4011.p65 ALCES VOL. 40, 2004 MCLAREN ET AL. - OVERABUNDANT MOOSE IN NEWFOUNDLAND 45 EFFECTS OF OVERABUNDANT MOOSE ON THE NEWFOUNDLAND LANDSCAPE Brian E. McLaren1,2, Bruce A. Roberts3, Nathalie Djan-Chékar4, and Keith P. Lewis5 1 Government of Newfoundland and Labrador, Department of Natural Resources, P.O Box 2222, Gander, NL, Canada A1V 2N9; 3 Natural Resources Canada, Canadian Forest Service, P.O. Box 960, Corner Brook, NL, Canada A2H 6J3; 4 Provincial Museum of Newfoundland and Labrador, Natural History Unit, P.O. Box 8700, St. John’s, NL, Canada A1B 4J6; 5 Programme in Cognitive and Behavioural Ecology, Departments of Biology and Psychology, Memorial University, St. John’s, NL, Canada A1C 5S1 ABSTRACT: The long-term effects of introduced and overabundant herbivores on community development must be monitored and managed in an ecosystem-based forest management approach. This paper builds on previously published ecological descriptions and hypotheses offered on the effects of moose overabundance in Newfoundland. The island, in the absence of wolves, provides a setting for study of local irruptions in moose populations, which now affect an increasing area of the forest. Moose effects occur most often after natural disturbances and logging, involving unique forest succession patterns. We describe some of these changes, along with anticipated and realised changes in associated forest biodiversity. We offer suggestions to improve or refine monitoring of moose populations, especially at local scales, to detect cases of overabundance. Finally, we offer recommendations for the management of overabundant moose populations. ALCES VOL. 40: 45-59 (2004) Key words: Alces alces, biodiversity, carrying capacity, conservation biology, disturbance, moose, Newfoundland, overabundance, population dynamics The long-term effects of introduced and overabundant herbivores on forest com- munity development must be monitored and managed in an ecosystem-based approach to forestry and wildlife interests. Adapting forest management to shifting baselines created by the effects of overabundant herbivores, especially in eastern North America, increasingly defeats the interest of biodiversity protection (Lindenmayer and Franklin 2003). Logging in the boreal forest and other forest management tailored to the spatial scale and frequency of wildfires or insect-related tree mortality are recom- mended to protect and/or restore ecological integrity (Hunter 1993, Niemelä 1999), as a minimum way of recognizing the adaptation of diverse organisms to the forests they occupy, by considering their disturbance regimes. Disturbance is usually defined as a rapid release or reallocation of resources in a forest community (White and Jentsch 2001), thereby possibly ignoring more gradual changes to forest community devel- opment caused by irruptive population phases or overpopulations of herbivorous mammals, especially deer (Cervidae). Such changes can be considerable, unpredict- able, and relatively irreversible (Davidson 1993, Côté et al. 2004). In this paper, we offer a case study of introduced moose (Alces alces andersoni), 2 Present address: Faculty of Forestry and the Forest Environment, Lakehead University, 955 Oliver Road, Thunder Bay, ON, Canada P7B 5E1 OVERABUNDANT MOOSE IN NEWFOUNDLAND – MCLAREN ET AL. ALCES VOL. 40, 2004 46 its overabundance, and its enduring effect on biodiversity in specific areas of boreal forest on the island of Newfoundland. We show that moose are significantly influenc- ing several aspects of some ecosystems, including forest succession and composi- tion, soils, and other wildlife. Previously it has been shown that moose are capable of producing negative economic effects on the forests of Newfoundland (Pimlott 1963, Thompson 1988), but such studies are often limited just to a portion of a forest rotation (Thompson and Curran 1993, McLaren et al. 2000a). This paper builds on published ecological descriptions and hypotheses of- fered for longer-term effects of moose over- abundance in Newfoundland (e.g., Bergerud and Manuel 1968, Thompson and Mallik 1989, Thompson et al. 1992). We make recommendations toward monitoring, con- servation planning, and management of overabundant moose from the perspectives of past examples and anticipated future challenges in Newfoundland. Moose in Newfoundland: A Background The island of Newfoundland, Canada, is a landmass of 112,000 km2 in the north- west Atlantic, of which about two-thirds is forested and/or qualifies as excellent moose habitat. Moose were introduced to central Newfoundland in 1878 with the release of a male and female from nearby Nova Scotia (Pimlott 1953). A second release of two males and two females from New Bruns- wick, into western Newfoundland, followed in 1904. Moose rapidly colonized New- foundland (Fig. 1), as they exploited new habitat and as wolves (Canis lupus) were e x t i r p a t e d ( P i m l o t t 1 9 5 9 ) . C a r i b o u (Rangifer tarandus terra-novae) are the only other ungulate in Newfoundland, pri- marily occupying non-forested habitat and existing prior to the arrival of Europeans. Where the two cervid species now co-exist, there has been no recorded direct competi- tion between them. Today, moose occupy all ecoregions on the island, at densities in primarily forested habitat in many instances exceeding 4 moose / km2 (> 1,000 kg / km2). Population reconstruction and aerial survey estimates for a moose management unit normally do not show this finer-scale spatial variation in density, but overabundance is suggested by our temporal series (Fig. 1), considering that approximately 75,000 km2 of habitat supported on average about 2 moose / km2 during two periods, in the late 1950s and late 1980s (Mercer and McLaren 2002). This average is met by considerable variation and any densities > 2 moose / km2 are considered above management targets (Newfoundland and Labrador Inland Fish and Wildlife Division, unpublished). The national parks in Newfoundland, where hunt- ing is prohibited, form special cases of over- abundance. The second and likely only successful moose introduction to Newfound- land was ca. 20 km from what is now the boundary of Gros Morne National Park, in the western part of the island. Local hunt- ing kept the population relatively low until this activity ceased with the establishment of the park in 1974. Within the park area of 