In defense of bare ground
On the road to Te Anau, a remarkable reserve sits incongruously at the side of the highway, as a reminder of the former face of some of our driest landscapes. The Wilderness Scientific Reserve is a strange, otherwordly place – a matrix consisting predominantly of bog pine (Halocarpus bidwillii) and what appears to be open ground.
My interest in this remnant habitat was sparked by a scientific paper that I had read on the ancient vegetation of dryland areas in the South Island, in which bog pine woodland/scrub was noted as an important (and in some areas, wide-ranging) plant community. Setting aside the rather appealing irony of a plant called ‘bog pine’ being dominant in extremely dry habitats, one of the other conspicuous features of this assemblage of plants is the part that ‘absence’ plays in its character.
This sense of ‘absence’ is provided by a matrix of slow-growing mosses and lichens, diminutive herbs and ground that is simply bare. Time plays a significant role in this kind of landscape, as the moss and lichen species that give the appearance of hoar frost on the ground develop very slowly when relatively undisturbed by human activities. The eerie character of Wilderness Scientific Reserve shows one face of what happens when we simply allow places to develop at their own pace, and according to their own rhythms (which are partially determined by the lack of fertility in the soil).
Similarly, steep banks in the north of New Zealand (especially within scrub and gumland areas) are often noticeably devoid of vegetation on the ground plane. In some cases, their state of undress exposes beautiful geological strata, as in the example of beautiful roadside banks leading into the Desert Road on its northern end. In others, a benefit quite unrelated to aesthetics is apparent in pockmarked mudstone faces, where native bees busily move in and out of their burrows.
Many of these sparsely-vegetated places provide habitat for plants that find an advantage in bare ground; in particular, certain of our native orchids (such as the sun orchids, or Thelymitra spp.) and several species of our native daphnes (Pimelea spp.). The proliferation of interesting species that occur in the northernmost zone of our mainland (around Te Paki and Spirits Bay) is, to a considerable degree, facilitated by the plethora of exposed habitats that plants like Pimelea xenica (a recently-described northern species of native daphne) or an attractive, weeping shrub, Pomaderris edgerleyi, prefer.
Throughout our natural landscapes, bare ground abounds. However, within consciously-designed plantings, it is (usually justifiably) perceived as a failure of design. For this reason, plants that we term ‘groundcovers’ are an important component of landscape work. Now obviously, I do not wish to challenge the legitimacy of covering the ground within this article. In a great deal of situations, that is a desirable goal (from both a practical and aesthetic point of view), and accordingly, we utilise a wide range of groundcover species within our work.
However, it is worth considering critically the different ways in which control is exerted within environments (natural or otherwise), and whether we achieve our goals through standard approaches - such as the assumption that clothing the ground is universally correct and natural. More importantly, as I have regularly contemplated whilst observing plants in our wild places, we should acknowledge that areas of predominantly bare ground are an intrinsic part of the diversity of habitats that make up a place (as well as reflecting their dynamism).
If a measure for performance anxiety in plants existed, I suspect that the stress levels of certain plants that we describe as ‘groundcovers’ would be rather elevated. With that title, language has conferred a single purpose upon a diverse range of plants, many of whom ironically require open ground and disturbance to maintain their tenure in their natural habitats (and whose tendencies are certainly not to divide and conquer).
Within natural ecosystems, many commonly-cultivated mat-forming groundcovers (such as species of Lobelia or Selliera) grow in particular sets of conditions that frequently exclude competing vegetation. In the case of Selliera, salt is the editing factor. For native members of genera such as Gunnera and Mazus, moisture (especially extreme seasonal inundation) is a major element in furnishing them with a competitive advantage. In the case of the attractive northern native, Leptinella rotundata, this threatened species endures in exposed coastal scarps, where wind and the dynamic nature of the habitat dictate that more aggressive plants cannot displace it.
However, when transposed into a planting in which none of the original limiting conditions exist, plants such as these lose their competitive advantage. This is one of the major reasons that the notion of a weed-suppressing mat of plants often works better in theory than it does in practice (unless regular and diligent maintenance is applied). If our goal is to establish a mantle over a given patch of earth, this raises questions about the best way to create that mantle.
For example, does it even need to be on the ground ? If the major battle with weeds takes place on the ground (especially in weedy places such as Auckland), why fight weeds on their own territory ? In recent years, we have looked into the use of a ‘suspended’ plane of diverse, shrubby planting to defuse the issue of the ground plane, at the same time as establishing a continuous mantle for the eye. In essence, this approach permits the ground to remain bare beneath the low, suspended canopy, and therefore more easily managed in plantings where low levels of maintenance are likely to be applied. This is, of course, similar to the rationale behind canopy closure within revegetation projects (especially in the north of the country), wherein the ground plane is controlled indirectly with a carpet of shade - rather than a direct physical carpet of plants.
