Naturalistic ecological woodland plantings
The key distinguishing feature of ecological plantings of trees and other woody species is the presence of one or more layers of understorey vegetation. Conversely, conventional urban parkland in the English Landscape style consists of mature trees limbed up to several metres above ground level in a setting of mown grass.
From the earliest days of landscape preference research in the 1960s, there have been a number of lines of research that have consistently found that images depicting multilayered woody vegetation of the kind one would expect to find within ancient woodland or along a woodland edge in a state of natural succession attract lower preference scores than images of parkland in the style of the English Landscape movement (Ulrich 1977; Kaplan 1985). Based on these studies, the assumption has grown that multi-layered woody vegetation itself is lower in preference than mature trees set in mown grass:
Thick undergrowth and dense stands of trees detract from the scenic beauty of forested environments. In particular, recent research suggests that humans may have a biologically prepared predisposition to associate negative consequences with spatially restricted natural environments.
(Ulrich et al. (1993) quoted in Parsons (1995))
Some of these studies are open to criticism. For example, it can be argued that that the images depicted simply do not compare like with like: a close-up of a woodland edge is quite different from a long view of an open woodland glade—one is an image of the. structure of the vegetation itself, the second is an image of the spaces defined by the vegetation. This is the case in the study by Ulrich cited above (1977). In a later paper (1986), Ulrich refers to two sample images from the high and low preference groups in the earlier study. The first is a typical parkland landscape in the English Landscape style. The second example is a much closer view of roadside scrubland. In the first image, the vegetation consists of mature trees limbed up to several metres from the ground, combined with what appears to be mown grass; in the second the vegetation consists of young trees with a dense understorey of thorny scrub and herbs. In the first image the vegetation appears healthy but in the second there are several leafless trees or shrubs that appear to be dead or dying. The topology in the two images is also completely different. In the first image the ground is predominantly level, whereas in the second the ground rises markedly away from the viewer, thus further reducing the visual permeability of the scene. There are in fact a number of variables that differ between the two scenes, variables that are not controlled for in the study.
In terms of aesthetic preference for the two different landscapes, it is arguable that most people would choose the long view for the simple reason that it is more interesting, because the image itself contains more. It is rather like comparing a photograph of a strip of wallpaper with a photograph of an entire room papered with different wallpaper. Whilst people may prefer scenes that contain long view distances over close views when comparing visual images of landscapes, such studies certainly do not support the hypothesis that certain kinds of vegetation are inherently lower in preference.
In one study focusing exclusively on near-view forest scenes, the degree of visual penetration was found to be a significant predictor of scenic beauty (Ruddell et al. 1989). However, visual penetration is not associated exclusively with certain kinds of vegetation. Visual penetration is also dependent on the spatial arrangement of vegetation and view distance. The relationship between view distance and vegetation density was explored by Purcell and Lamb (1998) who found an interesting interaction. They found that whereas sparser vegetation was preferred to denser vegetation in close views, the reverse applied in wide views. Here preference was related to view distance and not solely to the qualities of the vegetation itself.
Further, the bulk of the research relied upon by commentators such as Parsons (1995) was carried out in American forests, many of which were planted and managed for commercial purposes. The levels of tree density encountered during some of these studies (in excess of 1,000 trees per acre) (Hull et al. 1987) are far higher than one would normally expect to encounter in an urban public situation. Schroeder and Green (1985) investigated public perception of optimum tree density in American public parks and found that the preferred density varied from 40 to 65 trees per acre depending on whether the background was dense or open. Hence, the research carried out in American forests has to be viewed with some caution. Further, many of the findings from this research relate to coniferous rather than deciduous forests. In at least one of the studies relied upon by Parsons in support of his contention that ‘thick undergrowth and dense stands of trees detract from the scenic beauty of forested environments’, there was no significant relationship either way between understorey vegetation density and perception of scenic beauty, although the impact of this variable may have been represented by other stand characteristics in the study (Hull et al. 1987).
To some extent, however, these studies miss the point, because, as we have already seen, landscape aesthetics should not focus solely on preference for different views of landscapes rather they should embrace a whole gamut of different approaches, ranging from how we perceive landscapes in terms of their utility to the feelings they evoke in us. Further, we do not experience landscape solely from a series of static viewpoints. A great deal of our experience of landscape is dynamic: we get to know landscapes as we move through and interact with them, seeing them from different perspectives and experiencing them in different ways at different times.