1,805 km2, of which only ca. 30% is suitable habitat, moose have increased steadily from some 1,000 animals to > 7,000, with local densities as high as 7 moose / km2 by 1995 (McLaren et al. 2000b). Management of moose hunting in New- foundland, which began with the first sea- son in 1945 (Mercer 1995), has achieved a legal kill of about half a million, the majority of which is by resident hunting (McLaren 2004). Annual license issue since 1990 has been between 20,000 and 25,000 resident and non-resident tags combined, with a roughly equal number of either-sex and male-only licences issued (Mercer and McLaren 2002). Annual kill estimates, in- corporating poaching, crippling, and high- ALCES VOL. 40, 2004 MCLAREN ET AL. - OVERABUNDANT MOOSE IN NEWFOUNDLAND 47 way losses, have been 20,000 - 30,000 moose for several years. First Nations are cur- rently included in the resident licensing sys- tem in the island portion of Newfoundland and Labrador. Dating at least to 1934, the provincial government has recognized that sport hunting is a major attraction for visitors and the success of the moose intro- duction to the island has often been ap- plauded as a source of tourist revenue. Non-residents obtain about 10% of moose tags issued, and owing to the higher success offered by outfitters and guides, are respon- sible for > 10% of the annual moose kill. Newfoundland is also home to a dispro- portionately large part of the North Ameri- can moose population. The island popula- tion, at 125,000 moose, represents > 10% of the total continental number of moose (1.05 million), while the total island area, including areas unsuited to moose, is < 2% of the estimated continental moose range (6.44 million km2). Throughout North America, density of moose and other deer species varies according to four main factors: the availability of habitat, the availability of alternate foods created by agriculture, man- agement of hunting, and the presence of limiting factors like aridity or natural preda- tors (Crête and Daigle 1999). Overpopula- tion usually occurs following introduction into unexploited habitat and persists in situ- ations without natural predators (McShea et al. 1997), as is the case in Newfoundland. In these situations, moose appear to be limited primarily by the productivity of the boreal forest, as described for Québec by Crête and Courtois (1997). As productivity varies, so does the effect of moose, as expected in an unregulated trophic system. Thus, we have two arguments for moose overpopulation in Newfoundland: (1) as presented by Crête and Daigle (1999), New- foundland hosts an anomalous deer biomass compared with the rest of the continent, presumably because of the absence of wolves; and (2) as described by Mercer and McLaren (2002), a stable equilibrium be- tween the population and food resources does not appear to have occurred for New- Fig. 1. Moose population trends in insular Newfoundland since introduction, using McLaren’s (2004) assessment of the successful point of introduction, Pimlott’s (1959) estimate of increase rate during 1904-44 (solid squares), hunter success during 1945-65 (open circles), and hunter reports of moose seen during 1966-99 (solid circles). These estimates, scaled to consolidated aerial survey estimates in the 1980s and in the 1990s (Newfoundland and Labrador Inland Fish and Wildlife Division, unpublished), occur in a wide range of habitats and densities. OVERABUNDANT MOOSE IN NEWFOUNDLAND – MCLAREN ET AL. ALCES VOL. 40, 2004 48 foundland, particularly problematic where moose densities are higher, in more produc- tive areas, and/or in areas less accessible to hunting. Newfoundland thus allows us to illustrate specific cases where ecosystem management experiences new challenges as a result of moose overabundance. Effects of Moose on the Forest Ecosys- t e m Forest succession and composition — Most forests in Newfoundland consist of a c o m b i n a t i o n o f b a l s a m f i r ( A b i e s balsamea) and spruce (Picea spp.), with some pioneer and shade tolerant hardwoods. Fir is dominant in older forests, while insect outbreaks, fires, and logging have been fre- quent forest disturbances creating a gener- ally young-forest landscape. Since the ar- rival of moose, their consumption of balsam fir and hardwoods has affected forest re- generation following disturbance, particu- larly along edges and roadsides (Bergerud and Manuel 1968). In these affected for- ests, spruce and larch (Larix laricina) grow normally, since they are species not nor- mally found in moose diets, but balsam fir, a heavily-consumed species, can be prevented from reaching heights > 1 m. The resulting open ecotype has been described but not attributed specifically to moose in any gen- eral literature on forest management in New- foundland; the forest succession leading to the ecotype has been termed “old-field spruce succession” (Damman 1964). In updating the forest site classification (ecological land classification) for New- foundland, Roberts and Bajzak (1984) used the term “ungulate induced” to describe the succession specific to richer sites, in which a shift occurs from closed canopy balsam fir and white spruce (Picea glauca) to open-grown white spruce following distur- bances (Fig. 2). Roberts (1989a, 1989b) attributed this change to occupation of young forests by overabundant moose during the 1960s. Roberts’ (1989b) concern was for white spruce associations and other rare forest types. Similar examples of vulner- able forest types are yellow birch (Betula alleghaniensis) associations on the Avalon Peninsula and red maple (Acer rubrum) associations in central Newfoundland. These types, occurring with balsam fir, are fre- quently subject to logging and wind distur- bance and occur in management areas where moose are kept near target densities to offer steady hunting opportunities. Yellow birch and red maple are preferred species in a moose diet dominated by fir. In time, additional changes to natural forest succession caused by overabundant moose may become apparent. Since the 1980s, the total area disturbed by logging, as well as the secondary road network, has increased as forestry has become more extensive in Newfoundland. More common ecological associations may be threatened by the coexistence of regenerating com- mercial forest and overabundant moose, especially with declining interest in