4012bb.p65 ALCES VOL. 40, 2004 POTVIN AND COURTOIS - MOOSE IN CLEAR-CUTS 61 WINTER PRESENCE OF MOOSE IN CLEAR-CUT BLACK SPRUCE LANDSCAPES: RELATED TO SPATIAL PATTERN OR TO VEGETATION? François Potvin and Réhaume Courtois Ministère des Ressources naturelles et de la Faune, Direction de la recherche sur la faune, 930, chemin Ste-Foy, 4e étage, Québec, PQ, Canada G1S 2L4 ABSTRACT: Winter aerial surveys of moose (Alces alces) were completed on 14 landscapes (10– 256 km2 ) formed of aggregated black spruce (Picea mariana) clear-cuts logged 3–9 years ago in southcentral Québec. Moose were present in 8 landscapes (11 yards) and had a mean density of 0.20 moose/10 km2, which was 50% of the density observed in the same hunting zone with a similar forest composition. Based on previous work, effects of variability in hunting pressure and time since cutting were assumed not to influence distribution and abundance of moose. Browse density did not increase with age of cuts. Moose density was not related to the size of the clear-cut landscapes or the proportion of residual forest (18–40%) within each landscape (P = 0.14). Moose yards were not located close to uncut forest surrounding the landscapes and did not have a greater proportion of residual forest than clear-cut landscapes. Moose yards had a denser shrub layer and more browse available than random sites selected in the same landscapes. The presence of moose in large clear- cut black spruce landscapes is related to vegetation characteristics and not the spatial pattern of the forest. The authors propose two strategies to maintain moose populations and moose hunting activity in this type of forest after harvesting. ALCES VOL. 40: 61-70 (2004) Key words: Alces alces, black spruce, cover, food, habitat management, Picea mariana RÉSUMÉ: Nous avons réalisé l’inventaire aérien hivernal de l’orignal (Alces alces) dans 14 paysages formés de grandes coupes totales agglomérées (10–256 km2) effectuées au cours des 3 à 9 dernières années au centre-sud du Québec. La forêt d’origine était dominée par l’épinette noire (Picea mariana). La présence de l’orignal a été confirmée dans 8 paysages (11 ravages), pour une densité moyenne de 0,20 orignal/10 km2, soit 50% de celle retrouvée dans cette zone de chasse ayant le même type de forêt. D’après un travail antérieur, nous assumons que la variation de la pression de chasse et le nombre d’années après coupe n’ont pas influencé la distribution et l’abondance de l’orignal. La taille des paysages de coupe totale ou la proportion de forêt résiduelle (18–40%) à l’intérieur de ceux-ci n’a pas influencé la densité (P = 0,14). Les ravages d’orignaux n’étaient pas situés à proximité de la forêt intacte autour des paysages de coupe et ne contenaient pas davantage de forêt résiduelle que les paysages de coupe. Cependant, les ravages d’orignaux avaient une strate arbustive plus dense et une disponibilité de brout plus grande que des sites aléatoires choisis dans les mêmes paysages. La présence de l’orignal dans des paysages formés de grandes coupes totales en pessière noire est davantage reliée aux caractéristiques de la végétation qu’à la configuration spatiale de la mosaïque forestière. Nous proposons deux stratégies pour favoriser après coupe le maintien de l’orignal et de l’activité de chasse dans ce type de forêt. ALCES VOL 40: 61-70 (2004) Mots clés: Alces alces, aménagement forestier, couvert, épinette noire, nourriture, Picea mariana MOOSE IN CLEAR-CUTS – POTVIN AND COURTOIS ALCES VOL. 40, 2004 62 Although forest logging increases the amount of browse for moose (Alces alces), large clear-cuts can have detrimental ef- fects in the short term due to a lack of protective cover (Girard