From the 1960s onwards there has been a large tranche of research into forest landscapes, particularly in Scandinavia and the USA. This research has generally taken the form of collecting public responses to photographs depicting different forest conditions. Participants are shown a series of photographs of different forest scenes and are then asked to rate them for scenic beauty. The ratings are then compared to the content of the photographs to determine the relative preference for different factors: an approach known as the ‘psychophysical’ approach. A review by Ribe (1989) made the following findings. Comparisons of preference for managed as opposed to unmanaged or natural forests have yielded contradictory results (presumably because these definitions are fairly loose: managed and natural forests come in many different forms). High tree – density, particularly of young trees, is considered less attractive than medium densities (though one study found the optimum number of trees per acre to be 1,150 (Buyhoff et al. 1986). Vegetation structures that permit visual penetration are preferred to those that do not. The presence of a shrub or sapling understorey has been found both to enhance and detract from a scene (again, this may be because of the many different characteristics a woodland understorey can have in terms of variation in vegetation type and structure). A variety of species is preferred to a monoculture where it gives rise to visual diversity. The presence of large trees enhances preference as does a ground cover of grasses, ferns, forbs or seedlings. Slash (the stumps and offcuts that are the aftermath of tree-felling) is strongly disliked.
Thus, it would appear that multi-layered woody vegetation is not disliked per se but that public perception of it depends largely on other factors, such as view distance and visual penetration.
In the UK there has been a growing movement in favour of habitat creation schemes in urban settings since the 1970s. This has led to the conservation and creation of natural or semi-natural vegetation, including woodland, in discrete locations
|
|
11.6
The full aesthetic potential of urban woodlands is rarely realised—how often do the trees and ground flora come together to produce a rich display in a designed woodland?
in these settings. These initiatives are often led or supported by the community, and anecdotal evidence suggests that they are, on the whole, popular. However, because urban nature reserves or habitat creation schemes are popular, it does not follow that all naturalistic woodland ecological plantings in urban settings will automatically secure public acceptance. The urban nature conservation movement often builds on existing ecological capital so there is frequently a perceived need to protect what is already there: public support frequently rallies around such issues. Further, the overriding aim is to create or repair habitats for wildlife: also an objective that attracts popular support. These schemes are often restricted to distinct locations with recognisable qualities of their own, and constitute a small proportion of urban green-space overall, so that most urban dwellers have a clear choice about whether to visit them. There are major differences between these initiatives and using ecological woodland plantings in public urban landscapes for structural or aesthetic purposes (Figure 11.6).
The idea of using woodland ecological plantings in designed landscapes in public urban settings first emerged in the 1970s in the UK. One of the first pioneering examples was Oakwood in Warrington New Town. Oakwood was remarkable not only for the way in which the naturalistic woodland belts, landform and open spaces were used to structure the whole development, but also for the manner in which the vegetation and other landscape elements were skilfully and closely integrated with the built development. Oakwood marked a sea change that had come about in the minds of many landscape professionals.
The naturalistic landscapes created at Oakwood were seen as beneficial for its future residents for many reasons. Adults could enjoy nature on their doorstep in the form of vegetation and green spaces that were robust enough to withstand regular use. Children would have many opportunities for
|
|
11.7
In Oakwood naturalistic woodland is closely integrated with the built development
adventurous and creative play amongst the vegetation. For all age groups there were opportunities to interact with nature whilst carrying out the daily activities of living: going to school, work or to the shops (Tregay and Gustavsson 1983). There was a desire to enable residents to actively experience the interior of woodland rather than just looking at it from the outside (Figure 11.7). The designers of this new landscape were passionate and committed, and were strongly influenced by similar developments that were already taking place in Europe, particularly in Sweden and in the Netherlands. Under the circumstances, it is perhaps hardly surprising that public involvement or consultation was not considered to be an important part of the design process. In any event, this would have been difficult logistically as most of Oakwood’s residents came from far away, from Manchester and Liverpool.