hunting. Less accessible areas may be especially prone to a combination of natural distur- bance like insect outbreak and moose over- abundance. Areas of both natural and logging disturbance that fail to regenerate into closed canopy forest are already at a scale readily visible on aerial photographs (Fig. 2). Our photographs show white spruce associations. Thompson et al. (1992) and Thompson and Mallik (1989) extended their concerns to black spruce (Picea mariana) associations. Observations of forest composition change attributed to moose overabundance have been recorded in the national parks. Terra Nova National Park, 344 km2, has been unaffected by logging since the 1950s, but is a special case of forest disturbance, in which small (< 1 km2) landscape patches were disturbed by insect outbreaks in the late 1970s, just before a peak in moose ALCES VOL. 40, 2004 MCLAREN ET AL. - OVERABUNDANT MOOSE IN NEWFOUNDLAND 49 Fig. 2. Ungulate induced changes to the New- foundland landscape. The aerial photographs show cases where forest succession is inter- rupted by overabundant moose for (A) an area near Blue Hill in Terra Nova National Park that was disturbed by an insect outbreak in 1978 and has failed to regenerate in 25 years (photo date 1996), (B) a partially forested watershed in 1988, logged in the 1950s, where balsam fir has failed to regenerate in a white spruce association, following disturbances that also included insect outbreaks, and (C) an area near Halfway Cliff in Gros Morne National Park, where moose have created semi-open cover because they prevent a fir-dominated alpine forest understory from regenerating. The partial canopy in the three photographs (at arrows) is created by open-grown spruce. The scale bars in each photograph measure ca. 1 km. density during the 1980s. Moose density is low today, ca. 0.7 / km2, but very low recruitment observed in all of the last 5 midwinter surveys suggests a declining popu- lation, exceeding carrying capacity (Gosse et al. 2002). Carrying capacity in the park has itself declined over time as forests have matured and moose have nearly completely removed hardwoods, such as red maple and mountain ash (Sorbus americana) from mature-forest understories. In the forage- l i m i t e d , d i s t u r b e d p a t c h e s , p i o n e e r hardwoods, such as white birch (Betula spp.), aspen (Populus tremuloides), and pin cherry (Prunus pensylvanica), have also been affected. To illustrate the effect moose continue to have on limiting growth and survival of understory trees, park man- agers constructed several fenced areas or “exclosures” (Terra Nova National Park, unpublished). After only 3 growing sea- sons, red maple density inside exclosures was up to 3 times higher than outside. For white birch, similar densities occur inside and outside, because stump and root sprout- ing is common. However, growth of white birch has been affected. For example, up to 90% of white birch stems are > 0.5 m inside the exclosures (this includes 14% > 1 m after 3 growing seasons), but outside, only 5% are > 0.5 m. Outside the exclosures, reductions in stem density of 2–17% for balsam fir also occurred between 1999 and 2003. Measured annually, this effect was highest when snow cover was low and moose were able to uproot young stems. In C B A OVERABUNDANT MOOSE IN NEWFOUNDLAND – MCLAREN ET AL. ALCES VOL. 40, 2004 50 Gros Morne National Park, Canada yew (Taxus canadensis), once common in coastal forest associations (Robertson and Roberts 1982), is nearly completely removed from the forest understory, presumably by moose. In addition, areas with once fairly closed forest now have frequent understory openings and associated changes in forest structure (Connor et al. 2000). Experimental introduction of moose to offshore islands in eastern Newfoundland has provided other examples of their capac- ity to change forest composition very soon after arrival. On Brunette Island, off the south coast, several moose were introduced for the first time in 1974 and plant measure- ments using exclosures to assist compari- sons were undertaken during the 10 follow- ing years (Butler 1986). In this study, the ratio in annual production of balsam fir to faster growing hardwoods declined during the first 5 years from 39:1 to 15:1, measured in stems / ha (Wildlife Division, unpublished data). Less common species, like mountain ash and wild raisin (Viburnum cassinoides), declined faster than balsam fir as a ratio to the fastest growing hardwoods in the same period. These measurements were cor- rected for the “natural” succession changes observed inside the exclosures. The re- moval of forbs, grasses, and alder (Alnus spp.), not normally observed in retrospec- tive studies, was significant in observations of a tame moose, which consumed these plants in > 80% of observed bites in a summer period (Butler 1986). Recognizing the earliest changes created by an intro- duced species, often difficult to quantify, is important to biodiversity management. Soils — A review of the effects of herbivores on multi-trophic interactions in- cluding soil effects has recently been pro- vided by Bardgett and Wardle (2003). Pas- tor and Naiman (1992) discussed the sub- ject for moose. We provide one unique Newfoundland example. In many moun- tainous, serpentine plateaus in Gros Morne National Park, toxic levels of magnesium and heavy metals occur in soils (Roberts 1980, 1992). Before 1980, moose were rarely encountered in these areas of the park, but their use of the serpentine pla- teaus is now prevalent (B. A. Roberts, unpublished observations). In plants adapted to the specialized plateau ecosystems, such as dwarf birch (Betula glandulosa) and alder (Alnus crispa), nickel concentrations range from 48–77 ppm (Roberts 1992, Roberts and Proctor 1992). While not nor- mally including these species in their diet, moose are now consuming these toxic plants, presumably because of the lack of pre- ferred forage species, a consequence of overabundance. This interaction may ulti- mately be detrimental both to the animals and to the long-term persistence of some plant species. Our concern for landscape effects is that even the sparse cover of plants acts as a soil stabilizer. As moose consume and trample slowly growing plants, the rate of soil erosion increases, as once described by Leopold (1949), in “thinking like a mountain.” Other