and Joyal 1984, Courtois and Beaumont 2002, Courtois et al. 2002). Clear-cutting, the prevalent har- vesting regime in the boreal forest in Canada, remains a controversial issue (Bliss 2000). Provinces have adopted regulations to make this technique more acceptable ecologically and socially. For example, over the past 15 years, regulations in Québec have limited the maximum size of clear-cuts to 250 ha (up to 1995) or 150 ha (since 1996). Uncut forest strips (60–100 m wide) are left be- tween 2 adjacent clear-cut patches. Ripar- ian buffer strips, 20-m wide along lakes and on each side of permanent streams, are also mandatory and can be used to separate adjacent cuts, but their width must then be increased to 60 or 100 m. Aggregating cutover patches, a common management strategy, has resulted in clear-cut domi- nated landscapes that may exceed tens and even hundreds of km2. Large clear-cut landscapes are more frequent in black spruce (Picea mariana) boreal forests, character- ized by a more uniform forest mosaic, than in balsam fir (Abies balsamea) or mixed forests. A previous study in coniferous and mixed boreal forest in south-western Québec has shown that moose seldom use recent clear- cuts, except in areas providing dense high regeneration and where logging is more of a partial type (Courtois and Beaumont 2002, Courtois et al. 2002). Because clear-cut landscapes are larger in black spruce for- ests, we suspect that the short-term impact of logging might be more severe in this type of forest than where that previous study took place. Traditionally, higher moose densities have been associated with mixed stands (Crête 1988) or areas disturbed by forest fires, insect outbreaks, or logging some 15-40 years ago (Crête 1977, Peek 1998). Rarely has moose abundance been related to the pattern of the forest mosaic, except for the edge between food and cover (Thompson and Stewart 1998, Courtois et al. 2002). The original black spruce forest being a poor habitat for moose (Brassard et al. 1974, Girard and Joyal 1984), recent cutovers in that type of forest must conse- quently be of very low value for moose. In such clear-cut landscapes, the pattern of the residual forest, such as buffer strips, might play a greater role than vegetation for the persistence of moose in recent cuts. From 1998 to 2000, we surveyed moose in very large black spruce clear-cuts in central Québec. We hypothesized that clear-cut landscapes constitute poor habi- tats for moose. To verify this hypothesis, we predicted that (1) moose would be at a lower density in recently clear-cut land- scapes, especially the largest ones, than in similar uncut black spruce forests. We also hypothesized that (2) moose yards would be located close to uncut forest surrounding the clear-cut landscapes or in those parts of the landscapes having a higher proportion of residual forest, and that (3) moose yards would be located in parts of the clear-cut landscapes where the shrub layer is more dense and browse more abundant. STUDY AREA Aerial surveys were conducted in for- est management unit 25-03 (49°02'–50°00' N, 72°31'–74°17' W), located north-west of Lac Saint-Jean, Québec (Fig. 1). We selected all clear-cut landscapes 10 km2 in this unit, within the black spruce–moss eco- logical zone (Grondin 1996). These 14 landscapes had been clear-cut between 1991 and 1997 and ranged in size between 10 and 256 km2 (Fig. 2). The proportion of residual forest (uncut) within landscapes ranged between 18 and 40% of the productive area (land area minus wetlands/bogs), with a ≥ ALCES VOL. 40, 2004 POTVIN AND COURTOIS - MOOSE IN CLEAR-CUTS 63 Fig. 1. Location of the 14 aerial survey blocks for moose in forest