new banner
 

Home
Services of Scenic Solutions
Completed landscape projects
Papers and Presentations
Beautiful landscapes
Australian and International studies of landscape quality
The science of scenery
Landscape miscellany
Landscape resources
About Dr Andrew Lothian
Contact us
Site map

INTERNATIONAL LANDSCAPE STUDIES

Here we examine the international development of studies of landscape (or scenic) quality. Much of this material is based on research carried out during the 1990s in the course of my PhD studies. Although the material and references are slightly dated, much of it is as relevant today as it was then. In time, I may get around to updating the material. Only landscape preference studies are covered here, not the physical studies (see Protocol framework for the distinction)

Click on the following topics:

Twentieth century landscape studies
                Early 20th century landscape studies
                Typologies of landscape studies (summary)
                Characteristics of landscape preference studies
Typologies of landscape studies (extended version)
Findings from 20th century international landscape preference studies
        

TWENTIETH CENTURY LANDSCAPE STUDIES

Early 20th Century Landscape Studies

Compared with the last 30 years, the first 60-odd years of the 20th century were marked by few landscape studies and little apparent interest in landscape. Perhaps this is not surprising, given the dire economic situation of the 1930s, the impact of two world wars, and the major social upheavals associated with accelerating industrialisation, the shift of employment from the countryside and growth of the cities.

Prior to World War 2, the main term used in describing landscape was scenery, however the use of this term to describe theatre scenery was confusing, so increasingly the term landscape came to be used. In the United States, however, the term scenic quality is generally used.

Since the late 1960s, and paralleling the growth in community concern for the environment, there has been a very major increase in studies of landscape quality over the past 25 - 30 years. Much of the interest has been in North America [i.e. United States and Canada] and Britain, but the approaches of each has varied greatly.

In this section, the development of interest in landscapes is traced briefly by reference to Britain, the United States and Australia.

Britain

Geologists used the term scenery as the focus of geological explanations. Many books appeared with titles like The Scenery of England, but these did not describe the landscape in its aesthetic terms [other than generally] but in terms of the geological reasons for the appearance of the land.

These books included: Mackintosh, D., 1869. The Scenery of England and Wales, Longmans; Lord Avebury, 1906 [4th ed], The Scenery of England and the Causes to Which it is Due. MacMillan; Trueman, A.E. 1938. The Scenery of England and Wales. Victor Gollancz; Geikie, A., 1901. The Scenery of Scotland Viewed in Connection with its Physical Geology. MacMillan and Stamp, L.D.,  1946. Britain’s Structure and Scenery, Collins.

Geologists of the era focused on the effect of geology on the surface of the earth. Later geologists focused on the rocks and underlying structures and regarded the surface almost as an irrelevancy.

Geography was the main discipline in which an interest in landscape was kept alive in the early part of the 20th century. In 1920, the President of the Royal Geographical Society, Sir Francis Younghusband, addressed the Society on the theme, Natural Beauty and Geographical Science. Beginning his address with the words: “I have something to say which to old-fashioned geographers may appear revolutionary ...” he went on to argue that geographers should “regard the Earth as Mother-Earth, and the beauty of her features as within the purview of geography.” He claimed that, whereas mineral wealth and the Earth’s productivity is limited:

“the natural beauty is inexhaustible. And it is not only inexhaustible: it positively increases and multiplies the more we see of it and the more of us see it. So it has good claim to be considered the most valuable characteristic of the Earth” [Ibid, 4].

Following World War 1, the Council for the Protection of Rural England was formed, together with similar organisations in Wales and Scotland to safeguard the productivity and beauty of the countryside.

In the 1930s, a British geographer, Dr. Vaughan Cornish [who had been in Younghusband’s audience], responded to Younghusband’s challenge and wrote extensively and essentially descriptively about scenery, but all of his books are “now neglected by aestheticians and geographers alike” [Fuller, 1988]. His books on landscape included: The Poetic Impression of Natural Scenery [1931], The Scenery of England [1932], Scenery and the Sense of Sight [1935], The Preservation of our Scenery [1937], The Scenery of Sidmouth [1940], and The Beauties of Scenery, a Geographical Survey [1943]. Cornish wrote that the:

“combination of the English village, with the setting of field and hedgerow and coppice, is an Arcadian scene unrivalled elsewhere in Great Britain and unsurpassed in any part of the world” [Cornish, 1934].

In the period leading up to and after the World War 2, when the national parks were being established, Cornish’s works had some influence [Appleton, 1975].

In 1932, the Town and Country Planning Act gave local councils power to preserve scenic amenity. Although the World War 2 was a period of immense upheaval, it was also a period in which the English realised that their post-war society needed to change from what had gone before and to prepare for post-war reconstruction.

“A newer and better Britain was to be built. The feeling was one of intense optimism and confidence. Not only would the war be won: it would be followed by a similar campaign against the forces of want. That there was much that was inadequate, even intolerable, in pre-war Britain had been generally accepted. What was new was the belief that the problems could be tackled in the same way as a military operation.” [Cullingworth, 1985]

In 1942 an official inquiry on rural land use (Report of the Committee on Land Utilisation in Rural Areas (Scott Report), Cmd. 6378, HMSO, 1942) recommended the establishment of national parks for the enjoyment of the whole nation. The inquiry led to the 1944 White Paper, The Control of Land Use which referred to the establishment of national parks as part of a comprehensive post-war plan [Cullingworth & Nadin, 1994]. In 1945, the Dower Report (National Parks in England and Wales, HMSO, Cmd 6628, HMSO, 1945) defined national parks as “an extensive area of beautiful and relatively wild country, in which, for the nation’s benefit ... the characteristic landscape beauty is strictly preserved...” [Cullingworth, 1985]. The definition also provided for public enjoyment, wildlife, cultural heritage and farming. Dower also proposed protection for areas of high landscape quality.