forest wildlife — A review of deer overabundance in many parts of the world and its cascading (indirect) effects on other plant and animal species is provided by Côté et al. (2004). Changes to forests caused by overabundant moose in New- foundland are also very likely to affect many more forest-dwellers than the plants directly affected by browsing. For exam- ple, forest songbirds dependent on hard- wood and balsam fir trees (Setterington et al. 2000) and epiphytic tree lichens with specific habitat requirements (Yetman 1999) may actually be eliminated from portions of the landscape with overabundant moose. Alternatively, ungulate-induced modifica- tion of forest structure and composition after natural disturbance or logging may result in changes to habitat selection among ALCES VOL. 40, 2004 MCLAREN ET AL. - OVERABUNDANT MOOSE IN NEWFOUNDLAND 51 forest-dependent species. Our examples below further illustrate these hypotheses. Lichen diversity is generally related to the availability of different microhabitats (Gustafsson et al. 1992, Kuusinen 1995, Rosentreter 1995, Neitlich and McCune 1997). A mixture of deciduous and conifer- ous tree species provides an array of bark acidity and texture, offers diverse trunk structures, and creates a mosaic of mois- ture and light conditions for lichen growth. Consequently, altered forest composition and structure caused by moose, such as the elimination of hardwoods, may have an in- direct effect on epiphytic lichen community composition. This effect was illustrated through principal component analysis (PCA) of lichen cover in various forest types of Terra Nova National Park (Yetman 1999). Red maple in particular, in stands mixed with balsam fir, supports a unique lichen community according to the PCA, in which the principle components correspond to site richness and bark acidity. As discussed above, maple is one of the trees being limited in both density and height growth as a result of moose in the park. An end result may be the loss of the lichen community this tree supports. In the Avalon Peninsula, yellow birch is known to be a specific host for the rare cyanolichen, Degelia plumbia. As yellow birch is selectively browsed in this area, its subsequent failure to regener- ate may limit opportunities for this host- specific epiphyte. B o r e a l f e l t l i c h e n , E r i o d e r m a pedicellatum, a globally rare species, is now restricted to coastal Nova Scotia and the island of Newfoundland. It was re- cently listed as vulnerable under the New- foundland and Labrador Endangered Spe- cies Act. A species of oceanic affinity, it is found in moist, cool forests where it grows predominantly on balsam fir and forms part of a characteristic cyanolichen community (Ahti 1983, Maass 1983). The known New- foundland population of boreal felt lichen is concentrated in the Avalon Peninsula and Bay d’Espoir areas. The Avalon Peninsula has had a long history of land development, logging, insect outbreaks, wind disturbance, and fires; it is also an area frequently used by moose in winter. The percentage of juvenile lichens is much lower in this area than in the second area, where forest cover is more complete and moose are less abun- dant (N. Djan-Chékar, unpublished data). We speculate that in the Bay d’Espoir area more opportunity for lichen colonization exists as a result of fewer moose and better balsam fir regeneration. To assess this hypothesis, biologists now monitor both the boreal felt lichen by recording its occur- rence and abundance on the landscape, and also the use by moose of its critical habitat by annually counting pellet groups, an eco- system approach to biodiversity monitoring. Changes to vertebrate communities as a result of forest succession are probably best understood for birds (Helle and Niemi 1996). Plant structure and diversity influ- ence avian assemblages (e.g., MacArthur and MacArthur 1961, MacArthur et al. 1962, James and Rathbun 1981, Cody 1985, Willson and Comet 1996), foraging behav- iour (Parrish 1995), and nest site selection (Martin 1992). In addition, predation of nests can increase in areas with low foliage density (Martin 1993). However, the indi- rect effects that herbivores exert on avian assemblages through modifying vegetation are not well known (Rotenberry et al. 1995, McShea and Rappole 1997) and only a few researchers have explicitly studied the ef- fects of herbivores on birds (DeGraaf et al. 1991, Popotnik and Giuliano 2000). Indirect effects of moose overabun- dance on songbirds in the forest–heathland ecotone in central Newfoundland were con- sidered in a model of avian richness and abundance in black spruce-feathermoss forest and Kalmia angustifolia heath. Fire OVERABUNDANT MOOSE IN NEWFOUNDLAND – MCLAREN ET AL. ALCES VOL. 40, 2004 52 suppression, logging, and plant consumption by moose all contribute to a conversion from black spruce-feathermoss forest to transitional black spruce-Kalmia forests and, in cases of very poor regeneration, Kalmia heath (cf. Thompson and Mallik 1989, Thompson et al. 1992). A forest inventory describing the extent of this conversion has not been completed but the problem has demanded attention by silviculturists in the province (English and Hackett 1994). We observed that songbird abundance and spe- cies richness is significantly lower in Kalmia heath compared to black spruce-Kalmia forest (Lewis 2004). Most birds in Kalmia heath tended to be habitat generalists and were also common in the forests. Species abundance increased with increasing verti- cal structure in the Kalmia heath, indicating the importance of fire skips and snag reten- tion, current elements of forest manage- ment. For example, Common Yellow Throat (Geothlypis trichas) and Lincoln’s Spar- row (Melospiza lincolni) were associated with the fire skips. We were able to con- clude that the continued suppression of many plant species by overabundant moose, as well as the invasiveness of Kalmia with logging and disruption of a natural fire re- gime, has variable but significant effects. Initial conversion of forest to black spruce- Kalmia transition types will result in in- creased songbird abundance and species richness. However, if severe restriction of regeneration in black spruce forest contin- ues and heath increases in area, moose will have contributed to an increasingly impov- erished avifauna. Monitoring Moose