management unit 25-03 (gray shade) in Québec. Each block corresponds to a clear-cut landscape. mean value of 30%. The bulk of residual forest was made up of riparian and non- riparian buffer strips, but also included non- commercial forest patches (forests too young or at low density) and commercial forests in inaccessible areas (slope 40%) in some landscapes. Black spruce was the dominant tree species, with balsam fir, white birch (Betula papyrifera), trembling aspen (Populus tremuloides), and jack pine (Pinus banksiana) being locally present. The study area is part of hunting zone 18 west, which supports a density of 0.95 moose/10 km2 (Dussault 1999). In the northern portion of forest management unit 25-03 (black spruce-moss ecological zone), moose density is much lower than in the southern portion (balsam fir–white birch zone) (0.37 vs. 1.26 moose/10 km2). METHODS Aerial Surveys and Habitat Map Aerial surveys were conducted in Janu- ary of winters 1998 (1 landscape), 1999 (4), and 2000 (9). We used a Bell 206-B heli- copter flying at an altitude of 110 m and a speed of 160 km/hr (Courtois 1991). A navigator and 2 observers were on board. North-south survey lines were spaced 600 m along longitude transects (0.5 min). A 1- km distance was added to the beginning and end of each transect to delineate the survey blocks (Fig. 3). Moose sign was noted on 1:20,000 (1998) or 1:50,000 (1999, 2000) topographical maps showing recent clear- cut patches. A spatial database was built to produce the aerial survey maps and to analyse the data. The base map was the Québec 1:20,000 topographical map in numeric for- mat. It included watercourses, wetlands/ bogs, roads, and elevation contour lines. The delineation of recent cuts was obtained from the industrial forest company Abitibi- Consolidated Inc. We used ArcView 3.2 software and Spatial Analyst extension (ESRI 1996) to manage the database and to analyse the aerial survey data. The analysis of aerial surveys was done at 2 scales: clear-cut landscapes and moose yards. At the first scale, moose density was computed inside each landscape (excluding the 1-km buffers). We used the Pearson correlation coefficient to test if moose den- sity was related to the size of the landscape or to the proportion of residual forest within 0 50 100 150 200 250 300 1 2 3 4 5 12 13 14 15 17 18 19 20 B LANDSCAPE T O T A L A R E A ( km 2 ) 0 10 20 30 40 50 R E S ID U A L F O R E S T ( % ) . TOTAL AREA RESIDUAL FOREST Fig. 2. Total area of the 14 studied clear-cut landscapes and proportion of residual forest (uncut) within each one, expressed as the proportion of the productive area (land area minus WETLANDS/BOGS). ≥ MOOSE IN CLEAR-CUTS – POTVIN AND COURTOIS ALCES VOL. 40, 2004 64 each landscape. We also verified if moose yards tended to be located close to the uncut forest surrounding the landscapes. For that purpose, we generated 500-m buffers in- side each clear-cut landscape to evaluate the proportion of the surveyed area occu- pied by each distance class to surrounding uncut forest. Using a test, the actual number of moose yards by distance class was then compared with the theoretical number that corresponded to a distribution proportional to the area surveyed in each class. At the scale of the yards, we generated a 1-km2 circle around the center point of moose tracks to approximate the area used by moose. Using fresh moose tracks to delineate yards was not possible because the track network was too small and some- times consisted of a single linear track. To evaluate habitat selection of moose in cut- landscapes, we computed the Chesson- Manly selection index (Manly et al. 1993) for each yard for 4 habitat classes (WATER, WETLANDS/BOGS, RECENT