The focus on landscape beauty during the wartime is striking, perhaps reflecting a deep psychological comfort associated with the character and perceived beauty of their country during the trauma and hardship of war. 

While Dower’s emphasis was on “relatively wild areas of moor land and rough grazing” [Mosley, 1975, 66], the Hobhouse Report (Ministry of Town & Country Planning, 1947. Report of the National Parks Committee (England and Wales) HMSO), which followed in 1947, saw national parks as being larger and also covering areas of countryside, which had been changed by the imprint of human development and use. Hobhouse called Dower’s idea of areas of high landscape quality “conservation areas”.

In 1949, the National Parks and Access to the Countryside Act [the NPAC Act] was proclaimed under which, between 1951 and 1957, ten large tracts of private land were designated as national parks. The designation of areas as national parks was based on their perceived natural beauty and recreational potential. Natural beauty was defined as including “scenic beauty, flora, fauna and geological and physiographic features” [Mosley, 1975] - certainly a generous description extending well beyond landscape quality.

The NPAC Act incorporated Hobhouse’s concept of conservation areas with the power to designate Areas of Outstanding Natural Beauty [AONB] but it did not provide criteria for their selection. Although of comparable landscape quality with national parks, AONBs were generally smaller, less suitable for outdoor recreation, and did not include extensive areas of open countryside [Robinson, et al, 1976, 20]. While the National Parks comprised mainly highland landscapes, AONBs comprised more densely settled lowland landscapes.

By 1991, the national parks totalled 14,011 sq km, or 9% of England and Wales, while the 40 AONBs totalled 20,449 sq km or 11.6%. The parks and AONBs include about one-third of the coastline of England and Wales.

The system of national parks over private land and the designation of AONBs is a uniquely British solution to the competing desires to protect high quality environments with the need to provide food and fibre for a large population. The early emphasis on landscape parallels the experience in other countries. Scenic preservation, along with provision for public enjoyment of the parks, were the major initial reasons for the creation of national parks; concern about the protection of flora and fauna was generally a later factor.

Meanwhile in 1947, the Town and Country Planning Act had provided for the designation by councils of Areas of Great Landscape Value within council development plans. Areas were defined subjectively through surveys by one or two individuals, modified through discussions in-house and the results often varied widely from county to county, which became apparent at the county boundaries [Robinson et al, 1976].

The first real attempt to move beyond mere descriptions of the landscape and to analyse the British landscape more rigorously began with the work of David Lowenthal, a researcher with the American Geographical Society, and Hugh Prince, a geographer at University College, London. In two seminal papers, The English Landscape [1964] and English Landscape Tastes [1965] they described the content of the English landscape and English landscape preferences.

Lowenthal and Prince identified variety, openness and atmosphere as key visual qualities of the English landscape. They referred to it as “altogether so tamed, trimmed, and humanized as to give the impression of a vast ornamental farm, as if the whole of it had been designed for visual pleasure” [1964,, my emphasis].

In their later paper, Lowenthal and Prince identified components of what they considered epitomised the English landscape preferences: the bucolic, the picturesque, the deciduous, the tidy [i.e. order and neatness], façadism, antiquarianism (rejection of the present, the sensuous and the functional; having historical associations), and Pope’s genius loci - the spirit of the place. The list was derived from the authors’ interpretation of the literature and embodied the “past and present virtues of the inhabitants” [Ibid, 186].

In the late 1960s different approaches were developed by Fines [1968], in a survey of the East Sussex landscape, and by Hebblethwaite in a survey of the East Hampshire AONB [Hampshire C.C. et al, 1968]. Fines’ work in particular, though not without its critics for its subjectivity (e.g. Brancher, 1969), was influential in encouraging county councils to initiate similar surveys [see Penning-Rowsell, 1974]. A more sophisticated and objective study based on component measurement and statistical analysis was undertaken of the Coventry-Solihull-Warwickshire landscape in 1968 [Study Team, 1971].

The Forestry Commission in Britain acquired extensive tracts of barren highland areas on which it planted softwood plantations. This action provoked continuing controversy on landscape grounds:

”no other single issue has raised so much controversy as the conifer plantations in the Highland zone of Britain. Everywhere they have been condemned as unsuited to the landscape...” [Simmons, 1965]

Eventually, the pressure was such that the Commission engaged the leading landscape architect, Sylvia Crowe, to advise it [Crowe, 1966].

During the 1980s and 1990s, the Countryside Commission sought to fulfil its statutory obligations for maintaining natural beauty; it embarked on a series of studies. It identified the extent of change of the physical landscape, for example, finding that at the end of World War 2 woodlands covered 7800 square miles while pine forests were only 400 square miles. However by 1980, woodlands had contracted to 3000 sq miles and pine forests had expanded to 1600 sq miles. Hedgerows, which provide distinctive boundaries of fields, had reduced from 500,000 miles to 390,000 miles [Countryside Commission, 1986].

