the x, y photo coordinates were then determined for each corresponding ground control point. Between 9 and 15 ground control points were established for each photo. The relationship of the x, y photo coordinated to the real world and the ground control points were then used to determine the algorithm to orthorectify the image. Orthorectification of aerial photos was realized using mapinfo Professional (version 7.5; MapInfo Corporation, Toronto, ON, Canada) with a pixel size of 25 m2. The ratio between the total lichen woodland (LW) cover and the total forest (CF + LW) cover was then calculated [LW/(CF + LW)] for each transect. All stand disturbances (fires, epidemics, logging) were identified on the two sets of aerial photos and used for the identification of the origin and evaluation of the dynamics of the two dominant forest types, i.e. the lichen woodland and the closed-crown forest.

Mean elevation was calculated using contour lines on forest-survey maps from the Québec Department of Natural Resources (Ministère des Ressources Naturelles de la Faune et des Parcs 2005). The mean elevation of each lichen woodland stand was determined from 1/50,000 topographic maps (10-m precision) and then used to calculate the mean elevation of lichen woodlands in each transect. Vegetation changes across the landscape from the southern limit to the northern limit of the closed-crown forest zone were evaluated over the 50-year window of observation, in terms of an increase or a decrease of the lichen woodland cover over the study time interval. All changes in the proportion of lichen woodland relative to the total forest cover in each transect and in all the transects were used to calculate the resilience of the closed-crown forest across the closed-crown forest zone. The resilience index (RI) of the closed-crown forest is the proportion of this forest that self-regenerated after stand disturbances (logging and fires) over the last 50 years. The resilience index was calculated for each transect along the gradient of latitude.

The areas covered by lichen woodlands and also by logged, burned and insect-damaged sites were measured using mapinfo Professional (version 7.5; MapInfo Corporation, 2005). Then, regression analyses between lichen woodland area, proportion of lichen woodland in the closed-crown forest and latitude were performed using the sas/stat® statistical software package (proc reg, SAS Institute 2000) (Devore & Peck, 1994). The evaluation of the changing vegetation cover according to the latitudinal gradient was based on the area covered by lichen woodlands [LW/(CF + LW)] and the total area occupied by well-drained sites. An analysis of frequency distribution (Wilcoxon matched pairs test) using statistica (Statsoft, 1984–2006) was used to compare the closed-crown forest cover of the 1950s and of 2002 (Devore & Peck, 1994). This test was also used to compare the areas of logged, burned and insect-damaged sites of the 1950s and in 2002.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Study area
  5. Materials and methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. References
  10. Biosketches

Except for the southernmost transects, the mean altitude of all study sites increases steadily from south to north (Fig. 3). The highest altitudes are in the southern part of the study area (RFL and PGJ), with similar values only in the northern part, with a mean altitude of 750 m a.s.l. near the Otish Mountains (52°23′ N) (see no. 9 in Fig. 2). The lowest elevations correspond to the Lac-St-Jean (see no. 3 in Fig. 2) lowlands (48° N), with a mean altitude < 350 m a.s.l. The overall structure of the main cover types (terrestrial vegetation, peatlands and water bodies) is fairly homogenous all along the latitudinal gradient. Forest vegetation covers nearly 80% of the surface in most transects. However, in transects 7, 9 and 13 (at 49°12′ N, 49°42′ N and 50°42′ N, respectively), the forest cover is < 80% because of large hydroelectric reservoirs (Pipmuacan, Péribonka and Manouane) (see nos 5, 6, 7, respectively, in Fig. 2). Similarly, the reduced forest cover at 48°23′ N and 51°42′ N results from the presence of two large lakes, Lac St-Jean and Lac Plétipi (see no. 8 in Fig. 2). The peatland cover in the study area is fairly homogenous, except for three areas where the peatlands extend over a much larger surface at 48°42′ N (Lac St-Jean), 50°42′ N (Manouane reservoir), and 51°12′ N (Lac Plétipi).

image

Figure 3.  Surfaces occupied by water bodies (lakes, reservoirs and rivers), peatlands and forests in each transect along the latitudinal gradient. The mean altitude of each lichen woodland along the latitudinal gradient is also shown.