CUTS, and UNCUT FOREST): where U i = proportion of habitat class i inside the 1-km2 circle and A i = proportion of habitat i inside the clear-cut landscape. The Friedman test was applied to this index in order to test the influence of habitat class on moose yard selection (yard = subject, habitat class = treatment). Vegetation Survey Time since cutting influences vegeta- tion in cutovers. To account for that varia- tion, we measured habitat variables at 19 sites randomly selected within clear-cut patches of the studied landscapes. We also compared vegetation in moose yards (n = 11) with random sites (n = 11) selected within clear-cuts in the same landscapes. CM Index i = U i / A i ∑ U i / A i Measurements were made at 6 sampling stations for each random site or yard, sys- tematically distributed (50 × 50 m spacing). Vertical crown closure of the tree (> 4 m) and shrub (1.5-4 m) layers was measured by the presence/absence of canopy over 10 points spaced on a line at a 3-m interval at each station (adapted from Bunnell and Vales 1990). We used a 2× metric prism and DBH measurements to evaluate the basal area and the number of trees > 9 cm D B H ( 1 p o i n t s a m p l e p e r s t a t i o n ) (Grosenbaugh 1952). The number of stems in the shrub layer (1–9 cm DBH) was counted inside a 25-m2 circular plot. The lateral cover was measured on a 2-m profile board at a 15-m distance (2 readings in opposite directions per station) (Nudds 1977). The median height of forest regen- eration was estimated visually inside a 15-m radius circle at each station. Browse avail- ability was evaluated by counting stems having at least 1 twig (10 cm long) between 50 and 300 cm from the ground in three 1- m radius circular plots per station. Black spruce was not included as browse because it is seldom used by moose. To compare each variable among cuts of different ages, we used a 1-factor ANOVA. The same test was also applied between moose yards and random sites selected in the same land- scapes. Fig. 3. Aerial survey plan for a clear-cut landscape. North-south transects are spaced 0.5 min of lon- gitude (600 m) apart. χ2 ALCES VOL. 40, 2004 POTVIN AND COURTOIS - MOOSE IN CLEAR-CUTS 65 RESULTS Moose Densities Snow conditions were very good in all aerial surveys, with 40-80 cm of snow depth, 5-20 cm of fresh dry snow, and absence of crust. Wind was light or absent and a clear sky enabled easy detection of moose tracks in these open landscapes. Moose tracks were detected in 8 landscapes, for a total of 11 yards and 22 moose (4 adult males, 10 adult females, and 8 calves). The overall density for the 14 landscapes (1,089 km2) was 0.20 moose/10 km2. There was no significant relationship between moose den- sity by clear-cut landscape and the size of the landscape (r = -0.31, P = 0.14) or the proportion of residual forest within each landscape (r = 0.32, P = 0.14) (Fig. 4). Moose Yards The number of moose yards by 500-m distance classes from the uncut forest in- side each landscape had a similar distribu- tion to the proportion of the landscape's area by distance classes ( = 0.02, P = 0.99) (Fig. 5). The 14 clear-cut landscapes con- tained 4.2% WATER, 7.8% WETLANDS/BOGS, 61.8% RECENT CUTS, and 26.1% UNCUT FOR- EST. The proportions of these cover types were different between moose yards and landscapes, with moose yards containing less WATER (0.9%) and WETLANDS/BOGS (3.8%) (Friedman test, P < 0.01) (Fig. 6). When testing only for RECENT CUTS and UNCUT FOREST, there was no difference be- tween moose yards and landscapes (P = 0.76). Vegetation Vegetation was quite similar among cuts of different ages, with only white birch stems in the shrub layer being more numer- ous