Following this, the Commission issued Landscape assessment: a Countryside Commission approach [CCD 18, 1987]. The approach described the landscape character of areas and essentially comprised the subjective assessment of individual assessors. The method involved [Meredith, 1987]:

  • defining the purpose of the assessment and set criteria for judgement
  • compiling known information about the area
  • travelling throughout the area, recording observations, recording systematically what is seen, including sketches and descriptions
  • analysing what makes the landscape special and different from others
  • evaluating the landscape against the criteria set, for example an assessment of a landscape’s capacity to accommodate a proposed development
  • recording decisions arising from the analysis

Top of next column

By 1993 the application of the approach was judged a success [Martin, 1993]. The Commission had updated and expanded it in 1991 in CCP326: Environmental assessment:  the treatment of landscape and countryside recreation issues.

Concurrently the Commission embarked on a project called the New Map of England, which aimed to identify, describe and analyse landscape types at a broad regional scale. The Commission piloted the approach in southwest England before launching it across the country. It was expected to produce 150 maps, each with a detailed analysis of its landscape character.

The approach of the Countryside Commission focused solely on landscape character, assuming this to be a surrogate for landscape quality. Insofar as landscape quality is addressed, it is treated entirely  subjectively and descriptively by individual assessment.

In 1999, the Countryside Commission became the Countryside Agency and subsequently, Natural England.

United States

In contrast to England, the movement earlier this century to establish and protect national parks was driven largely concerns about scenic preservation. The loss in 1913 of Hetch Hetchy Valley in Yosemite National Park for a dam to provide water for San Francisco spurred the preservationists to gain wider recognition of the natural scenic attractions of the national parks. Arguing against the utilitarians who wanted to use the water, timber and other resources of these “waste” areas, the scenic preservationists argued that scenery also brings dollars. In 1910 there were some 20 distinct organisations directly advocating scenic protection [Runte, 1979].

“Let it not be forgotten that Switzerland regards its scenery as a money-producing asset to the extent of some two hundred million dollars annually”, said Allen Chamberlain, an advocate from New England. He and others argued that Americans should see the beauties that America had to offer first, before spending their money overseas, thus combining patriotism, aesthetics and economics. Chamberlain’s figures were cited in the Senate in arguments for the Glacier National Park in 1910, and five years later the figure being spent overseas by Americans was said to have soared to $500 million annually. Congressman Taylor argued that Switzerland gained between $10,000 and $40,000 per square mile of scenery per year and that America stood to gain much more..

Alliances were struck between the park authorities and railroad companies to provide better access to the parks. As early as the 1870s, the Northern Pacific railroad company had endorsed scenic protection, not for altruistic reasons but rather to promote tourism and patronage of their lines. In the post World War 1 era, private automobiles gradually overtook the railroads as the preferred means of travel, expanding access to the parks. While the interwar years saw the use of automobiles confined largely to the wealthier, following World War 2 the automobile moved from a luxury to a necessity.

In 1951, J.B. Jackson founded the periodical Landscape at the University of New Mexico, the journal played a key role in influencing a new generation about landscape aesthetics. This period saw the spread of cities and expanded industrial production, often accompanied by widespread environmental degradation and aesthetic loss. It was a period which demanded concern about the nation’s landscape.

David Lowenthal joined other critics of the degradation of the American landscape:

“The pristine landscape of aboriginal America was a fitting home for the brave and the free. But the brave was exterminated, and freedom became license; far from appreciating the glorious wilderness, the pioneer tore it apart and replaced it with a sordid landscape designed solely for profit.” [1966]

Peter Blake’s God’s Own Junkyard: the Planned Deterioration of America’s Landscape [1964] was, according to the author, “written in fury” [Lowenthal, 1966]. Similar books documented the wasteland of the American landscape. These included: Christopher Tunnard & Boris Pushkarev, 1963. Man-made America: Chaos or Control?; Stewart Udall, 1964. The Quiet Crisis; and Ian Nairn, 1965, The American Landscape: a Critical View.

In 1962, Stewart Alsop wrote:

“Out of the frontier past has grown a sub-conscious consensus that there is something manly about messiness and ugliness, something sissified about whatever is handsome, or well ordered, or beautiful.” [Lowenthal, 1966]

Numerous books have chronicled the origins of American’s love-hate relationship with the environment, and landscape in particular including: Michale Conzen [Ed], 1990, The Making of the American Landscape, Unwin Hyman and Mick Gidley & Robert Lawson-Peebles, 1989, Views of American Landscapes, CUP.

Concern about environmental blight did not go completely unheeded. In February 1964, President Johnson, partly at the urging of his wife, Lady-Bird Johnson, delivered to Congress a Message on Natural Beauty, a program to beautify America [Jackson, 1965]. Included were proposals to beautify rural and urban landscapes, the nation’s highways, and to remove billboards and automobile junkyards along the highways. J.B. Jackson was sceptical about the likelihood of anything permanent resulting from the program. He said that, in a country where “whatever is old is obsolete, and whatever is obsolete is discarded” [Ibid], the wonder of the American landscape is not that it contained such mess but that it also contained so many attractive suburbs and towns. In 1965, the President convened the White House Conference on Natural Beauty [Beauty for America, 1965] to stir the nation to action.