Overabundance Given that herbivores can influence forested ecosystems in a variety of ways, it is important to have programs to monitor their abundance and effects. Increasingly, moose management plans must ask for lo- cal knowledge and advice on appropriate mitigative measures in cases of overabun- dance. We offer some suggestions to im- prove or refine monitoring of moose populations, especially at local scales, in Newfoundland and in other jurisdictions where overabundance may be a concern. Improved use of aerial surveys — Surveys are not generally useful in assess- ing moose overabundance, because they are prioritized to areas of low, not high abundance, providing information to justify changes to licence quotas, the largest con- cern of hunters. In addition, survey areas are often much larger than areas of local overabundance. However, data from past surveys are often readily available and their innovative use, such as for identification of very high density areas or for location of census blocks of high individual counts, can be made to assess local overabundance and to verify interpretations from land capabil- ity indices. Land inventories — Management for forest sustainability must recognize land capability as the critical long-term factor determining productivity of any biological species. Land inventories can assist in ecosystem management at various scales. Most readily available maps are based on timber inventory, i.e., predicting economic value of the trees, and have limited applica- bility to predicting wildlife habitat (e.g., Proulx and Joyal 1981, Potvin et al. 1999, McLaren and Mahoney 2001). However, the soil-based Canada Land Inventory (CLI) is a more comprehensive classification that includes specific references to the habitat requirements of deer and other wildlife. It offers ecological comparisons of areas with varying forest capability and consequently shows relative capability to support ungu- late populations. The inventory can become a reference for predicting “ungulate-in- duced” changes to forest succession on the landscape. For example, Dryopteris–Lyco- podium-balsam fir and Hylocomium–bal- ALCES VOL. 40, 2004 MCLAREN ET AL. - OVERABUNDANT MOOSE IN NEWFOUNDLAND 53 sam fir, classed uniquely in the CLI, are the dominant types where stand conversion to white spruce has taken place as a result of disturbance and overabundant moose (Roberts 1989a, b). Pellet-group counts — Neff (1968) first described pellet-group (deer defeca- tion) counts as a means of assessing local deer density. Jordan et al. (1993) advise that pellet-group counts for moose are an effective means to make relative density comparisons over short periods and that long-term averages also compare well to information obtained from aerial surveys. The fact that moose can migrate seasonally, sometimes over long distances (McLaren et al. 2000b), may make pellet groups a more relevant index of local overabundance than an aerial survey conducted in one season. Important factors in designing a monitoring program using pellet-group counts are to replicate counts in each area of interest and to conduct counts at the same time in successive years (Jordan et al. 1993). Pellet-group counts are subjective, because group definition is variable be- tween observers and the season in which defecation occurs in a pellet form is vari- able between years and individuals. Pellet groups also preserve longer in dry relative to humid soil conditions that vary with to- pography. Simple attempts at calibration to correct for such errors are likely to be unsuccessful. Browse surveys — Often, plant-based protocols for monitoring herbivore effects are too labour-intensive or they include in- accurate assessments like browsing “se- verity indices” determined by visual inspec- tion. Indices designed for one plant archi- tecture may not apply well to another. Telfer (1967, 1972) advises on more accurate measurement of forage yield and browsing effects based on twig counts. Literature on optimal foraging theory (e.g., Gross et al. 1993) offers additional advice. Dendrochronology — Roberts (1989a) used dendrochronology to describe “old- field spruce succession” by means of sam- pling balsam fir trees from several areas within one forest type to measure stem growth and ring width. From this work, a specific period of moose overabundance was determined. Dendrochronology has since been applied to several studies of the long-term effects of mammals on ecosys- tems (e.g., Sinclair et al. 1993, McLaren and Peterson 1994). The technique can provide an accurate description of the cu- mulative effects of moose consumption on tree or shrub biomass (McLaren and Peterson 1996). Experimental exclosures — Several ecologists have constructed exclosures (fences to keep mammalian herbivores out of an experimental area, e.g., McInnes et al. 1992). Exclosures may not produce immediate changes in forest succession but they may reveal other ecological effects. Changes in the trophic pyramid may be most apparent at levels directly above and below the plants, i.e., in soil or in the abun- dance of herbivores that are not excluded by the fence. For example, McInnes et al. (1992) gained insight into effects of moose on the boreal forest by measuring not only browsing of trees and shrubs, but also changes to leaf litter. There are several cautions in designing an exclosure system: 1. The fence itself introduces ecological effects. Attracted by the forage regenera- tion inside the exclosure, herbivores may circle the fence and cause more extreme damage to plants adjacent to it. Conse- quently, the “control” (unfenced) monitor- ing area should be well outside the fence. As vegetation in the area outside the fence continues to be browsed, it may retain an open or semi-open canopy depending on conditions at the time of fence construction. Plants just inside the fence thus benefit OVERABUNDANT MOOSE IN NEWFOUNDLAND – MCLAREN ET AL. ALCES VOL. 40, 2004 54 from a higher light supply than in the ab- sence of herbivores and their measurement would constitute a bias. The “experimen- tal” (fenced) monitoring area should there- fore be located far enough inside the fence to eliminate this bias. Finally, the fence material should be chosen so that it pro- duces minimum soil change; e.g., galva- nized material adds toxic zinc to soils. Crib- bing works well on fence posts in shallow soil. 2. Sufficient replication to control for differ- ences in site history and productivity is often difficult, both because of the difficulty in placing exclosure sites randomly and because of the expense of fence construc- tion and maintenance. An obvious compro- mise is to construct larger exclosures; we suggest that 35 m on a side is minimum construction because it allows at least a 500-m2 unbiased sampling area. Larger exclosures may be required to incorporate topographic variation. Numerous plots in- side exclosures may be unnecessary and they do not provide real landscape-scale replication. 