It is interesting to reflect that the assumption of the desirability of permanent groundcover is, to a certain extent, a product of climate and horticultural norms. In many high-profile North American and European public plantings (such as the work of the Swiss landscape architect, Günther Vogt, or the planting designer, Piet Oudolf), a desire for textured, naturalistic plantings and seasonality means that ground is regularly exposed between growing seasons. For entirely different reasons (as my friend Andy Hamilton recounted to me from his time designing landscapes in India), ground is deliberately left bare around houses in India, where the luxuriance of tropical vegetation pressing in towards residences becomes overwhelming - especially when water is everywhere during the monsoon.
Obviously, we are not about to experience a rush towards plantings consisting of substantial areas of conspicuously bare ground, and that is certainly not the intention of this article. However, it is important to acknowledge the validity of bare ground (whether permanent or periodic) at the point where landscape architecture intersects with natural plant communities, or where active land management is required to assist threatened species or plant communities. This goes against a basic instinct to chase bare ground away (as disapproval towards ground that is neither cultivated nor naturally clothed) - a desire that can act as a factor in the decline of threatened herbs.
Furthermore, it is extremely useful to get our heads inside the matter of how ground-covering species actually behave within natural environments, and reflect that diversity in a variety of treatments within designed plantings (or designed processes). Even if the endgame is (as in most plantings) to bedeck the ground in green, greater understanding of the ground as an interface upon which many different outcomes are played out makes us better designers.
Plant communities associated with bare (or disturbed) ground
Contrary to the old saying, I would contend that nature does not actually abhor a vacuum. In fact, on the basis of regular meteorological evidence, nature seems on occasions to love nothing more than a good ‘vacuum’, as the landscape becomes ruptured by storms and other events. Nature is in a constant state of flux in which landscapes are being alternately clothed and stripped.
This introduces a temporal dimension to how we view landscapes, wherein the current, seemingly stable appearance of a place may not be as permanent as it seems. When the next big storm comes through, that corridor of riverine forest or verdant slope standing before you might look very different.
Perhaps more accurately, it also introduces an issue of scale to how we perceive places, as ground that we perceive as being bare may in fact support life in ways that don’t conform to notions of planted sites – for example, in the form of tiny, non-vascular plants. Just because we cannot see a process at play, it does not mean that something is not developing. Viewed thus, the seemingly bare surface of a rock becomes an embroidered garden of mosses, lichens and bare surface.
As previously noted, water is a major agent of disturbance, whether by beating a path through or inundating a place. In the case of the critically endangered Mazus novaezeelandiae subsp. impolitus f. hirtus (a worthwhile landscape species), damp ground and shade combine to create open conditions in which this diminutive, pale-flowered species can find a place. At one of the two sites where I have seen this beautiful little herb, it currently occupies the zone along the margin of damp forest near East Cape where its largest population endures. My friend who showed me the site (Graeme Atkins, a DoC ranger who achieves remarkable feats with conservation on the East Coast/Tairawhiti) informed me that the Mazus had spread much further out into the pasture in the previous growing season, but had mostly retrenched to the forest margin by the time that we visited.
This kind of flux is a regular part of the growth cycle of plant populations, based on whether a year has been wet or dry, whether livestock have been grazing, or other environmental factors. Conditions that favour a thick sward of exotic grasses are unfavourable to the Mazus, whereas the inverse (such as extreme inundation causing pasture to thin out) provides a matrix of pockets of bare ground in which the native herb may thrive. Seasonal flux of a more predictable nature takes place on the margins of our high country tarns (of which the Glenmore Tarns are an outstanding example), where the alternating periods of winter inundation and dry summers maintain conditions that furnish many diminutive native herbs (including the sweet-scented Lobelia perpusilla) with the advantage that they need.
On the opposite end of the moisture spectrum, Auckland’s lavafields constitute another kind of habitat with large areas of bare ground; simply due to the conspicuous presence of countless volcanic rocks upon the surface. Rangitoto Island’s otherworldly landscape provides an example of how parts of Central Auckland appeared before being cleared to accommodate the city’s growth. One of the most distinctive qualities of lavafields is the presence of plants that would normally make a life on the comparatively bare surfaces of tree branches (such as Astelia solandri and Kirk’s tree daisy, Brachyglottis kirkii subsp. kirkii), based on the similarity between these two well-drained habitats. In southern parts of the country, scree slopes and other rocky habitats provide equally specialised environments, with distinctive aesthetics and plants associated with them (such as the remarkable penwiper, Notothlaspi rosulatum, which has taken on an appearance similar to the stones surrounding it).