Following his English model, Lowenthal [1968] identified the following characteristics of the American landscape:

    • size - the sheer vastness of the land
    • wildness - relative to European this is particularly apparent to visitors
    • formlessness - “compared with Old World landscapes, those of America appear generally ragged, indefinite and confused; parts stand out at the expense of a unified whole”
    • insiders and outsiders - viewing the landscape, not as visitors but as inhabitants engaged in its development
    • the present sacrificed to the glorious future - a traveller in 1837 noted “they do not love the land of their fathers, but they are sincerely attached to that which their children are destined to inherit”
    • the present diminished by contrast with an idealised past - the Disneyland image of history
    • individual features emphasised at the expense of aggregates
    • the nearby and the typical neglected for the remote and the spectacular - the prominence given to National Parks
    • scenic appreciation serious and self-conscious - landscapes improved and signposted

    In his whimsical paper “You’ll Love the Rockies” [1983], J.A. Walter [an Englishman] recounted his impressions of the American landscape, contrasting it with that of England and the Continent. He found the vast scale of the American landscape daunting, yet the high position of the sun actually served to flatten the landscape compared with the low sun in England that emphasised the smallest undulation. The forest trees of America he found boring and frustrating, in that they blocked the view - yet Americans obviously loved them. In contrast to European landscapes that comprise a delicate balance of forest and village, mountain and meadow, he found American landscapes comprising vivid contrasts - rock spire and desert, ice-clad volcano and forest. The American landscape comprises large-scale examples of pure landscapes - all desert, all forest, all mountain ranges, each of enormous extent, in contrast to the incredible variety apparent in English and European landscapes in small areas.

     Australia

    The appreciation of the Australian landscape was slow to develop. Initially settled mainly with convicts from England and people from various European countries escaping religious persecution [e.g. Germans in Barossa Valley and Hahndorf in South Australia], the antipodean landscape was viewed with eyes used to the temperate climate and soft folds of the English or European lowlands. With the priority being survival, this drove the exploitation imperative until well into the 20th century. The prevailing ethos was “if it moves, shoot it; if it doesn’t, cut it down”.

    Massive change to the original Australian landscape ensued, with the felling and ringbarking of the forest trees, the drainage of swamps, the opening up of roads and railways, and the settling of towns and cities. Grazing by introduced stock, as well as by feral pests such as rabbits, removed the shrub layer and native grasses. Under Aboriginal occupation, fire had been used regularly to open the vegetation and drive out game. With European settlement, fires were controlled when they occurred but were often devastated the land and its inhabitants.

    This is not the place to survey the attitudes of the explorers and settlers to the Australian landscape (See "The Explorer's Eye: seeing new landscapes" paper in Landscape Miscellany part of this website) or to trace the way in which artists and writers have interpreted it. However it is worth noting that it was not until the 1880s that an Australian-born view of the landscape emerged, epitomised by the Heidelberg school of painting which established a distinctively Australian ‘feel’ to their landscape paintings. In a short space of a few years, the en plein air style of painting produced by artists such as Charles Conder, Frederick McCubbin, Arthur Streeton and Walter Withers helped to transform the way in which Australians viewed their landscape.

    The Sydney periodical, The Bulletin, which was founded in 1880, quickly established itself with a “character of outspokenness, incisiveness, and sardonic radicalism” [Heseltine, 1988, 1789]. Complementing the impact of the Heidelberg school of painting, The Bulletin published works by the growing band of Australian writers during the 1890s, including works by Henry Lawson and Banjo Patterson. The bush ballad was a major form of writing at that time. The characteristic attitude of the period was “ardent patriotism, the equally ardent socialism, the belief in mateship, and the superiority of bush life to that of the coastal cities.”

    The Australian art historian, Bernard Smith, has remarked that the idealisation of rural labour was a global phenomenon of the late 19th century and, although by then most Australians lived in the cities, they identified themselves “with the life and attitudes of the Australian rural worker”. The “frontier exercised an enormous influence upon the imagination of all Australians.” [Smith, 1971].

    Thus through art and the written word, the landscape and bush became an Australian icon, representing the best or the ideal, cloaked by the rose-tinted glasses of idealism and patriotism. That this conception of Australia was largely accomplished in the decade 1885 - 1895 is remarkable.

    With the beginning of the 20th century, together with Federation of the nation in 1901, a new confidence was apparent, one that was built on the image of Australia that had already been established. Paradoxically, the century has witnessed the gradual growth of the cities at the expense of rural areas and although Australians still exhibit nostalgia for the bush, and experience it in the comfort of their air-conditioned 4WDs, the reality of the bush is something remote to most Australians.

    Nevertheless a distinctive love for the Australian landscape is apparent, evident in the popularity of tourism and recreation to experience it [however remotely], of the abundance of beautifully illustrated books, calendars, videos and films of the landscape, the extent of its use as iconic symbols in advertising, and of the many conflicts which have occurred when forestry, mining and other forms of ‘desecration’ of the landscape have been proposed.

    As in England and the United States, Australia initially established national parks to protect outstanding scenery and to provide areas for public enjoyment, with protection of flora and fauna a more recent motivation. Unlike the English model, there are no nationally designated areas of outstanding beauty. However, many State and local government planning strategies provide for development control of developments that might degrade or impair the landscape [e.g. Haynes, 1975,Di I ; O’Neil, 1975].

    Prior to 1965 there had been no attempt to analyse the Australian landscape in an aesthetic or visual quality sense, although there had been numerous works that had examined it from a landform or biological sense - e.g.  C.F. Laseron’s Face of Australia.

Typologies of Lan:dscape Studies

This section reviews typologies or classifications of the international landscape studies. It reviews current typologies and then presents an alternative typology. A more extensive review of the typologies is available here. Table 1 summarises a range of typologies of landscape studies that have been proposed.