By the 1980s there was a growing awareness of the desire to incorporate more natural landscapes into towns and cities. The approach first used at Warrington became known as ‘the ecological approach’ and passed into mainstream thinking among the planners and designers of local authorities and New Town Corporations. One of the first evaluations of the validity of this approach, in terms of public perception, when compared to more traditional approaches to green-space planning, design and management, was the study of TartagliaKershaw (1980) into the role of urban woodland in residents’ daily life. Tartaglia-Kershaw carried out a detailed study of the Gleadless area of Sheffield, a housing area planned around an existing mature woodland. Although the woodland in Gleadless was generally within 500 m of the housing, and often considerably closer, it was not closely integrated with the housing as in Warrington. In Gleadless, the woodland and the housing formed two distinct and separate areas. 72% of the sample in the study said that the woods were important to them. An overwhelming 90% liked living on the estate, and 94% said that they liked the way the area had been planned. However, Tartaglia-Kershaw (1980) concluded that the overall findings did not support the approach used in Warrington:
The residents do not want woodland to the door as many figures in the
‘Nature in Cities’ movement suggest, and which is happening in New
Towns based on woodland structure planning [sic].
In another early study responding to the need for research on the impact of the nature and character of urban green-space, Burgess et al. (1988) examined the views of urban dwellers about their local green-spaces. They found that traditionally managed urban green-spaces characterised by isolated trees and mown grass were not valued as much as natural or semi-natural urban landscapes characterised by woodland, multiple layers of vegetation and an un-mown grass/herb layer. However, they also found that many people had ambivalent feelings about the landscapes they most valued: these landscapes were also the ones that aroused the most fear. They concluded that what people really want is a range of opportunities provided simultaneously in as many different green spaces as possible, and not zoned between different parks and green spaces.
Burgess’ findings about the value that people place on natural or semi-natural urban landscapes were confirmed and explored in more detail by Bussey (1996) in another landmark study. Bussey carried out an extensive study of urban-dwellers’ attitudes and feelings towards their local woods in Redditch, England, and found that woods were ranked above parks and were second only to open countryside as the preferred landscape for informal recreation. These findings are mirrored in an extremely large Dutch study of 3,118 respondents throughout the Netherlands. In this study, 57% of respondents said they would prefer small areas of nature and green-space close to home as opposed to a large nature area further away (Reneman et al. 1999).
The studies of Tartaglia-Kershaw (1980), Burgess et al. (1988) and Bussey (1996) were unusual and pioneering within the genre that can be loosely called landscape perception research. Through their mixture of qualitative and quantitative techniques, they were able to examine the ambivalent and sometimes conflicting feelings that we hold towards naturalistic landscapes in urban settings: such landscapes inspire negative as well as positive feelings.
Despite the innovative work done by researchers such as Burgess and Bussey, the idea that ‘woodland structure planting’ is regarded as unsafe by members of the general public, and is therefore unsuitable for use in urban situations, has persisted among local authorities (Thompson 2000):
Fear of crime can be as disabling as crime itself. One of the most unfortunate results of this widespread apprehensiveness is that vegetation has come to be regarded with mistrust by many urban residents. It is seen as providing hiding places for potential assailants. Landscape architects have had to take account of this fear. Some local authorities have actually been taking shrubberies out of parks and residential areas, and when considering new plantings designers are urged to use low-growing shrubs and to keep shrub beds back from the edges of paths. This defensive approach is in many ways the antithesis of the ecological ideals which were being imported from Holland in the 1970s. These called for mass plantings, more relaxed plantings, and an altogether shaggier, more naturalistic style of landscape design.
There is clearly a danger that, in seeking to reassure the general public by the removal of shrubby vegetation, local authorities are also destroying the landscapes that people most value, despite their understandable fears. However, it may also be the case that the ‘ecological approach’ was too wholesale, in that naturalistic vegetation was used too indiscriminately and too close to people’s homes, as predicted by Tartaglia-Kershaw (1980). There may well be an appropriate gradient of planting styles, ranging from more formal and manicured to ‘shaggier’ and naturalistic, corresponding roughly with distance from people’s homes and the places they have to visit daily. Arguably what is needed is an element of choice, as proposed by Burgess. People may well tolerate or even welcome more naturalistic treatments provided they can choose when to interact with them. These issues are explored in more detail in the final section.