in 8–9 year-old clear-cuts than in younger ones (P = 0.04) (Table 1). Moose yards had quite a different vegetation than the clear-cut landscapes where they were r = -0.31 P = 0.14 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 0 50 100 150 200 250 SIZE OF THE LANDSCAPE (k m 2) M O O S E / 1 0 k m 2 r = 0.32 P = 0.14 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 15 20 25 30 35 40 RESIDUAL FOREST (%) M O O S E / 1 0 k m 2 Fig. 4. Relationship between moose density in clear-cut landscapes and the size of the land- scape or the proportion of residual forest within each landscape. located (Table 2). Moose yards had greater vertical cover for the shrub layer (37 vs. 18%, P < 0.01), greater lateral cover (69 vs. 53%, P < 0.01), a more abundant shrub layer (2,600 vs. 1,300 stems/ha, P < 0.01), especially balsam fir (P = 0.02), and 3 times more browse available (23,200 vs. 7,100 stems/ha, P < 0.01). Conversely, black spruce was less abundant in the tree layer (basal area and stems/ha, P = 0.02) and shrub layer (P = 0.01) of moose yards. The height of the regeneration was the same in moose yards and in random sites (2.2 m). DISCUSSION Black spruce forests are poor habitats for moose, as opposed to mixed or decidu- ous stands (Brassard et al. 1974, Girard and Joyal 1984). The general density in this type of forest in the hunting zone was only 0.37 2 2 MOOSE IN CLEAR-CUTS – POTVIN AND COURTOIS ALCES VOL. 40, 2004 66 0 1 2 3 4 5 0-500 500- 1000 1000- 1500 1500- 2000 2000- 2500 2500- 3000 3000- 3500 3500- 4000 4000+ DISTANCE (m ) M O O S E Y A R D S . 0 5 10 15 20 25 30 35 40 L A N D S C A P E A R E A ( % ) . MOOSE YARDS LANDSCAPE AREA Fig. 5. Distribution of the number of moose yards in relation to the distance from the uncut forest surrounding each clear-cut landscape and proportion of the landscape area by 500-m distance classes. moose/10 km2, a low value (Dussault 1999). As expected from our first prediction, moose density was even lower inside our 14 clear- cut landscapes (0.20 moose/10 km2). How- ever, such density still represents about 50% of the density before cut. In a previous study, Courtois and Beaumont (2002) meas- ured a smaller decrease in moose density (23–30%) in 2 blocks that had been partially clear-cut (29–43% of their total area). In our study, the treatment was more severe because surveys were restricted to clear- cut landscapes instead of larger blocks con- taining a proportion of uncut forest. In the early 1980s, Girard and Joyal (1984) con- ducted aerial surveys in an area located at the same latitude as our study area. They reported densities of 0.36 moose/10 km2 in 60-km2 plots located in clear-cuts, as op- posed to 0.50 moose/10 km2 in uncut forest. All these results indicate that moose can persist in landscapes that have been re- cently logged (< 10 years) and where large clear-cuts (= 250 ha) are aggregated. Moose densities in such landscapes are much lower though than in uncut forest. This confirms part of our first prediction. On the other hand, moose density was not higher in the smaller clear-cut landscapes (Fig. 4). While 0 10 20 30 40 50 60 70 WATER BOGS CUTS FOREST % O F T H E A R E A . LANDSCAPES MOOSE YARDS 5 landscapes smaller than 100 km2 had the highest densities ( ≥ 0.4 moose/10 km2), 5 other landscapes in the same area class had no moose. In Ontario, increase in hunting pressure in logged areas had a negative effect on moose densities (Eason et al. 1981, Eason 1989, Rempel et al. 1997). Conversely, in southwestern Québec, the accessibility of- fered by new logging roads had a minor impact on harvest rates in blocks recently clear-cut (Courtois and Beaumont 1999). We did not measure hunting pressure