Table 1 Typologies of Landscape Quality Studies

Author Classification of paradigm or model
Penning-Rowsell 1973
  • Independent of landscape users – preferences defined
  • Dependent on landscape users – preferences observed
Brush 1976
  • Preferential judgement – likes and dislikes
  • Comparative appraisal – context of group preferences
Dearden 1977
  • Measurement of landscape’s physical attributes
  • Preferences of observers from photographs
Arthur, Daniel & Boster, 1977
  • Descriptive inventories
  • Public preference models
  • Economic analyses
Dearden, 1980
  • Field-based methods – physical descriptions
  • Surrogate methods – use of photographs
  • Measurement methods – improved field measurements
Penning-Rowsell, 1981
  • Intuitive methods
  • Statistical sophistication
  • Landscape preference approaches
Porteous, 1982
  • Humanists (or theorists)
  • Environmental activists
  • Experimentalists
  • Planners
Punter, 1982
  • Landscape perception – psychology of perception
  • Landscape interpretation – meaning in landscapes
  • Landscape quality – visual quality and formalism
Fenton & Reser, 1988
  • Objective measurement of physical attributes
  • Ratings of landscape variables
  • Description in phenomenological terms
Dearden & Sadler, 1989
  • Based on landscape judgement as external (E) to observer (i.e. objects) or internal (I) to observer (i.e. perceptual, affective, cognitive). Where E>I consensus will be high – objectivist; where I>E consensus low – subjectivist.
Gobster & Chenoweth, 1989
  • Physical descriptor type
  • Artistic descriptor type
  • Psychological descriptor type
Dakin, 2003
  • Expert, experimental, and experiential approaches

In the early 1980s, two key typologies of landscape studies were published:

  • Zube, E.H., Sell, J.L., and Taylor, J.G., 1982. Landscape perception: research, application and theory (Landscape Planning 9, 1 – 33)
  • Daniel, T.C. and Vining, J., 1983: Methodological issues in the assessment of landscape quality. (In: Altman, I. and Wohlwill, J.F. (Eds) Behavior and the Natural Environment. Plenum Press, New York

These typologies were remarkably similar though derived independently. They are compared in Figure 1.

Typologies

Figure 1 Comparison of Landscape Typologies

Broadly these typologies identified a continuum: measuring physical characteristics directly from the landscape, through measuring the human perception of these characteristics in aesthetic terms, to identifying the feelings and experiences of interacting with the landscape.

  • Using the ecological model, an expert assesses the natural amenities of the landscape based particularly on its naturalness.
  • The expert paradigm and formal aesthetic model assess the landscape’s formalist features including line, form, colour and texture by persons skilled in making such judgements, generally landscape architects.
  • In the psychophysical paradigm/model quantitative relationships are established between physical features and human responses through analysis of observers’ preferences.
  • In the cognitive paradigm/psychological model the feelings and perceptions of people who interact with the landscape are measured together with the meaning that landscapes hold for people.
  • The experiential paradigm/phenomenological model measures the individual experience of the human-landscape interaction, a person’s subjective feelings, expectations and interpretations in encountering the landscape.

Table 2 summarises the ten typologies that have sought to classify the landscape studies. It lists them under the four paradigms defined by Zube, et al, 1982 for comparison. In some cases it is difficult to assign the typologies as the descriptions used differ greatly, however the Table indicates my best judgement as to their placement. Most of the typologies define only two or three categories and in some instances, several of these amount to the same thing, namely the psychophysical paradigm.

Table 2    Summary of Landscape Analysis Typologies

Author Expert Psychophysical Cognitive Experiential
Penning-Rowsell, 1973 Independent Dependent on users    
Brush, 1976   Comparative appraisal Preferential
judgement
 
Dearden, 1977   Measurement techniques Preference
techniques
 
Dearden, 1980 Field based Surrogate methods
Measurement techniques
   
Arthur et al, 1977 Descriptive inventories Public preference models    
Penning-Rowsell, 1981 Intuitive methods Statistical sophistication
Preference approaches
   
Porteous, 1982 Planners Experimentalists   Humanists
Punter, 1982 Landscape quality Landscape perception   Landscape
perception
Daniel & Vining, 1983 Formal Aesthetic Psychophysical Psychological Phenomen-
ological
Fenton & Reser, 1988 Objective measurement                Normative judgements   Phenomen-
ological
Gobster & Chenoweth, 1989 Physical
Artistic
  Psychological  

Note: Arthur et al, 1977 also include economic analyses. The classification by Porteous [1982] is of the type of researcher rather than the product of their work. His activist category does not fit any of the above paradigms. Daniel & Vining [1982] also included the ecological model.

Go to top

Characteristics of Landscape Studies

In this section the characteristics of studies of landscape preferences undertaken over the period 1966 - 1996 are described. The findings of these studies are described elsewhere. Over this 30-year period, some 191 separate preference studies have been published describing a total of 227 surveys.

For the purposes of this section, studies refer to the complete paper describing the research while surveys refer to individual parts of the research project. 

Particular emphasis is given to the research instruments developed and used as the effectiveness of these determines the understanding gained regarding landscape preferences. The characteristics examined here are (Click on topic of choice):

The findings of the studies are presented subsequently.

The compilation covers studies that have as their subject non-urban landscapes; it does not cover studies of wilderness, recreation or urban landscapes, except where these relate directly to rural or natural landscapes [e.g. of parklands]. Each of these areas comprises an extensive literature in their own right and while some benefit would derive from their inclusion, it is beyond the scope here. Descriptions of these landscape preference studies may be found here.