A more recent study (Jorgensen et al. 2002) examined the impact of the spatial arrangement of woodland and the nature of the woodland edge on public perception of safety and preference in an urban park. Several different naturalistic edge treatments (flowering herb layer, dense understorey, flowering herb layer combined with dense understorey and, finally, native woodland edge) were contrasted with a more conventional parkland vegetation of specimen trees and mown grass in three different spatial arrangements (full enclosure, partial enclosure and no enclosure). Respondents were asked to rate digital images of the 15 combinations of edge treatment and spatial arrangement for safety and then preference. Although the respondents found the native woodland edge to be the least safe of all the edge treatments, there were some interesting findings in relation to the interaction between edge treatment and spatial arrangement. Reactions to the three different spatial arrangements of the woodland varied dramatically according to the nature of the woodland edge in the case of the spatial arrangements known as full enclosure and no enclosure, but not in the case of partial enclosure, when all edge treatments received similar ratings for safety and preference. The most dramatic variation was in the case of the dense understorey edge treatment: rated most unsafe in the full enclosure spatial arrangement but most safe in the no enclosure spatial arrangement. These findings suggest that, whilst safety issues are undoubtedly an important issue when working with naturalistic woody vegetation, design can play an important role in mitigating these issues.
There seems to be something particularly powerful, if not shocking, about a certain kind of urban nature, namely the kind of nature that takes over when an urban building or plot has been abandoned. Perhaps it is the speed with which the transformation takes place or perhaps it has something to do with the palpable power of nature to invade and even destroy man-made structures, splitting concrete and rapidly colonising seemingly inhospitable horizontal and vertical habitats. In Germany, this kind of nature has been given its own name: ‘industrial nature’. Up until recently, industrial nature was not tolerated. In the UK, where funds were available, industrial sites were dismantled, carefully graded and levelled, covered with topsoil and replanted with what came to be known as ‘woodland structure planting’. Evidence of this approach can be found on the outskirts of many of our towns and cities in the form of the Country Park. In Germany, an appreciation of the special qualities of these post-industrial sites has led to a whole new design approach:
The visitor first has to set aside his preconceptions, his knowledge that the land was once a workplace full of labour and toil, and is now nothing more than a devastated polluted site. Only then will he be able to appreciate its peculiar attraction, its atmosphere of decay, disorder, wilderness and chaos in a basically urban setting.
(Dettmar 1999)
These properties have been exploited to the full in the Ruhr district of Germany, in Emscher Park, other woodlands in the Ruhr District, and in the Stidgelande nature park in Berlin. These woodlands are all the effect of natural succession on derelict industrial sites. At some point during the late twentieth century, various public authorities in Germany came to the conclusion that these areas of industrial dereliction combined with spontaneously occurring vegetation had special qualities of their own worth preserving. In the case of Emscher Park, they conceived the idea of retaining the existing industrial structures and working with the spontaneously generated vegetation to create a framework for a large number of different recreational activities, as opposed to razing the entire site to the ground and starting again from a tabula rasa (Figure 11.8). An even more radical approach has been taken in other parts of the Ruhr where the ‘wild industrial
woods’, as they are now known, are simply left to their own devices, subject only to a fairly minimal management regime (Dettmar 1999). There has been no attempt so far to evaluate the success of these projects in terms of public attitudes but the very fact that such measures are taking place on such a vast scale (Emscher Park
|
|
11.8
An example of ‘industrial nature’— derelict industrial structures combined with spontaneously occurring vegetation in Emscher Park, Germany
alone covers 300 km2) must surely indicate a degree of public acceptance?
The use of the former power station at Bankside to house the Tate Gallery’s collection of modern art in the gallery known as Tate Modern, and the reincarnation of the former steel plant at Rotherham as Magna, a new interactive centre for recreation, suggests that former industrial structures are being re-evaluated in the UK. So far, however, these developments have focused on the built form, rather than the surrounding landscape, and it remains to be seen whether the radical approach to industrial nature exemplified by Emscher Park would find favour in Britain, given the differences in outlook referred to above.