Download figure to PowerPoint

The main disturbances that have influenced the vegetation cover over the last 50 years varied significantly along the latitudinal gradient. The southern part of the study area has been affected by clear-cuts since the end of the 19th century (Fig. 4a), whereas the northern part of the study area (from 50°53′ to 52°12′ N) has never been logged. Logging disturbance has changed significantly since the 1950s (Table 1). The RFL and PGJ areas corresponded to the heart of the logging zone during the 1950s. At present, the expansion to the north of logging corresponds approximately to the northern limit of logged areas during the 1950s. In the 1950s, the middle part of the study area (Péribonka Reservoir, 50° N) was subjected to clear-cuts near the Péribonka river, which was used for log floating over a distance of 250 km. Tree harvesting has doubled with the introduction of mechanical practices since 1970 (Table 1). The northern limit of clear-cut logging is at present located near 50°53′ N.

image

Figure 4.  Surface (percentage) occupied by (a) logged sites, (b) insect-damaged sites and (c) burned sites in each transect in the 1950s and in 2002 across the closed-crown forest zone.

Download figure to PowerPoint

Table 1.   Difference in the cover of lichen woodlands, logged areas, insect-outbreak areas and burned areas between the 1950s and 2002.
VariableMeanSDDiff.SD diff.Td.f.P-value
  1. *LW and CF refer to lichen woodlands and closed forests, respectively; T, Wilcoxon matched pairs statistic; d.f., degrees of freedom.

Logged areas 200215.49214.956 Logged areas 2002 vs. 1950
Logged areas 19503.4626.380 12.03  14.1283.711180.002
Insect-outbreak areas 20023.3875.044 Insect-outbreak areas 2002 vs. 1950
Insect-outbreak areas 19506.6869.779−3.30   6.971−2.062180.053
Burned areas 20023.4674.736 Burned areas 2002 vs. 1950
Burned areas 195017.43710.930−13.97  11.041−5.51518< 0.001
*LW area 20022448.9922867.005 LW area 2002 vs. 1950
LW area 19501530.9752131.242918.021157.6303.457180.002
Proportion LW/CF + LW 200220.61023.961 Proportion LW/CF + LW 2002 vs. 1950
Proportion LW/CF + LW 195012.91917.658  7.69   9.8333.410180.003

Of the three major spruce budworm outbreaks occurring in eastern Canada during the last century, the aerial photos taken between 1950 and 1955 show the immediate impact of the 1944–1951 outbreak (Table 1). The RFL was the most insect-affected area across the closed-crown forest zone. The Monts Valin area (49°23′ N) (see no. 4 in Fig. 2) was also heavily damaged in the 1950s. Both the RFL and Monts Valin areas are largely dominated by balsam fir–paper birch stands, which are more subject to attacks of the spruce budworm. Balsam fir–paper birch stands are seldom found in the northern part of the study area. During the 1950s, however, the forest was damaged by the spruce budworm even at these northern latitudes (51° N) (Fig. 4b). The recent aerial photos have not shown such degrees of disturbance. The RFL region was again affected by the insect during the 1975–1985 outbreak, with signs of stand deterioration. The forests that have regenerated over the last 20 years showed less damage by the spruce budworm than forests that regenerated during the 1950s.

Fire occurrences were widespread in the 1950s (Fig. 4c). Even in the humid areas of the PGJ and RFL, fires burned nearly 20% of the land. Although fires have affected both areas recently, the current distribution of burned sites was significantly different from that of the 1950s (Table 1). In the logging zone corresponding to the southern and the middle parts of the study area (48°53′ N to 50°42′ N), 1950s fires burned approximately 30% of the forest. Currently, extensive logging creates large firebreaks, and recently burned areas are practically absent. The northern part of the study area was less influenced by human activity, and the most common disturbances were large fires. In the 1950s, fires burned nearly 40% of the total forest cover from latitude 51°N to 52°12′ N. Even in 2002, large fires occurred in this area but with a far lower impact than in the 1950s (P < 0.001).