Year of Studies

Data on the year in which the papers on landscape preference studies were published [Figure 2] indicates that the eleven year period, 1979 - 89 was the period of greatest activity accounting for 60% [114 studies] of the total.

year of studies

Figure 2    Year of Landscape Preference Study

The reasons for the decline in research over recent years is unclear and may be due to several factors: a belief that the field had been thoroughly researched, leaving nothing of significance to find, that the field remains incomprehensible and that further research is unlikely to pay dividends, or perhaps something as simple as the principal researchers moving on to other areas of research.

Go to list

 Location of Studies

Table 3 summarises the location of studies by country. The United States has dominated in studies of landscape preferences, accounting for 71% [159] of those studies. This table includes eight that compare preferences of two countries [e.g. Australia and Italy], so both countries are counted.

Table 3 Country of Study Location

Country or Region Number of Studies
United States
135
Canada
8
England & Scotland
17
Spain
7
Other Europe
11
Asia
7
Australia
10
New Zealand
2
Africa
1
Total
191

Sweden accounted for four of the studies in the rest of Europe [it is probable that there are more studies in languages other than English] and Thailand accounted for four in the studies in Asia. The predominance of studies in the United States reflects the degree by which quantitative studies have become standard in that country compared with other countries where qualitative studies were more prevalent.

Go to list

Principal Researchers and Centres

Most of the prolific researchers have been American. On the basis of the number of papers published on surveys the following were the most prolific:

  • Gregory J. Buhyoff, Associate Professor of Forestry & Quantitative Methods, School of Forestry, Virginia Polytechnic Institute & State University, Blacksburg, Virginia [19 studies].

  • R. Bruce Hull IV, Associate Professor in Dept of Forestry, Virginia Polytechnic Institute & State University; Doctorate at Virginia Polytechnic Institute under Buhyoff and post doctoral work in Australia. Following a stint as Associate Professor, College of Architecture, Texas A & M University he moved to his current position [13 studies].
  • Ervin H. Zube (late), originally at the Institute for Man and Environment at the University of Massachusetts, Amherst and later at the Landscape Resources Division in the School of Renewable Natural Resources, University of Arizona [10 studies].
  • Terry Daniel, Professor of Psychology, University of Arizona, Tuczon, Arizona [10 studies].
  • Herbert W. Schroeder, an environmental psychologist with the Forest Service at the North Central Forest Experiment Station, Chicago [9 studies].
  • J. Douglas Wellman also an Associate Professor of Forestry, School of Forestry, Virginia Polytechnic Institute. Often published as second author with Buhyoff [8 studies].
  • Stephen and Rachel Kaplan, respectively Professor of Environmental Psychology and Professor of Psychology, School of Natural Resources, University of Michigan. S. Kaplan is the theoretician [3 studies] while his wife, R. Kaplan, has conducted empirical analyses  [5 studies].

It is noteworthy that all but one of these researchers were university based, Schroeder is the only one based in a Government research organisation. A leading centre has been the School of Forestry at the Virginia Polytechnic Institute & State University in Blacksburg, Virginia. The University of Arizona has also been another key centre with many other researchers.

Purpose of Studies

The purpose of the studies can be categorised into seven broad groups:

  1. Theory development and testing.
  2. Techniques for measuring landscape preferences - development, testing and refinement.
  3. Influence of human observer factors on preferences - factors such as culture, socio-economic factors and familiarity with the landscape. These are factors that are dependent on the person viewing the landscape. It also includes the assessment of landscapes by expert vs ‘lay’ observers.
  4. Landscape factors - factors influencing preferences that are dependent on the physical landscape, including naturalness, presence of development, complexity and mystery.
  5. Affective effects - the feedback of the landscape upon the observer.
  6. Influence of the mode of presentation mode of the landscape including photographs vs field observations.
  7. Assessment of the landscape qualities of specific landscape areas or features such as the coast or water.

Because many studies were assessed to have several purposes [e.g. development of techniques while also determining the influence of landscape factors on preferences], the total number of purposes exceeded the number of studies. Table 4 summarises the frequency of these purposes.

Table 4    Purposes of Preference Studies

Purpose of study

Number

%

Theory development & testing
10
2.9
Techniques for measuring preferences
65
19.0
Human observer factors influencing preferences
59
17.2
Landscape factors influencing preferences
170
49.7
Affective influence of landscapes on observers
13
3.8
Mode of landscape presentation
16
4.7
Specific landscape areas or features
9
2.6
Total
342
100.0

Table 5 summarises the findings of the studies. The empirical nature of the majority of surveys is apparent in these figures: 56% defined landscape perception factors; 24% assisted in technique development; and only 11% focussed on theory development.

Table 5 Assessment of the Findings of Surveys

Assessment Frequency %
Supported theory
21
7.0
Changed/refined theory
6
2.0
Opposed theory
7
2.3
Supported technique
51
17.1
Changed/refined technique
14
4.7
Opposed technique
7
2.3
Defined participant characteristics
26
8.7
Defined landscape perception factors
166
55.7
Total
298
100.0

Note: the findings of many surveys applied to more than one category.

Various techniques for assessing landscape preferences have been developed including the Scenic Beauty Estimation [SBE] method of Daniel & Boster, Shafer’s measurement of landscape photographs, visitor employed photography, landscape adjective checklist, and physiological measures.