The word ‘erosion’ has rather dramatic connotations; generally being perceived as a destructive (and disruptive) process associated with extreme weather events. However, in many places, extreme weather might be defined as those rare occasions when a place isn’t actually being pummelled by the elements. In locations like Hokianga’s South Head, constant erosion provides exposed ground for species such as Pimelea orthia to develop. The commonly-cultivated form of Coprosma known as C. ‘Hawera’ takes this inclination to extremes by perching right on the edge of the permanently unstable cliffs found along stretches of the South Taranaki coast. The broken rocks that accumulate at the base of cliffs and slopes (known as talus) can also provide suitable bare ground in which plants like the endangered taxon, Lobelia ‘Woodhill’ (on Auckland’s West Coast) or Myosotis colensoi (a forget-me-not from Castle Hill) are able to develop, in the absence of competitors.
In the north of the country, large stretches of infertile habitats provide consistently bare ground on which a host of specialised plants develop significant communities. It should come as no surprise that certain of these genera (such as Leucopogon, Pomaderris and Pimelea) are also associated with eastern Australian plant communities, given the weathered, infertile character of many Australian ecosystems. Indeed, one of our most beautiful native flowers, Hibiscus richardsonii, is shared with New South Wales, whilst a particularly graceful meadow-like grass, Dichelachne crinita, is also common in Australia, where it has received considerable use within gardens and landscape work (as opposed to New Zealand, where it is practically unknown within cultivation). All of our 4 native species of Dichelachne (of which 3 have significant potential for cultivation) are well suited towards open situations, whether on bare banks (such as mudstone cliffs on Auckland’s Hobson Bay, where Dichelachne crinita grows) or in the small gaps afforded at the base of open tree specimens (as at Waikumete Cemetery, where Dichelachne inaequiglumis maintains a foothold in the ‘worst’ areas).
Also associated with the north of New Zealand (although equivalent landscapes occur in a number of seams in other parts of the country) are the ultramafic cliffs and plateau around North Cape. The nature of the serpentine-rich geology that lies beneath these Spartan ecosystems dictates that the soils are high in minerals/elements (notably iron and magnesium) that have an adverse effect on plant growth. As a result, bare ground abounds, whilst the plants that do make a home on the Surville Cliffs (including Pittosporum pimeleoides subsp. majus and Coprosma spathulata ssp. hikuruana) exhibit stunted growth forms in response to the hardships that they face. In this case, the prospect of less bare ground is injurious to the distinctive plant communities found here, with some of the major offenders being the Australian shrubs/trees from the genus, Hakea, which (due to their evolution on extremely infertile soils) can create new systems that exclude the plants that have developed here over millennia.
Similarly, ultramafic landscapes elsewhere in the country, such as the Red Hills near Nelson or their namesake (the Red Hills) in Southland, are characterised by low, open habitats in which distinctive species have evolved. Of course, cliffs don’t need to be toxic to limit vegetation. They just need to be full of solid bedrock, like the bizarre landscapes of the Denniston Plateau, or marble domes like Hoary Head in Kahurangi National Park (where the diminutive Clematis marmoraria creeps within fissures of marble). On Maungaraho Rock (one of several volcanic plugs within Northland), Hebe saxicola emerges from the dense rock of this eroded volcano.
In contrast with the mainly exposed landscapes mentioned above, bare ground is also a common fixture within forests, whether as a result of dense leaf litter, deep shade or the undesirable editing influence of introduced animals. It could be contended that leaf litter (as a layer of plant-derived material) does not really fit into the category of bare ground. However, as an alternative to actively-growing vegetation, it is interesting to consider the place of leaf litter in designed landscapes. It has struck me as rather unusual that people are often keen to assiduously remove leaf litter from plantings, only to replace it with mulch, when leaf litter is part of the overall aesthetic and ecology of a place. In the case of taraire, cabbage trees or puriri, the futile removal of leaf litter takes on a positively Sisyphian dimension.
Leaf litter can provide a beautiful effect, as demonstrated by the ethereal understorey that develops in our native beech forests (or, for that matter, in European beech forests as well). The skeletal, partially decomposed leaves of Hoheria contribute a distinctive layer to a place, whilst the forest floor beneath kauri (and other native conifers) often takes on an attractive character. This leads on to the interesting prospect that we might actively consider the effect of particular types of leaf litter upon the ground as a result of the choices made around which trees populate a woodland planting.
One final point pertaining to the place of bare ground within natural environments is the influence of animals as agents of continual disturbance within landscapes. As my friend Terry Hatch once informed me, in pre-human times within these islands, millions of seabirds were returning to the land every evening when roosting (flying as far inland as the Volcanic Plateau).