Both in the 1950s and in 2002, the lichen woodland cover increased significantly with latitude (R2 = 0.73, P < 0.0001, and R2 = 0.81, P < 0.0001, respectively) (Fig. 5a). The distribution of lichen woodlands has changed significantly over the last 50 years in the closed-crown forest zone (Table 1). Lichen woodlands currently cover only small areas in the southern part of the study area and in the logging zone. The shifting point of the respective dominance of the closed-crown forest in the south and the lichen woodland in the north is at latitude 51o N. However, the northernmost transects near the Otish Mountains around 52o N included a smaller lichen woodland cover both in the 1950s and today. Globally, from south to north, 19 500 ha of closed-crown forest has shifted into lichen woodland following 1950s disturbances, which represents a loss of about 9% of dense forest. The proportion of the lichen woodland cover relative to the overall forest cover shows the same trend (Fig. 5b & Table 1). Lichen woodland stands increased significantly across the whole study area (2002: R2 = 0.81, P < 0.0001; 1950: R2 = 0.61, P = 0.0002). In the southern part, lichen woodlands covered only 1% and 2.2% of the forest in the 1950s and in 2002, respectively, whereas in the logging zone (48°53′ N to 50°42′ N) this cover was between 1% and 6% in the 1950s and between 5% and 11% in 2002. Currently, the cover of lichen woodland varies between 20% and 77% in the northern part of the study area.

image

Figure 5.  (a) Lichen woodland cover and (b) proportion (percentage) of the lichen woodland relative to the total forest cover (LW/LW + CF) for each transect along the latitudinal gradient in the 1950s and in 2002. LW and CF refer to lichen woodlands and closed forests, respectively.

Download figure to PowerPoint

A small proportion of the closed-crown forest shifted into lichen woodland following logging (Fig. 6a). This shift represents 6% in the extensive logging zone and 7% in the PGJ area. Several closed-crown forests shifted to lichen woodlands following the 1950s fires. Closed-crown forests of the PGJ and RFL were less susceptible to shifting towards lichen woodlands than those in the central (48°53′ N to 50°42′ N) and northern (50°42′ N to 52°12′ N) zones. About 3% of the burned forests in the 1950s shifted to lichen woodlands in the PGJ and RFL areas. Between 8% and 30% of the closed-crown forests have shifted to lichen woodlands since the 1950s in the logging zone, and between 4% and 66% in the northern part of the study area. A proportion of the closed-crown forest shifted to lichen woodland following the 1950s fires, but another proportion remained stable; that is, the closed-crown forests self-regenerated after fire (Fig. 6b). In the RFL and PGJ areas, a large proportion (98%) of the closed-crown forest has remained stable over the last 50 years. In the logging zone, between 67% and 100% of the closed-crown forest self-regenerated, whereas only 15% to 56% of the northernmost closed-crown forest self-regenerated following the 1950s fires. The northern zone had a proportion of lichen woodlands significantly greater than those in the other zones before the 1950s fires. When these lichen woodlands burned, they self-regenerated without any shift towards closed-crown forests.

image

Figure 6.  (a) Proportion (percentage) of the closed-crown forest (CF) that has shifted to lichen woodland (LW) following fire and logging during the 1950s. (b) Stability of the closed-crown forest and the lichen woodland following fires during the 1950s. The percentages correspond to the proportion (percentage) of the total area in each transect.

Download figure to PowerPoint

The percentage of closed-crown forests that self-regenerated after stand disturbance over the 50-year period varies greatly according to latitude (Fig. 7). In the southern zone, the closed-crown forest is highly resilient to disturbances, with an RI varying from 0.8 in the RFL to 1.0 in the Lac St-Jean area. In the central zone, the RI decreases with increasing latitude, with higher values in areas dominated by extensive balsam fir forests and in high-altitude forests. In the northern zone, the RI drops dramatically to reach about 0.2; that is, in the forest areas most susceptible to shifting to lichen woodlands following fire disturbance. Near the Otish Mountains (52oN), the RI increases sharply to reach a value similar to or slightly lower than that of the logging zone.

image

Figure 7.  Resilience (percentage) of the closed-crown forest according to latitude. The resilience index (%) of the closed-crown forest is the proportion of this forest that self-regenerated after stand disturbance (fire and logging) over the last 50 years.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Study area
  5. Materials and methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. References
  10. Biosketches