3. Monitoring protocols are often too ambi- tious or flawed. Good advice is to begin fence construction only after very specific research hypotheses have been outlined. 4. Exclosures will not mimic the forest succession that would occur as if the her- bivore had never been present in the eco- system. While this statement is a truism, it is often ignored in discussions of the results of exclosure monitoring. For example, trees regenerating inside an exclosure often grow vigorously from rhizomes, stump sprouts, or layering. Such vegetative reproduction is obviously enhanced when herbivores domi- nate an ecosystem for many years allowing plants to allocate biomass below ground or to lateral branch growth. Meanwhile, seed fall into an exclosure declines when plants outside the exclosure are not replaced be- cause herbivores continue to suppress api- cal or floral growth, and seeds with longer persistence will have a competitive advan- tage inside the exclosure area. Local ecological knowledge — For- est managers rely on conservation officers, field technicians, and the general public, especially hunters and naturalists, to collect new field observations. Access to wilder- ness areas and our ability to change them through industry, sport, and deliberate or accidental species introductions increase in tandem. Thus, we have two related reasons to improve the reporting and analysis of local ecological knowledge. In many in- stances, descriptions of changing forest structures or landscapes through natural succession can only be made following di- rect field observations repeated over many years, usually by naturalists and field tech- nicians (e.g., Robertson and Roberts 1982, Crête et al. 2001). Good examples of the systematic collection of this information already exist (e.g., Ecological Monitoring and Assessment Network, www.eman- rese.ca, Bird Map Canada, www.bsc- eoc.org/birdmap_e.htm, etc.). These can serve as baselines for new or improved monitoring systems. Similarly, diligent re- porting by hunters has allowed efforts to summarize overabundance at a continental scale (e.g., Crête and Daigle 1999), or at local scales (e.g., Mercer and McLaren 2002). Local ecological knowledge is in- creasingly incorporated into research and management plans (e.g., Ferguson and Messier 1997). Moose in Newfoundland: Some Man- agement Recommendations Côté et al. (2004) challenged ecologists and wildlife managers to reduce deer num- bers before and not after long-term impacts become difficult to reverse. Angelstam et al. (2000) specifically reviewed manage- ment issues involving high moose densities. Mercer (1995: 92) correctly took the posi- ALCES VOL. 40, 2004 MCLAREN ET AL. - OVERABUNDANT MOOSE IN NEWFOUNDLAND 55 tion for Newfoundland that more emphasis be placed on relationships between hunter moose-kill and moose density to stabilize populations. In areas of similar habitat, hunter kill density should be similar and proportional to moose density. In one area of Newfoundland, local moose overabun- dance was effectively managed by tempo- rarily directing resident hunting into the area (McLaren et al. 2000a). In other areas, where resident hunter demand is not high, or in less accessible areas, non-resi- dent hunters might be encouraged to use permanent or temporary hunting camps. Other innovations that may be part of future management of overabundant moose in Newfoundland include implementation of a commercial hunt for the restaurant and luxury export trade and targeted areas of reduction, currently being considered for the national parks. CONCLUSION Considerable interest was generated by the ecological literature discussing when and where a species becomes a “keystone” for its effect on structuring ecosystems (Paine 1995, Power et al. 1996). We sug- gest that this discussion also be applied to introduced herbivores to determine where the effects of their introduction on the struc- ture and function of native communities accumulate to the extent that they can be described as “wrecking balls”. Understand- ing and predicting the continuous effects of herbivores on forest ecosystems following disturbance and managing these effects must be part of sustainable forest manage- ment. ACKNOWLEDGEMENTS The material in this paper was first presented at the Fourth International Work- shop on Disturbance Dynamics in Boreal Forests, University of Northern British Co- lumbia, Prince George, B.C., August 9–14, 2002. The authors wish to thank the Cana- dian Forest Service for support in preparing this paper, Joe Brazil of the Endangered Species Program and John Maunder of the Provincial Museum, Government of New- foundland and Labrador, for their early dis- cussions and review, and Gordon Eason, Richard Ward, and an anonymous reviewer for later critical reviews. REFERENCES AHTI, T. 1983. Lichens. Pages 319–360 in G. R. South, editor. Biodiversity and Ecology of the Island of Newfound- land. Junk Publishers, The Hague. ANGELSTAM, P., P.-E. WIKBERG, P. DANILOV, W. E. FABER, and K. NYGRÉN. 2000. Effects of moose density on timber qual- ity and biodiversity restoration in Swe- den, Finland, and Russian Karelia. Alces 36:133–145. BARDGETT, R. D., and D. A. WARDLE. 2003. Herbivore-mediated linkages between aboveground and belowground commu- nities. Ecology 84:2258–2268. BERGERUD, A. T., and F. MANUEL. 1968. Moose damage to balsam fir-white birch forests in central Newfoundland. Jour- nal of Wildlife Management 32:729– 746. BUTLER, C. E. 1986. Summer food utiliza- tion and observations of a tame moose, Alces alces. Canadian Field-Naturalist 100:85–86. CODY, M. L. 1985. Habitat Selection in Birds. Academic Press, San Diego, California, USA. CONNOR, K. J., W. B. BALLARD, T. DILWORTH, S. MAHONEY, and D. ANIONS. 