in our study but Bertrand and Potvin (2002) ana- lysed moose harvest statistics in those same 14 clear-cut landscapes and in the entire management unit 25-03 from 1981 to 2000. A similar trend between both data sets suggests that the effect of hunting on moose Fig. 6. Cover types of 14 clear-cut landscapes and of 11 moose yards located in the same landscapes. ALCES VOL. 40, 2004 POTVIN AND COURTOIS - MOOSE IN CLEAR-CUTS 67 Table 1. Vegetation of clear-cut black spruce landscapes according to the time since cutting. ( x SE) abundance was not different in the clear- cut landscapes than outside. Contrary to our second prediction, spa- tial analysis was unable to explain the pres- ence or the location of moose yards inside clear-cut landscapes. Moose yards were not located close to the uncut forest sur- rounding the landscapes and did not have a higher proportion of residual forest. Forest strips between clear-cut patches or riparian buffer strips along streams are not attrac- tive to moose in winter probably because of their small size (60–100 m width) and low browse production (black spruce forest). Furthermore, as suggested by Courtois et al. (2002), cover does not appear to be a major component of habitat for moose even in late winter, at least in regions where snow depth is usually < 90 cm. In the clear-cut landscapes in general, browse availability and density of the shrub layer were still poor 3–9 years after log- ging. In moose yards, the vegetation was quite different from that of random sites within the same landscapes in that shrubs and small trees were more abundant, espe- cially balsam fir, and there was greater vertical cover, lateral cover, and more browse. Before conducting the ground sur- vey in recently logged areas, we did not expect to find this type of vegetation, which is more characteristic of balsam fir or mixed 3–4 years 5–7 years 8–9 years (n = 10) (n = 6) (n = 3) Crown closure (%) Trees (> 4 m) 2 ± 1 2 ± 1 7 ± 3 0.91 Shrubs (1.5-4 m) 7 ± 1 13 ± 3 24 ± 6 0.25 Tree layer (> 9 cm DBH) Basal area (m2 / ha) Black spruce 0.3 ± 0.2 0.9 ± 0.3 0.1 ± 0.1 0.97 Balsam fir 0.0 ± 0.0 0.0 ± 0.0 0.3 ± 0.2 1 White birch 0.1 ± 0.1 0.3 ± 0.2 0.7 ± 0.4 0.96 Total 0.4 ± 0.2 1.4 ± 0.4 1.0 ± 0.6 0.99 Stems / ha Black spruce 32 ± 18 87 ± 35 9 ± 9 0.86 Balsam fir 0 ± 0 0 ± 0 18 ± 15 1 White birch 3 ± 2 13 ± 7 18 ± 11 0.85 Total 35 ± 18 120 ± 37 44 ± 25 0.92 Shrub layer (1–9 cm DBH) Black spruce 500 ± 110 520 ± 170 170 ± 97 0.91 Balsam fir 93 ± 35 0 ± 0 470 ± 160 0.14 White birch 7 ± 7 89 ± 48 390 ± 170 0.04 Trembling aspen 7 ± 7 480 ± 200 13 ± 13 0.69 Total 610 ± 110 1,200 ± 240 1,300 ± 340 0.22 Lateral cover (%) 42 ± 2 52 ± 4 57 ± 4 0.11 Height of regeneration (m) 1.8 ± 0.1 2.0 ± 0.2 1.8 ± 0.2 0.22 Available browse (stems / ha) 6,200 ± 750 10,700 ± 1,800 11,900 ± 1,900 0.07 Variable P MOOSE IN CLEAR-CUTS – POTVIN AND COURTOIS ALCES VOL. 40, 2004 68 Table 2. Vegetation of moose yards located in clear-cut black spruce landscapes and of random sites selected within clear-cuts of the same landscapes. ( x SE) Moose yards Random sites (n = 11) (n = 11) Crown closure (%) Trees (> 4 m) 4 ± 1 5 ± 2 0.83 Shrubs (1.5-4 m) 37 ± 3 18 ± 3 < 0.01 Tree layer (>9 cm DBH) Basal area (m 2 / ha) Black spruce 0.1 ± 0.1 0.7 ± 0.2 0.02 Balsam fir 0.1 ± 0.0 0.1 ± 0.1 0.56 White birch 0.5 ± 0.2 0.4 ± 0.2 0.91 Total 0.6 ± 0.2 1.4 ± 0.4 0.1 Stems / ha Black spruce 10 ± 6 71 ± 25 0.02 Balsam fir 8 ± 5 8 ± 7 0.97 White birch 16 ± 7 14 ± 6 0.82 Total 33 ± 12 100 ± 27 0.02 Shrub layer (1–9 cm DBH) Black spruce 240 ± 60 590 ± 130 0.01 Balsam fir 550 ± 160 140 ± 60 0.02 White birch 410 ± 120 150 ± 70 0.07 Trembling aspen 850 ± 280 250 ± 110 0.05 Total 2,600 ± 400 1,300 ± 200 < 