Studies of how human observer factors affected preferences examined the influence of such factors as culture, education, age and familiarity. The influence of culture was the subject of 13 studies. A key issue was the extent to which so-called ‘expert’ preferences differ from the general population and 17 studies examined this issue. By far the principal focus of the studies was the influence of landscape factors on preferences. This aimed to identify what physical factors could explain landscape quality as assessed by subjects.

The extent to which the landscape itself affected observers was examined by a relatively small number of studies [13]. The principal issue in the presentation category was the use of photographs versus field visits for evaluating preferences and 17 studies examined this issue. Very few studies [9] actually aimed to assess the landscape quality of a given area based on preferences; this is surprisingly low as it might be expected to be one of the principal purposes for studying landscapes. However this is probably due to the academic nature of most of the research studies.

Theoretical Basis of Studies

Porteous’ pithy observation that landscape preference studies are “rampantly empirical” [1982, 63] is borne out by an analysis of the theoretical basis of surveys. Of 227 surveys, only 19% [43 studies] were considered to have a theoretical basis [Table 6]. The remainder were classed as empirical.

Table 6 Theoretical Basis of Surveys

Theoretical Basis Number of Surveys
Prospect & refuge theory [Appleton]
11
Information processing theory [Kaplans]
35
Habitat theory [savanna] [Orians]
8
Affective theory [Urlich]
5

Note: Some surveys included more than one theory

Some studies included a theoretical basis with an otherwise heavily empirical approach and test landscape preferences with reference to these theories. These were not included in the category relating to theory in Table 4. This category applied only where the primary purpose of the study was to develop or test theories and application of this criterion limited the total to the following six studies (Table 7).

Table 7 Studies to test theories

Study
Theory
Balling & Falk, 1982
Habitat theory
Clamp & Powell, 1982               
Appleton’s prospect & refuge theory
Gimblett, Itami & Fitzgibbon, 1985
Kaplans’ mystery component
Herzog & Smith, 1988
Kaplans’ mystery component
Kaplan, Kaplan & Brown, 1989
Kaplans’ four predictors
Naser, et al, 1983
Appleton’s prospect & refuge theory

Research Instruments

The form of research instruments used in undertaking research of landscape preferences is vital to understanding the findings of the studies. Landscape research studies often involve two sets of variables:

  • independent - the elements which comprise the landscape
  • dependent - the rating of the landscape’s visual quality

Studies may seek to relate the dependent variable - landscape preferences, to the elements of the landscape as shown by the independent variables. This was achieved through statistical measures including correlation, multiple regression and factor analysis. Other studies compared the preferences of one population sample with another, no independent variable was used.

Most of the following instruments were used to evaluate the dependent variables, measuring the preferences of observers. Shafer’s method of measuring the parameters in landscape photographs was one of the few methods available that provided an objective measure of the independent variables.

Landscape studies utilise a diverse range of instruments to measure the preferences of observers. Approximately 30 different types of instruments are identified in the studies [Table 8]. The majority use photographs or similar surrogates rather than requiring observers to view the landscape in the field because research has shown that surrogates can provide similar results as field assessments. While 20 surveys used field assessments, 218 used surrogates.

Top of next column

Go to top

Table 8 Research instruments to measure landscape preferences

Instrument
Surveys
%
Psychophysical Methods
 
 
Law of Comparative Judgement method
21
6.1
Scenic Beauty Estimation method
43
12.5
Surrogates
Rating of photographs
99
28.7
Paired photographs
17
4.9
Q sort of photographs
19
5.5
Visitor employed photographs
6
1.7
Film, video & computer graphics
13
3.8
Total Surrogates
218
63.2
Descriptive Methods
 
 
Semantic differential
19
5.5
Adjective checklist
12
3.5
Interview or questionnaire
33
9.6
Descriptive rating
3
0.9
Other Methods
Field assessment
20
5.8
Maps
5
1.4
Physiological tests
4
1.2
Miscellaneous
19
5.5
Independent variable measurement
Measurement of features on photographs
12
3.5
Total
345
100.0

Among the most sophisticated and solidly grounded techniques are the psychophysical methods. These include two key methods used in landscape assessment, known as the Law of Comparative Judgement [LCJ] method and the Scenic Beauty Estimation [SBE] method, developed from psychophysics.

Each of these research instruments are examined in detail in chapter 7 of my PhD thesis, available here.

Participants

The characteristics of the participants in the preference studies are summarised in Table 9.

Table 9 Participants in landscape surveys

Participants Frequency %
Tertiary students
129
41.1
General community
71
22.6
Visitors to park or site
35
11.1
Natural resource professionals
24
7.6
Design professionals
14
4.5
University staff
14
4.5
Landowners & residents
10
3.2
Children
7
2.2
Other
10
3.2
Total
314
100.0

Rather than use samples of the general community, which may be expected, a sizeable proportion of surveys utilised students from the university campus where the study was being undertaken. Reviewing the surveys, it is easy to gain the impression, that landscape preferences have been overwhelmingly determined by American students. Well over half [57%] of the surveys used students, and in 38% of surveys they provided the sole subject [Table 10].

Table 10 Student participation in landscape surveys

 Participants No %
Total surveys
227 100%
Students used in surveys 129 57%
Students used in surveys (US) 94 41%
Only students used 87 38%
Only students used (US) 63 28%

Of the surveys carried out in the United States [135], 41% used students, and in 28% of surveys students only were used. Although the proportion of surveys in the US that used students was lower than outside the US, in absolute terms the bulk of studies using students were US based.