2000. Changes in structure of a boreal forest community following intense herbivory by moose. Alces 36:111–132. CÔTÉ, S. D., T. P. ROONEY, J.-P. TREMBLAY, C. DUSSAULT, and D. M. WALLER. 2004. Ecological impacts of deer overabun- dance. Annual Review of Ecology, OVERABUNDANT MOOSE IN NEWFOUNDLAND – MCLAREN ET AL. ALCES VOL. 40, 2004 56 Evolution and Systematics 35:113–147. CRÊTE, M., and R. COURTOIS. 1997. Limit- ing factors might obscure population regulation of moose (Cervidae: Alces alces) in unproductive boreal forests. Journal of the Zoological Society of London 242:765–781. ______, and C. DAIGLE. 1999. Manage- ment of indigenous North American deer at the end of the 20th century in relation to large predators and primary production. Acta Veterinaria Hungarica 47:1–16. ______, J.-P. OUELLET, and L. LESAGE. 2001. Comparative effects on plants of caribou/reindeer, moose and white- tailed deer herbivory. Arctic 54:407– 417. DAMMAN, A. H. W. 1964. Some forest types of central Newfoundland and their relationship to environmental factors. Forest Science Monograph 8. DAVIDSON, D. W. 1993. The effects of herbivory and granivory on terrestrial plant succession. Oikos 68:23–35. DEGRAAF, R. M., W. M. HEALY, and R. T. BROOKS. 1991. Effects of thinning and deer browsing on breeding birds in New England oak woodlands. Forest Ecol- ogy and Management 41:179–191. ENGLISH, B., and R. HACKETT. 1994. The impact of Kalmia on plantation per- formance in central Newfoundland. Silviculture Notebook No. 2, Govern- ment of Newfoundland and Labrador, St. John's, Newfoundland, Canada. FERGUSON, M. A. D., and F. MESSIER. 1997. Collection and analysis of traditional ecological knowledge about a popula- tion of arctic tundra caribou. Arctic 50:17–28. GOSSE, J., B. MCLAREN, and E. EBERHARDT. 2002. Comparison of fixed-wing and helicopter searches for moose in a mid- winter habitat-based survey. Alces 38:47-53. GROSS, J. E., L. A. SHIPLEY, N. T. HOBBS, D. E. SPALINGER, and B. A. WUNDER. 1993. Functional response of herbivores in food-concentrated patches: tests of a mechanistic model. Ecology 74:778– 791. GUSTAFSSON, L., A. FISKESJÖ, T. INGELÖG, B. PETTERSSON, and G. THOR. 1992. Fac- tors of importance to some lichen spe- cies of deciduous broad-leaved woods in southern Sweden. Lichenologist 24:255–266. HELLE, P., and G. J. NIEMI. 1996. Bird community dynamics in boreal forests. Pages 209–234 in R. M. DEGRAAF and R. I. MILLER, editors. Conservation of Faunal Diversity in Forested Land- scapes. Chapman and Hall, Norwall, Massachusetts, USA. HUNTER, M. L., JR. 1993. Natural fire regimes as spatial models for managing boreal forests. Biological Conservation 65:115–120. JAMES, F. C., and S. RATHBUN. 1981. Rar- efaction, relative abundance, and diver- sity of avian communities. Auk 98: 785–800. JORDAN, P., R. O. PETERSON, P. CAMPBELL, and B. MCLAREN. 1993. Comparison of pellet counts and aerial counts for esti- mating density of moose at Isle Royale, a progress report. Alces 29:267–278. KUUSINEN, M. 1995. Epiphytic lichen diver- sity on Salix caprea and Populus tremula in old-growth forests of Fin- land. Mitteilungen der Eidgenössischen Forschungsanstalt für Wald, Schnee und Landschaft 70:125–132. LEOPOLD, A. 1949. A Sand County Alma- nac and Sketches Here and There. Oxford University Press, London, Eng- land. LEWIS, K. P. 2004. Processes Underlying Nest Predation by Introduced Red Squir- rels (Tamiasciurus hudsonicus) in the Boreal Forest of Newfoundland. Ph.D. ALCES VOL. 40, 2004 MCLAREN ET AL. - OVERABUNDANT MOOSE IN NEWFOUNDLAND 57 Thesis, Cognitive and Behavioural Ecol- ogy Programme, Memorial University of Newfoundland, St. John’s, New- foundland and Labrador, Canada. LINDENMAYER, D., and J. FRANKLIN. 2003. Towards Forest Sustainability. Island Press, Washington, D.C., USA. MAASS, W. 1983. New observations on Erioderma in North America. Nordic Journal of Botany 3:567–576. MACARTHUR, R. H., and J. W. MACARTHUR. 1961. On bird species diversity. Ecol- ogy 42:594–598. _____, _____, and J. PREER, JR. 1962. On bird diversity. II. Prediction of bird census from habitat measurements. American Naturalist 46:167–174. MARTIN, T. E. 1992. Breeding productivity considerations: what are the appropri- ate habitat features for management? Pages 455–473 in J. M. HAGAN and D. W. JOHNSTON, editors. Ecology and Con- servation of Neotropical Migrant Landbirds. Smithsonian Institution Press, Washington, D.C, USA. _____. 1993. Nest predation and nest sites, new perspectives on old patterns. BioScience 43:523–532. MCINNES, P. F., R. J. NAIMAN, J. PASTOR, and Y. COHEN. 1992. Effects of moose browsing on vegetation and litter of the boreal forest, Isle Royale, Michigan. Ecology 73:2059–2075. MCLAREN, B. E. 2004. Social and natural history of moose introduced to New- foundland in A. J. Gaston, T. E. Golumbia, J.-L. Martin, and S. T. Sharpe, editors. Lessons from the Islands: In- troduced Species and What They Tell Us About How Ecosystems Work. Canadian Wildlife Service Occasional Paper. In Press. _____, and S. P. MAHONEY. 2001. Com- parison of forestry-based remote sens- ing methodologies to evaluate caribou habitat in non-forested areas of New foundland. The Forestry Chronicle 77:866–873. _____, _____, T. S. PORTER, and S. M. OOSENBRUG. 2000a. Spatial and tempo- ral patterns of use by moose of pre- commercially thinned, naturally-regen- erating stands of balsam fir in Central Newfoundland. Forest Ecology and Management 133:179-196. _____, C. MCCARTHY, and S. P. MAHONEY. 2000b. Extreme moose demographics in Gros Morne National Park, New- foundland. Alces 36:217–232. _____, and R. O. PETERSON. 1994. Wolves, moose, and tree rings on Isle Royale. Science 266:1555–1558. ______, and _____. 1996. Seeing the f o r e s t w i t h t h e t r e e s : u s i n g dendrochronology to investigate moose- induced changes to a forest understory. Alces 31:77–86. MCSHEA, W. J., and J. H. RAPPOLE. 1997. Herbivores and the ecology of forest understory birds. Pages 298–309 in W. J. MCShea, H. B. Underwood, and J. H. Rappole, editors. The Science of Over- abundance, Deer Ecology and Popula- tion Management. Smithsonian Institu- tion Press, Washington, D.C., USA. _____, H. B. UNDERWOOD, and J. H. RAPPOLE (editors). 1997. The Science of Over- abundance, Deer Ecology and Popula- tion Management. Smithsonian Institu- tion Press, Washington, D.C., USA. MERCER, W. E. 1995. Moose management plan for Newfoundland. Report on file with Inland Fish and Wildlife Division, Newfoundland and Labrador, Corner Brook, Newfoundland and Labrador, Canada. _____, and B. E. MCLAREN. 