Coastal Barrens of Nova Scotia
J. Lundholm, with help from E. Oberndorfer and S. Burley.
We have been working on coastal barrens in Nova Scotia since 2004. Coastal barrens occur mainly along the Atlantic coast of Nova Scotia, but are isolated from one another by forested habitats and by development. I was surprised to find almost nothing published about these habitats when I relocated from Ontario, so our lab began the first quantitative descriptions of coastal barrens in Nova Scotia (Oberndorfer and Lundholm 2009). This “paper” summarizes what we know about the ecology and natural history of this habitat.
1. Description of Vegetation and Physical Landscape
Coastal barrens could be called heathlands, in that they are relatively unproductive, have poorly developed organic soils and are dominated by plants in the Ericaceae. They have similarities to heathland habitats in northern Europe and share at least a handfull of species. The are also similar to tundra habitats and several of the species common on Nova Scotia coastal barrens are found in greater abundance further north (e.g. Empetrum nigrum) Coastal barrens are not very common in Nova Scotia, with the majority being concentrated on the Chebucto Peninsula (especially around Peggy’s Cove) and in the Canso area. These kinds of habitats are much more abundant in Newfoundland where barrens cover large areas. There is apparently very little of this vegetation in New Brunswick and PEI although there is a small barren near the wind power installation at North Cape (I believe it is called Black Marsh—it is definitely more of a bog). The barrens habitat in Nova Scotia is really a mosaic of several different habitats occurring in patches of various sizes across the barrens landscape: bogs occur in depressions where water accumulates. These are characterized by many typical bog species including Sarracenia purpurea, Sphagnum spp., Carex trisperma, Carex nigra, Vaccinium oxycoccus, etc. The larger bog patches are clearly “real” bogs with peat accumulation, but we don’t know whether very small pocket bogs also have the same kind of peat accumulation. At the other extreme, on ridges, rock outcrops or erratics, plants occur mainly in cracks in the rock. Typical species in these areas are Sibbaldiopsis tridentata, Danthonia spicata, and Empetrum eamsii (the red form of Empetrum nigrum: I believe the jury is still out on the taxonomic of the crowberries—some authors believe they are just different forms of the same species; In our habitats, they appear to occupy different habitats: E. eamsii is only found close to the coast (less than 500m) and typically occurs on rocks where there is greater wind exposure and less soil. E. nigrum can sometimes be found in the same habitat, but appears to have a much broader tolerance for environmental conditions. E. nigrum is found in bogs throughout the province (including bog patches on the barrens) and is the most abundant species on coastal barrens. Between the moisture extremes of bog and rock outcrop, most of the barrens are covered with a thin layer of organic substrate, contributed by the plants themselves and these intermediate areas are covered with low shrubs, E. nigrum, Corema conradii, Gaylussacia baccata and G. dumosa among others. There are also patches of trees within the barrens, giving way to more or less continuous forest as you go further inland (see section 2. Dynamics, below). These tree patches are most commonly spruce (Picea glauca or P. rubens), balsam fir (Abies balsamifera), jack pine (Pinus banksiana) or even deciduous. Areas of tall shrubs are common in some barrens (Gaylussacia baccata, Viburnum cassinoides, Ilex verticillata, Nemopanthusmucronata etc.), may represent an intermediate stage between open barrens with rock outcrops or low shrub assemblages, and fully treed patches (again, see below 2. Dynamics).
The crowberries contribute much of the characteristic sweet scent of the barrens, and this appears to come from the foliage. Other less abundant, but fragrant species include Morella pennsylvanica and Myrica gale. An equally pungent, but less pleasant plant odour on the barrens comes from the decaying leaves of wild raisin/witherod (Viburnum nudum (cassinoides in some taxonomies)). This is mainly apparent in the fall: the fresh leaves of the species have little scent; the sour milk/feces odour appears to develop when the leaves begin to senesce in the fall. The litter layer below the shrubs also stinks. (It took us years to figure out that we were not completely nuts in associating these musty smells with wild raisin).
The barrens are full of edible plants as well. Many of the Ericaceae produce edible berries (although the edibility of plants like teaberry/wintergreen (Gaultheria procumbens) are debated in our lab): cranberries, blueberries, huckleberries are probably the best known and most edible. Crowberries are edible but are lacking in the flavour department. The large crowberry crop has been a source of some mystery: these plants produce a huge amount of biomass in berries, pretty much every year, but seedlings are rarely seen. Even after fires, most of the recruitment comes from vegetative spread from existing plants. We saw this after a burn swept through the entire coastal barren at Keji Seaside annex (Port Mouton side) in 2006 or 2007. There were very few seedlings, mostly suckers from Gaylussacia. The main species that recruited from seeds were asters and goldenrods (see Dynamics below).