0.01 Lateral cover (%) 69 ± 3 53 ± 2 < 0.01 Height of regeneration (m) 2.2 ± 0.1 2.2 ± 0.1 0.8 Available browse (stems / ha) 23,200 ± 2,400 7,100 ± 1,300 < 0.01 Variable P forests than of black spruce stands, at such northern latitudes. In conformity with our third prediction, this confirms that food pro- duction is the main factor driving habitat selection by moose, as suggested by Crête (1977), Peek (1998), and Courtois et al. (2002). MANAGEMENT IMPLICATIONS In black spruce forest, it takes more than 10 years for browse availability to become attractive to moose in recent clear- cuts. In our study, moose density was very low and in most cutovers, even the oldest ones, and food production was much lower in control sites than in moose yards. These observations confirm our hypothesis that clear-cut black spruce landscapes are poor moose habitats. In balsam fir or mixed forests, 10-year cutovers are much more productive (Potvin et al. 2004). Special measures are therefore needed in the planning process if the goal is to maintain moose populations and moose hunt- ing activity in this type of forest. As pro- posed by Courtois et al. (2002), there can be 2 strategies: (1) identify stands that have greater value for moose and use silvicultural techniques that will maintain their charac- teristics, or (2) leave a higher proportion of residual forest (50–60%) by applying a dis- persed patch cutting strategy. Since mixed ALCES VOL. 40, 2004 POTVIN AND COURTOIS - MOOSE IN CLEAR-CUTS 69 and deciduous stands are rare within the black spruce–moss ecological zone, the first strategy might be implemented as a basic step where moose production is a moderate priority issue. The second strategy is more aimed at maintaining moose hunting activ- ity. Although moose density in our study was not related to the proportion of residual forests, moose hunters have a negative perception of forest harvesting systems (Courtois et al. 2001). They prefer land- scapes where residual forests are dominant over those where clear-cut patches are aggregated and narrow uncut strips are the sole residual forest. The second strategy might be suited to areas where moose pro- duction is a higher priority (outfitters areas, areas devoted to forest–wildlife integrated management). In other areas, the goal might be to decrease moose numbers. For example, the conservation of woodland caribou (Rangifer tarandus) may be at risk be- cause of predation, if a large moose popula- tion exists that can sustain high wolf num- bers (Courtois 2003). In this case, caribou habitat management involves preserving large forest blocks (> 250 km2) intercon- nected with a corridor network. In order to diminish moose and predator densities, clear- cuts outside preserved areas should aim at regenerating black spruce and very few deciduous species. In this context, stands that may be suitable to moose (e.g., mixed forests) might even be transformed towards coniferous stands. ACKNOWLEDGEMENTS We would like to thank Charles Faucher (deceased), who conducted most of the spatial analyses, wildlife technicians of Saguenay-Lac-Saint-Jean region at Société de la faune et des parcs du Québec, who took part in the aerial surveys, and Normand Bertrand, Laurier Breton, and Claude Paquet, who implemented the vegetation survey. Special thanks to Abitibi-Consoli- dated Inc. for providing the maps of the recent cuts. REFERENCES BERTRAND, N., and F. POTVIN. 2002. Utili- sation par la faune de la forêt résiduelle dans de grandes aires de coupe: synthèse d’une étude de trois ans réalisée au Saguenay-Lac-Saint-Jean. Québec Ministère des Ressources naturelles, de la Faune et des Parcs, Report DEF- 0231. BLISS, C. J. 2000. Public perceptions on clearcutting. Journal of Forestry 98(12):4-9. BRASSARD, J. M., É. AUDY, M. CRÊTE, and P. GRENIER. 