2002. Evi- dence of carrying capacity effects in Newfoundland moose. Alces 38:123– 141. NEFF, D. J. 1968. The pellet-group count technique for big game trend, census, OVERABUNDANT MOOSE IN NEWFOUNDLAND – MCLAREN ET AL. ALCES VOL. 40, 2004 58 and distribution, a review. Journal of Wildlife Management 32:541–545. NEITLICH, P. N., and B. MCCUNE. 1997. Hotspots of epiphytic lichen diversity in two young managed forests. Conser- vation Biology 11:172–182. NIEMELÄ, J. 1999. Management in relation to disturbance in the boreal forest. For- est Ecology and Management 115:127– 134. PAINE, R. T. 1995. A conversation on refining the concept of keystone spe- cies. Conservation Biology 9:962–964. PARRISH, J. D. 1995. Effects of needle architecture on warbler habitat selec- tion in a coastal spruce forest. Ecology 76:1813–1820. PASTOR, J., and R. J. NAIMAN. 1992. Selec- tive foraging and ecosystem processes in boreal forests. American Naturalist 139:690–705. PIMLOTT, D. H. 1953. Newfoundland moose. Transactions of the North American Wildlife Conference 18:563–581. _____. 1959. Reproduction and productiv- ity of Newfoundland moose. Journal of Wildlife Management 23:381–401. _____. 1963. Influence of deer and moose on boreal forest vegetation in two areas of Eastern Canada. Transactions of the 6th International Union of Game Biolo- gists Congress, Bournemouth, England. POPOTNIK, G. J., and W. M. GIULIANO. 2000. Response of birds to grazing of riparian zones. Journal of Wildlife Management 64:976–982. POTVIN, F., L. BÉLANGER, and K. LOWELL. 1999. Validité de la carte forestière pour décrire les habitats fauniques à l’échelle locale: une étude de cas en Abitibi-Témiscamingue. The Forestry Chronicle 75:851–859. POWER, M. E., D. TILMAN, J. A. ESTES, B. A. MENGE, W. J. BOND, L. S. MILLS, G. DAILY, J. C. CASTILLA, J. LUBCHENCO, and R. T. PAINE. 1996. Challenges in the quest for keystones. BioScience 46:609–620. PROULX, G., and R. JOYAL. 1981. Forestry maps as an information source for de- scription of moose winter yards. Cana- dian Journal of Zoology 59:73–80. ROBERTS, B. A. 1980. Some chemical and physical properties of serpentine soils from western Newfoundland. Cana- dian Journal of Soil Science 60:231– 240. _____. 1989a. The growth and develop- ment of white spruce (Picea glauca) in Newfoundland. Page 167 in The Silvics and Ecology of Boreal Spruces. Pro- ceedings of the 11th IUFRO Northern Forest Silviculture and Management Working Party S 1.05–12 Symposium, St. John's, Newfoundland, Canada. _____. 1989b. The origin and development of white spruce, Picea glauca, forest stands in central Newfoundland. Ameri- can Journal of Botany 76:138. _____. 1992. The serpentinized areas of Newfoundland, Canada. A brief review of their soils and vegetation. Pages 53– 66 in A. J. M. Proctor and R. D. Reeves, editors. The Vegetation of Ultramafic (Serpentine) Soils. Inter- cept Scientific Publication, France. _____, and D. BAJZAK. 1984. Forest site classification for the boreal forest of central Newfoundland, Canada (B.28a) using a bio-physical soils approach. Joint meeting of the Working Parties No. 1.02–06 and No. 1.02–10 of IUFRO on Q u a l i t a t i v e a n d Q u a n t i t a t i v e Asseessment of Forest Sites with Spe- cial Reference to Soil. Birmensdorf, Switzerland. _____, and J. PROCTOR, editors. 1992. The Ecology of Areas with Serpentinized Rocks, a World View. Kluwer Aca- demic Publishers, Netherlands. ROBERTSON, A., and B. A. ROBERTS. 1982. Checklist of the alpine flora of Western ALCES VOL. 40, 2004 MCLAREN ET AL. - OVERABUNDANT MOOSE IN NEWFOUNDLAND 59 Brook Pond and Deer Pond areas, Gros Morne National Park. Rhodora 84:101– 115. ROSENTRETER, R. 1995. Lichen diversity in managed forests of the Pacific North- w e s t , U S A . M i t t e i l u n g e n d e r Eidgenössischen Forschungsanstalt für Wald, Schnee und Landschaft 70:103– 124. ROTENBERRY, J. T., R. J. COOPER, J. M. WUNDERLE, and K. G. SMITH. 1995. When and how are populations limited? The roles of insect outbreaks, fire, and other natural perturbations. Pages 55– 84 in T. E. Martin and D. M. Finch, editors. Ecology and Management of Neotropical Migratory Birds. Oxford University Press, New York, USA. SETTERINGTON, M. A., I. D. THOMPSON, and W. A. MONTEVECCHI. 2000. Wood- pecker abundance and habitat use in mature balsam fir forests in Newfound- land. Journal of Wildlife Management 64:335–345. SINCLAIR, A. R. E., J. M. GOSLINE, G. HOLDSWORTH, C. J. KREBS, S. BOUTIN, J. N. M. SMITH, R. BOONSTRA, and M. DALE. 1993. Can the solar cycle and climate synchronize the snowshoe hare cycle in Canada? Evidence from tree rings and ice cores. American Natural- ist 141:173–198. TELFER, E. S. 1967. Comparison of a deeryard and a mooseyard in Nova Scotia. Canadian Journal of Zoology 45:485–490. _____. 1972. Forage yield and browse utilization on two logged areas in New Brunswick. Canadian Journal of Forest Research 2:346–350. THOMPSON, I. D. 1988. Moose damage to pre-commercially thinned balsam fir stands in Newfoundland. Alces 24:56– 61. _____, and W. J. CURRAN. 1993. A reexamination of moose damage to bal- sam fir – white birch forests of central Newfoundland: 27 years later. Cana- d i a n J o u r n a l o f F o r e s t R e s e a r c h 23:1388–1395. _____, _____, J. A. HANCOCK, and C. E. BUTLER. 1992. Influence of moose browsing on successional forest growth on black spruce sites in Newfoundland. Forest Ecology and Management 47:29– 37. _____, and A. U. MALLIK. 1989. Moose browsing and allelopathic effects of Kalmia angustifolia on balsam fir re- generation in central Newfoundland. Canadian Journal of Forest Research 19:524–526. WHITE, P. S., and A. JENTSCH. 2001. The search for generality in studies of dis- turbance and ecosystem dynamics. Progress in Botany 62:399–450. WILLSON, M. F., and T. A. COMET. 1996. Bird communities of northern forests, ecological correlates of diversity and abundance in the understory. Condor 98:350–362. YETMAN, D. 1999. Epiphytic Lichen Diver- sity and Abundance Based on Forest Stand Type in Terra Nova National Park: Implications for Lichen Conser- vation and Forest Management, B.Sc. Thesis, Department of Biology, Memo- rial University of Newfoundland, New- foundland and Labrador, Canada.