Where petrels perform their clumsy landings and form burrows, the ground is consistently being opened up (as well as being thoroughly covered with guano). These habitats provide the ideal conditions for a range of herbs like Cook’s Scurvy Grass (Lepidium spp.), some of which have become rare due to the decline of the animals that pave the way for them. Whether by water, drought, shade, infertility or the fumbling travails of breeding seabirds, bare ground is natural within our landscapes, and much more common than many of us might generally assume.
Agriculture’s influence on landscape culture
Our attitude towards bare ground is not just a function of aesthetics, but also of protocols. With respect to the latter, many of our protocols about land management (whether for productive or amenity purposes) are born out of agriculture. The presumed natural cover of a sward is based on ideals from grassland and pasture; agrarian models that sit at the heart of the development of civilisation. As a result, many traditional values around landscapes are based on greenness and uniformity as markers of a functioning place.
This situation has, of course, changed with the appreciation of the diversity of natural landscapes within New Zealand. However, a residue of it can endure in certain places that don’t make the marketing brochure. In these cases, the legacy of agriculture can be an assumption that fullness is desirable, and apparent ‘emptiness’ is either symptomatic of a paucity of natural heritage or a situation to be remedied. There is something akin to guilt attached to allowing ground to lie ‘unimproved’. It is in the cases where diversity is difficult to perceive (such as the gumlands of the north, or the grasslands of the Mackenzie Basin or Manuherikia) that ambiguity can creep in.
Paradigms for cultivation are also based on the template of agriculture, especially in relation to the manner in which ground is prepared and fertiliser is applied. We operate on a system of continual addition, which optimises the conditions for both desirable plants and weeds. Raising fertility has major repercussions on some plant communities, whether via active addition of fertiliser, the addition of nitrogen-fixing plants into a system (such as the weed, Lotus pedunculatus) or the influence of animals like rabbits. When managing or planting certain landscapes, we need to take a critical approach towards the issue of fertility, and ask ourselves whether universal ‘good practice’ really is the best thing.
The place of ‘absence’ and bare ground in designed landscapes
The relevance of observing (and accepting) bare ground in the design of large-scale landscapes, especially those that encompass significant natural areas, is self-evident. On a larger scale, design is heavily focussed on processes, and the gestures are larger. Dynamism is also more likely to be accepted, and it is necessary to ascertain the extent to which certain types of control may be exerted upon landscapes.
On a finer scale, it opens up a varied conversation. For me, the most important consideration is the place that dynamism and seasonal flux are allowed to occupy in designed landscapes. With respect to this, it is worth mentioning the dichotomy between the efforts of private individuals in their gardens (where seasonal flux and happenstance are more readily permitted) and the introduction of a professional dimension to plantings (in which seasonal variation can be viewed as a dereliction of duty).
However, it is worth mentioning once again the example of many Northern Hemisphere designers, whose work has distinct cycles of growth that allow for the space and light requirements of flowering species that would otherwise underperform in a solid, densely-planted, evergreen mantle. Evocative images of flower-laden meadows like the High Line regularly make their way in to the renders of students and professional designers, yet the system of gardening required to create such plantings is generally anathema within landscape architecture.
Bare ground does not need to be visible. It may be concealed within a matrix of structural planting. In such cases, it is worth asking the question of whether bare ground invites more work than groundcover species, for the latter can sometimes function as miniature refugia for weeds, from whence they radiate out between episodes of maintenance.
As noted previously, the question of fertiliser is an important factor in the development of landscapes, and the kinds of plant communities that develop on them. It is interesting to consider how superior outcomes may be achieved by adopting approaches that are not based on the agricultural paradigm of adding significant amounts of fertiliser. In one project that we have carried out in recent years, our approach towards managing the site was to steer the soil towards low fertility (a situation that most soils naturally veer towards in isolation), in order that desirable species are not bullied by plants more capable of dominating in high fertility. Consciously promoting low fertility raises questions about our expectations around speed of growth (and by extension, coverage of the ground), and how this balances with the availability of resources for undesirable species.
It is not my intention within this article to provide a full range of scenarios in which bare ground may be permitted to play a role in landscape architecture (whether in the form of lavafields, leaf litter, low-fertility meadows or other settings). My concern is more with thinking about the different ways in which we approach the ground, beyond viewing it as a canvas to be adorned or a problem to be covered. Whether the intention of a design is to have a verdant sward or a more dynamic matrix, it is important to simply look at the ground and think critically about how it may develop. When we look objectively at the true face of natural and planted environments, we see that, far from being an aberration, bare ground is an intrinsic (and inevitable) part of landscapes.