The short- and long-term stabilities of biomes have been estimated based on palaeoecological data (Plöchl & Cramer, 1995; Lavoie & Payette, 1996; Williams et al., 2000; Masek, 2001; Yemshanov & Perera, 2001). In this study, direct observations based on a comparison of air photos taken at different times have shown lichen woodland expansion across the closed-crown forest zone since the 1950s. As hypothesized, the area covered by lichen woodlands increased significantly from south to north to reach a maximum at the limit of the closed-crown forest zone. According to Hare (1959), the northern limit of the closed-crown forest zone is at about 52° N. Our data show that the maximum increase of the lichen woodland cover is located south of Hare’s limit (52° N) and north of the central zone (48°53′ N to 50°42′ N) dominated by logging. The increase of lichen woodland cover within the closed-crown forest zone is the consequence of recurrent disturbances such as fire and logging, and if the trend is maintained in the long term it will result in a southward migration of the lichen woodland zone.

The overall resilience of the closed-crown forest is a function of stand disturbances, latitude and altitude. The nature of the disturbance varies spatially. In the southern zone, spruce budworm periodically affects black spruce stands (Blais, 1983; Jasinski & Payette, 2005). In the central zone, 75% of the area was subjected to logging, and in the northern zone, extensive fires were common. Closed-crown forests are less resilient to disturbances with increasing latitude. Although the reversion of lichen woodlands to dense black spruce stands is possible, no field evidence has yet been found across the closed-crown forest zone. Over the last 50 years, about 19,500 ha of the 225,000 ha of dense coniferous forest has shifted to lichen woodland, which represents about 9% of the forest. Lichen woodlands are expanding in the heart of the closed-crown forest zone and are in a process of massive establishment after fire in the northern zone.

The shift of the closed-crown forest takes place mainly in the northern part of the study area. Except for the PGJ area, the closed-crown forest in the southern part of the zone is more resilient to disturbance. During the 20th century, spruce budworm outbreaks or logging impacted on balsam fir forests several times (Blais, 1983; Payette et al., 2000; Jasinski & Payette, 2005). In this area where balsam fir dominates, the shift from closed-crown forest to lichen woodland is marginal because forest gaps are rapidly filled by balsam fir, paper birch, eastern larch (Larix laricina (Du Roi) K. Koch) and trembling aspen (Populus tremuloides Michx.). Indeed, the shift of the closed-crown forest to open forest occurs where black spruce is dominant and where fire intervals are relatively short (such as in the PGJ area). During spruce budworm outbreaks, most trees are killed, and survivors show reduced growth and seed production for several years (Morin & Laprise, 1990; Simard & Payette, 2001). If fire occurs shortly after an epidemic, the closed-crown forest is more susceptible to a shift to lichen woodland (Payette et al., 2000).

Closed-crown black spruce stands dominate the landscape in the central zone (48°53′ N to 50°42′ N), with logging as the main stand disturbance, and insect outbreaks and fires restricted spatially. In monospecific black spruce stands, logging can induce a shift from closed-crown forest to open forest. As a result, degradation of the closed-crown forest and reduction of tree density occurs when the latter species (balsam fir, paper birch or trembling aspen) are absent. Closed-crown forests of the central zone are less resilient to disturbance than those of the southern zone but more resilient than those of the northern zone. The northern zone is dominated by extensive fires, which are currently shifting closed-crown forests to lichen woodlands. In the northernmost closed-crown forests near the Otish Mountains, the reduced expansion of lichen woodland stands is related to topography and altitude, which is often above 1100 m a.s.l. Wetter conditions in the Otish Mountains, caused by orography, probably reduce fire ignition and frequency. With longer fire intervals, the inception and expansion of lichen woodlands are reduced greatly.