Nova Scotia coastal barrens host rare species of vascular plants (provincially rare in any case). These are more common at the more northernly and isolated sites such as Baleine (CapeBreton). Lichen and moss assemblages are less well known, but our surveys have turned up rare species as well. One of our honours students found over forty species of lichens growing on exposed boulders at a single field site. Several of these species represent new records for the province. There is much more work required to adequately document cryptogamic biodiversity in this habitat.
2. Dynamics
The long-term persistence of open barrens within forested landscapes cannot be taken for granted. Soil development eventually results in forest ecosystems in most parts of maritime Canada. Rock barrens and heathlands often occur as islands within a matrix of forest and can contain rare habitats or species. Determining the relative longevity of open barrens within forested landscapes is an important conservation problem, as some barrens seem to require periodic disturbances whereas others seem to be maintained by harsh edaphic conditions. I have involved several students in projects that examine limestone barrens in Ontario and coastal barrens in Nova Scotia. Some authors have argued that the presence of limestone ‘islands’ in a ‘sea’ of coniferous forest is due to periodic fires, and that prescribed burns might be necessary to maintain limestone pavement vegetation. Using a chronosequence from pavements burned recently (~80 years ago) to pavements not burned for 300-500 years, we showed that there were significant differences between the species composition and abundance of seed banks, above-ground vegetation, and soil parameters of pavements and adjacent forests, regardless of pavement age (Stark and Lundholm 2003, 2004). This suggests that the old pavements were not experiencing forest succession (or significant development of soil), at least over ~500 years. We found no evidence that forests are invading the pavement patches via seeds, nor did adjacent forests contain a repository of seeds for typical pavement species. This indicates that harsh conditions on the limestone pavements (shallow soil, frequent drought) prevent forest colonization.
In contrast, our work on coastal barrens in Nova Scotia reveals that up to 16% of coastal barren area has become forested in the last 80 years, thus parts of these systems are successional (Burley and Lundholm 2009). In general, it appears that coastal refugia that resist forest encroachment are persistent, while inland or more sheltered barrens undergo periodic expansion due to disturbance, and subsequent contraction due to forest succession. We recently also showed that tree islands in coastal barrens in Nova Scotia do act as nuclei for forest succession (collaboration with treeline specialist Dr. Karen Harper). We have also collected data to examine the role of seed rain and seed banks, and the availability of beneficial mycorrhizae (collaboration with mycorrhiza specialist Dr. Gavin Kernaghan) on the potential for trees to colonize open barrens. We mainly believe that the more stressful parts of the coastal barrens can maintain themselves over long periods of time due to shallow soils, and coastal effects such as high winds and salt spray. Barrens vegetation can colonize nearby areas when forest cover is removed by fires, sheep grazing or other disturbances, but where soils can build up and conditions are less harsh, these areas become forested again over time. Some barrens contain evidence of past fires as charcoal in the soil, but others have no such evidence, thus it appears that fire is not required for the creation or maintenance of barrens, but it may have been the dominant force shaping some of our coastal barrens.
The availability of exposed bedrock seems to play a key role in providing optimal conditions for the persistence of barrens vegetation: the largest and longest-lasting barrens are invariably associated with large outcrops, such as those found at Peggy’s Cove and Canso. In other parts of the coast where glacial till overlies the bedrock, forest often extends right to the edge of the ocean, with only a very small area where typical coastal barrens plants such as Empetrum nigrum can be found. In the north of Cape Breton Island, there is a 30-50m strip of barrens vegetation with Empetrum eamsii, Campanula rotundifolia, and Plantago maritima on many of the headlands, but the lack of hard bedrock near the surface results in forest cover as soon as the strongest coastal effects fade.
Many of the dominant plant species on coastal barrens are also common in adjacent forests (Gaultheria procumbens, Vaccinium vitis-idaea, Cornus canadensis, Maianthemum canadense etc.), suggesting that these populations on open barrens may be simply remnants of a previous forest. On the other hand, barrens populations of these species are often morphologically distinct from those in adjacent forests (thicker leaves etc.), leading to the suggestion that there may be genetic differentiation in response to the coastal environment. Genetic analyses do not show clear separation between barrens and forest populations of Vaccinium vitis-idaea, indicating that there is likely gene flow between the populations. Further work could identify whether this is true of more of the species that straddle both open barrens and closed canopy forests. There are many questions left to answer about the landscape dynamics of coastal barrens: as sea level rises, will the coastal effects spread further inland, creating a larger area of persistent barrens? What happens to all the seeds produced by plants on the barrens? What is the role of winter conditions in survival of tall shrubs on the barrens?