1974. Distribution and winter habitat of moose in Québec. Naturaliste canadien 101:67-80. BUNNELL, F. L., and D. J. VALES. 1990. Comparison of methods for estimating forest overstory cover: differences among techniques. Canadian Journal of Forest Research 20:101-107. COURTOIS, R. 1991. Normes régissant les t r a v a u x d ’ i n v e n t a i r e s a é r i e n s d e l’orignal. Québec Ministère du Loisir, de la Chasse et de la Pêche, Report 1907. _____. 2003. La conservation du caribou forestier dans un contexte de perte d’habitat et de fragmentation du milieu. Ph.D. Thesis, Université du Québec à Rimouski, Rimouski, Québec, Canada. _____, and A. BEAUMONT. 1999. The influence of accessibility on moose hunt- ing in northwestern Québec. Alces 35:41-50. _____, and _____. 2002. A preliminary assessment on the influence of habitat composition and structure on moose density in clearcuts of north-western Québec. Alces 38: 167-176. _____, C. DUSSAULT, F. POTVIN, and G. DAIGLE. 2002. Habitat selection by MOOSE IN CLEAR-CUTS – POTVIN AND COURTOIS ALCES VOL. 40, 2004 70 NUDDS, T. D. 1977. Quantifying the veg- etation structure of wildlife cover. Wild- life Society Bulletin 5:113-117. PEEK, J. M. 1998. Habitat relationships. Pages 351-373 in A. W. Franzmann and C. C. Schwartz, editors. Ecology and Management of the North Ameri- can Moose. Smithsonian Institution Press, Washington, D.C., USA. POTVIN, F., L. BRETON, and R. COURTOIS. 2004. Response of beaver, moose and snowshoe hare to clearcutting in Québec boreal forest: a reassessment 10 years after cut. Canadian Journal of Forest Research 35: 151 - 160. REMPEL, R. S., P. C. ELKIE, A. R. RODGERS, and M. J. GLUCK. 1997. Timber man- agement and natural-disturbance ef- fects on moose habitat: landscape evalu- ation. Journal of Wildlife Management 61:517-524. THOMPSON, I. D., and R. W. STEWART. 1998. Management of moose habitat. Pages 377-401 in A. W. Franzmann and C. C. Schwartz, editors. Ecology and Man- agement of the North American Moose. Smithsonian Institution Press, Wash- ington, D.C., USA. moose (Alces alces) in clear-cut land- scapes. Alces 38: 177-192. _____, J. P. OUELLET, and A. BUGNET. 2001. Moose hunters’ perception of forest harvesting. Alces 37:19-33. CRÊTE, M. 1977. Importance de la coupe forestière sur l’habitat hivernal de l’orignal dans le sud-ouest du Québec. Canadian Journal of Forest Research 7:241-257. _____. 1988. Forestry practices in Que- bec and Ontario in relation to moose population dynamics. Forestry Chroni- cle 64:246-250. DUSSAULT, C. 1999. Inventaire aérien de l’orignal (Alces alces) dans la zone de chasse 18 ouest à l’hiver 1998. Québec Ministère de l’Environnement et de la F a u n e , D i r e c t i o n r é g i o n n a l e d u Saguenay-Lac-Saint-Jean. EASON, G. 1989. Moose response to hunting and 1 km2 block cutting. Alces 25:63- 74. _____, E. THOMAS, R. JERRARD, and K. OSWALD. 1981. Moose hunting closure in a recently logged area. Alces 17:111- 125. ESRI. 1996. ArcView GIS. Environmen- tal Systems Research Institute Incor- porated, Redlands, California, USA. GIRARD, F., and R. JOYAL. 1984. L’impact des coupes à blanc mécanisées sur l’orignal dans le nord-ouest du Québec. Alces 20:3-25. GRONDIN, P. 1996. Écologie forestière. Pages 133-279 in Manuel de Foresterie. Presses de l’Université Laval, Québec, Québec, Canada. GROSENBAUGH, L. R. 1952. Plotless timber estimates - new, fast, easy. Journal of Forestry 50(1):32-37. MANLY, B., L. MCDONALD, and D. THOMAS. 1993. Resource Selection by Animals. Statistical Design and Analysis for Field Studies. Chapman and Hall, London, U.K.