Lichen woodlands began to expand in the southern part of the closed-crown forest zone around 1500 years ago (Jasinski & Payette, 2005). This is supported by pollen diagrams of the study area, which show a decrease of conifer pollen in the closed-crown forest zone during the late Holocene (Garralla & Gajewski, 1992). The major increase of lichen woodland cover in the closed-crown forest zone as recorded in this study appears unusual, given the short time interval of 50 years. If this trend is maintained, closed-crown forests could disappear within about 550 years. However, it is likely that the conversion of closed-crown forest to lichen woodlands is part of a natural process of long-term biome changes associated with changing environmental conditions during the Holocene.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Study area
  5. Materials and methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. References
  10. Biosketches

For field and laboratory assistance, Vincent Beaulieu, Damien Côté, Ann Delwaide and Mathieu Tremblay are gratefully acknowledged. We are also grateful to Gilles Houle for his valuable comments on the manuscript. This research was financially supported by the Fonds Québécois de la Recherche sur la Nature et les Technologies (FQRNT, Québec), the Fond Forestier du Saguenay-Lac-St-Jean Région 02, The Consortium de Recherche sur la Forêt Boréale Commerciale and the Natural Sciences and Engineering Research Council of Canada (NSERC).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Study area
  5. Materials and methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. References
  10. Biosketches
  • Bergeron, J.-F. (1996) Domaine de la pessière noire à mousses. Manuel de Foresterie (ed. by J.Bérard), pp. 135240. Les Presses de l’Université Laval, Québec.
  • Bergeron, Y., Gauthier, S., Flannigan, M. & Kafka, V. (2004) Fire regimes at the transition between mixedwood and coniferous boreal forest in northwestern Québec. Ecology, 85, 19161932.
  • Blais, J.R. (1983) Trends in the frequency, extent and severity of spruce budworm outbreaks in eastern Canada. Canadian Journal of Forest Research, 13, 539545.
  • Boisclair, J. (1990) Parc des Grands-Jardins. Le plan directeur. Direction du plein air et du Parc des Grands-Jardins. Ministère des Loisirs, de la Chasse et de la Pêche du Québec, Québec.
  • Devore, J. & Peck, R. (1994) Introductory statistics, 2nd edn. West Publishing Company, St Paul, MN.
  • Dussart, E. & Payette, S. (2002) Ecological impact of clear-cutting on black spruce-moss forests in southern Québec. Écoscience, 9, 533543.
  • Environnement Canada (2003) Normales et moyennes climatiques, Chibougamau-Chapais, année 2002. Service de l’environnement atmosphérique, Environnement Canada, Ottawa.
  • Garralla, S. & Gajewski, K. (1992) Holocene vegetation history of the boreal forest near Chibougameau, central Québec. Canadian Journal of Botany, 70, 13641368.
  • Hare, F.K. (1959) A photo-reconnaissance survey of Labrador-Ungava, Memoir 6. Department of Mines and Technical Surveys, Ottawa.
  • Hare, F.K. & Ritchie, J.C. (1972) The boreal bioclimates. Geographical Review, 62, 333365.
  • Holling, C.S. (1973) Resilience and stability of ecological systems. Annual Review of Ecology and Systematics, 4, 123.
  • Jasinski, P. & Payette, S. (2005) The creation of alternative stable states by compounded disturbances in the southeastern boreal forest, Quebec, Canada. Ecological Monographs, 75, 561583.
  • Jayen, K., Leduc, A. & Bergeron, Y. (2006) Effect of fire severity on regeneration success in the boreal forest of northwest Québec, Canada. Écoscience, 13, 143151.
  • Johnson, E.A. (1992) Fire and vegetation dynamics: studies from the North American boreal forest. Cambridge University Press, Cambridge.
  • Johnson, E.A. & Miyanishi, K. (1999) Subarctic lichen woodlands. Savanna, barren and rock outcrop plant communities of North America (ed. by R.Anderson, J.Fralish and J.Baskin), pp. 421436. Cambridge University Press, Cambridge.
  • Johnstone, J.F. & Kasischke, E.S. (2005) Stand-level effects of soil burn severity on postfire regeneration in a recently burned black spruce forest. Canadian Journal of Forest Research, 35, 21512163.