3. Plant Species Coexistence
Plant species coexist despite requiring only a handful of limiting resources. In theory, the best competitor for the limiting resource in an area should eventually outcompete all the others. On the ground, monocultures are rare in nature, thus ecologists have sought to understand how different species coexist. It is clear that different species prefer different habitats, so on a landscape comprised of patches with different environmental conditions different species coexist on the landscape, because they can outperform other species in their ‘home’ habitat (there is a good amount of evidence in support of this). This phenomenon is highly likely to partly explain the persistence of high plant biodiversity on the barrens: the barrens landscape itself is comprised of patches characterised by distinct sets of environmental conditions (bogs, rock outcrops etc.) and there are characteristic sets of species that appear to prefer these areas. However, within a patch or plant neighbourhood, where individual plants can actually interact, it is unlikely that individuals from different species are partitioning microscale differences in the local environment. For one, even if micro-scale patches have consistent environmental conditions over time, the appropriate species have to be guaranteed access their preferred patches, and in plant communities, seed limitations often result in the superior species not making to a particular patch. Within plant neighbourhoods on the barrens, we found no evidence that greater environmental diversity (e.g. plots that had greater [horizontal] spatial variability of soil depth or moisture, compared to others) led to greater plant species diversity, thus more environmentally heterogeneous plots do not appear to contain more niches than more homogeneous plots. Other ways in which plant species can coexist include partitioning different forms of the same resource, which has been shown to allow coexistence of tundra species specializing on the use of different forms of nitrogen, and temporal resource partitioning, where species differ in their growth rate or production of seeds at different times. These niche explanations for coexistence all rely on the idea of trade-offs, that no single species can be good at all possible conditions, and that species differ in key traits related to resource acquisition such that they specialize on a subset of all possible conditions in a region. We are now beginning to explore these two mechanisms as possible, niche-based explanations of species coexistence in plant neighbourhoods on the coastal barrens.
In contrast to these niche-based hypotheses to explain neighbourhood coexistence, neutral theories assume that species are functionally equivalent, such that niches are not required to explain the coexistence of large numbers of species. Low mortality rates and slow growth can result in apparent coexistence for long periods of time even though species are not differentiated into resource niches. Even when plant species clearly show resource-use trade-offs associated with niche differentiation, if dispersal limitations prevent individuals from reaching preferred microsites then niche differences may make less of a contribution to long-term coexistence than the classical theories would predict. One of the key assumptions of niche-based coexistence theories is that intraspecific competition should be more intense than interspecific (conspecifics are more similar to one another than to individuals of different species, thus compete more directly for the same resources at the same time). Using size-distance regressions in forest understories, Marina Neytcheva found no evidence that intra-specific competition is more intense than interspecific competition, thus other mechanisms besides niche-differentiation may be required to explain coexistence. On the coastal barrens, it seems fairly clear that non-random assemblages of species are differentiated in terms of their preferred types of patches on the barrens landscape. On the other hand, the ordinations show great overlap between what I have called distinct types of patches on the landscape. Thus, while bog-dwelling species may outcompete most non-bog species in bog patches, several species have wide tolerance of environmental conditions and thus challenge this narrow interpretation of niches. Within a patch defined by fairly spatially consistent environmental conditions, explaining coexistence becomes more difficult. We speculate that within a neighbourhood, all the species that can handle conditions within that patch type are competitively equivalent, and coexist unstably over long periods of time due to slow mortality and growth rates. We are now planning experiments to evaluate some further niche mechanisms and compare these to the expectations under neutral models.
In the past, I focused on experimentally submitting plant communities in microcosm to varying levels of temporal variability in resource supply, to see if these alone could produce the shifts in diversity that would signify temporal niche differention. First, these studies showed that losses of species were partly not random, thus some species were more suited to some sets of conditions. Second, random losses of species were also detected, thus both species-specific processes linked with niche differentiation and species-neutral processes linked with random events contributed to overall diversity patterns. These studies suggest that temporal niches may be worth looking for, although alone they do not provide direct evidence of phenological differences that might represent stable coexistence mechanisms (stabilizing mechanisms).
4. Conservation and Restoration
Many of the key coastal barrens areas on the Atlantic side of the province are protected. These included National Parks, provincial Nature Reserves and Wilderness areas. The main unprotected barrens area is the large barrens complex around Peggy’s Cove. While the protected areas represent an important step toward truly protecting these habitats from human influence, during our fieldwork we discovered active ATV trails throughout each of these areas. ATV trails can damage sensitive vegetation, ruin soil and alter hydrology. The worst damage appears to be in wet areas, where in many areas, ATV users create multiple parallel trails, and wreck large bogs by “doing doughnuts”. Some of these trails can be seen clearly on Google Earth. Since this kind of damage occurs in protected and unprotected areas alike, it appears that more enforcement is required in the protected areas to ensure that recreational vehicles are kept out of ecologically sensitive areas.