  • Lavoie, C. & Payette, S. (1996) The long-term stability of the boreal forest limit in subarctic Quebec. Ecology, 77, 12261233.
  • Masek, J.G. (2001) Stability of boreal forest stands during recent climate change: evidence from Landsat satellite imagery. Journal of Biogeography, 28, 967975.
  • Ministère des Ressources Naturelles de la Faune et des Parcs (2005) 3e inventaire écoforestier. Direction des inventaires forestiers, Québec.
  • Morin, H. & Laprise, D. (1990) Histoire récente des épidémies de la tordeuse des bourgeons de l’épinette au nord du Lac Saint-Jean, Québec: une analyse dendrochronologique. Canadian Journal of Forest Research, 20, 18.
  • Morneau, C. & Payette, S. (1989) Postfire lichen–spruce woodland recovery at the limit of the boreal forest in northern Quebec. Canadian Journal of Botany, 67, 27702782.
  • Payette, S. (1992) Fire as a controlling process in the North American boreal Forest. A systems analysis of the global boreal forest (ed. by H.H.Shugart, R.Leemans and G.B.Bonan), pp. 144169. Cambridge University Press, Cambridge.
  • Payette, S. & Delwaide, A. (2003) Shift of conifer boreal forest to lichen–heath parkland caused by successive stand disturbances. Ecosystems, 6, 540550.
  • Payette, S., Bhiry, N., Delwaide, A. & Simard, M. (2000) Origin of the lichen woodland at its southern range limit in eastern Canada: the catastrophic impact of insect defoliators and fire on the spruce-moss forest. Canadian Journal of Forest Research, 30, 288305.
  • Payette, S., Fortin, M.-J. & Gamache, I. (2001) The subarctic forest–tundra: the structure of a biome in a changing climate. BioScience, 51, 709718.
  • Plöchl, M. & Cramer, W. (1995) Possible impacts of global warming on tundra and boreal forest ecosystems: comparison of some biogeochemical models. Journal of Biogeography, 22, 775783.
  • Riverin, S. & Gagnon, R. (1996) Dynamique de la régénération d’une pessière à lichens dans la zone de la pessière noire à mousses, nord du Saguenay-Lac-Saint-Jean (Québec). Canadian Journal of Forest Research, 26, 15041509.
  • Rowe, J.S. (1972) Forest regions of Canada. Canadian Forestry Service Publication 1300. Department of the Environment, Ottawa.
  • Rowe, J.S. (1984) Lichen woodland in northern Canada. Northern ecology and resource management (ed. by R.Olson, F.Geddes and R.Hastings), pp. 225237. University of Alberta Press, Edmonton.
  • Simard, M. & Payette, S. (2001) Black spruce decline triggered by spruce budworm at the southern limit of lichen woodland in eastern Canada. Canadian Journal of Forest Research, 31, 21602172.
  • Timoney, K.P., La Roi, G.H. & Dale, M.R.T. (1993) Subarctic forest-tundra vegetation gradients: the sigmoid wave hypothesis. Journal of Vegetation Science, 4, 387394.
  • Williams, J.W., Webb, T., III, Richard, P.J.H. & Newby, P. (2000) Late Quaternary biomes of Canada and the Eastern United States. Journal of Biogeography, 27, 585607.
  • Yemshanov, D. & Perera, A.H. (2001) Synthesizing published knowledge of boreal forest cover change for large-scale landscape dynamics modelling. Forestry Chronicle, 79, 132146.

Biosketches

  1. Top of page
  2. Abstract
  3. Introduction
  4. Study area
  5. Materials and methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. References
  10. Biosketches

François Girard is a PhD student at the Départment de Biologie and Centre d’Études Nordiques, Université Laval. His dissertation focuses on the origin and dynamics of lichen woodlands in the closed-crown forest zone.

Serge Payette is a professor of plant ecology and palaeoecology at the Département de Biologie and Centre d’Études Nordiques, Université Laval. As Chairman of the NSERC Northern Research Chair on disturbance ecology, he studies the relationships between boreal and subarctic ecosystems, climate and stand disturbances at various temporal and spatial scales.

Réjean Gagnon is a professor of forest ecology at the Département des Sciences Fondamentales and Chairman of the Consortium de Recherche sur la Forêt Boréale Commerciale. He studies the ecology of closed-crown forests and the dynamics of natural disturbances.

Editor: Glen MacDonald