Participant Characteristics

Studies varied widely in the extent to which they sought data about the characteristics of the participants they used in assessing landscape preferences, and sought to correlate these characteristics with the preferences.

Out of the 227 surveys, 142 [63%] sought no data from the participants, a surprisingly high proportion. Table 11 summarises the most frequent characteristics sought.

Table 11 Participant Characteristics

Participant Characteristics Frequency %
Age
49
22.7
Gender
49
22.7
Education & training
24
11.1
Employment
20
9.2
Socio-economic status
20
9.2
Childhood residence
19
8.8
Culture & ethnicity
8
3.7
Expert/non-expert
6
2.8
Race [i.e. black]
6
2.8
Other
15
6.9
Total
216
100.0

Note: Other included landowner [4], personality [3], resident/non-resident [3], visitor knowledge [2] and religion, marital status and distance travelled [1 each].

The first five characteristics in Table 10 - age, gender, education, employment and socio-economic status were the most frequently sought. Childhood residence or home origins were sought to relate preferences to where the participant’s early acculturation occurred. Studies examining the effect of culture on preferences and those examining the influence of race sought data on these factors.

Table 12 summarises the number of characteristics of the participants sought by the surveys. This indicates that most surveys sought relatively little data [usually none] on their participant characteristics which is perhaps surprising in preference studies.

Table 12 Frequency of Participant Characteristics

Characteristics Surveys
1 25
2 21
3 18
4 14
5 5
6 2
8 1
Total 86

In some cases certain characteristics may be implicit, because of where the study was done [e.g. age group, gender, education, employment, socio-economic status, culture, expertise]. The absence of explicit mention does not mean the characteristics were not taken into account.

Landscape Characteristics

The landscape characteristics were defined by interpretation of the purpose, scope and coverage of the surveys. Inevitably a degree of judgement and subjectivity is implicit in the characteristics identified for each survey, particularly where they covers a range of characteristics [Table 13].

Table 13 Landscape Characteristics Covered by Surveys

Landscape Characteristic Frequency %
Naturalness 138 19.2
Forests or trees 183 25.4
Water 89 12.3
Rural 87 12.1
Urban areas/ development 65 9.0
Mountains & rugged topography 57 7.9
Environmental damage, impacts 29 4.0
Coast or lake shore 24 3.3
Urban parks and gardens 17 2.3
Roads 13 1.8
Other characteristics 19 2.6
Total 721 100.0

Note: Other covers wildlife [3], distance [2], number of people [5], sound [4], air quality [1] and unstated [4]. Many surveys had multiple landscape characteristics.

Forest-based landscape research was dominant with the majority of surveys in US Forest Service areas that were subject to research to assist in their management to ensure cognisance was given to scenic values.

The environmental damage category was frequently associated with forests; for example, the research of Buhyoff on the aesthetic impact of southern pine beetle on ponderosa pines on forested lands. Some surveys covering urban parks and gardens were included where these addressed wider issues than just the urban environment; for example, testing of Appleton’s prospect refuge theory [e.g. Naser, et al, 1983] or where they formed part of a wider study [e.g. Kaplan & Talbot, 1988].

Naturalness, or the natural character of landscapes, was the next dominant theme, frequently in combination with forests, water, coast, mountains or other natural features.

It is perhaps surprising that mountains and similarly rugged terrain have not comprised a dominant theme in landscape research. Nevertheless, it was included with other characteristics in nearly 60 studies. The mountain landscapes included in the surveys were mainly in the US - the Rockies and the eastern Appalachians and Adirondacks. Other mountain landscapes covered included: Canary Islands [Bernaldez et al, 1987], Spain [De Lucio & Mugica, 1994], Bali [Hull & Revell, 1989], Thailand [Tips & Savasdisara, 1986]. Fines [1968] used mountain scenes from the Himalayas, the Alps and Britain in his survey. Apart from Fines’ survey, none used scenes from the well-known mountain landscapes of the Alps. Until this author completed his survey of the Flinders Ranges (Lothian, 2009), no Australian mountainous areas had been subject to preference studies.

Water was a major theme of landscape surveys because with few exceptions it has been found to contribute positively to landscape quality. Together with coasts and lake shores, water is an important aspect of landscape research.

Table 14 summarises the number of characteristics examined by the surveys. The number of characteristics examined is surprisingly small - over 80% examined four or fewer characteristics. This indicates that surveys have generally examined landscapes which are fairly uniform; for example, many examined just forests. A large number examined the forest, water and mountains combination.

Table 14   Number of Characteristics Covered by Landscape Preference Surveys

Number of characteristics Frequency %
1
44
19.4
2
34
14.9
3
50
22.0
4
56
24.7
5
28
12.3
6
11
4.8
7
4
1.7
Total
227
100.0

Landscape Representation

As has already been described, the majority of studies used surrogates of the landscape rather than survey participants in situ in the landscape, which is generally too difficult and expensive. Field surveys were used in 31 studies. Table 15 summarises the form of other representations of the landscape. The dominance of photographs as a representation medium is evident from these figures.

Table 15 Representation of Landscapes by Surveys

Representation Frequency %
Photographs
189
89.6
Film or video
8
3.8
Artists drawings
9
4.2
Computer graphics
4
1.9
Model
1
0.5
Total
211
100.0

Note: Many surveys used more than one form of representation.

The types of photographs - whether black and white, or colour are indicated by Table 16. This indicates the dominance of colour photographs.

Table16   Form of Photographs Used in Surveys

Form of Photograph Frequency %
Black & white
25
12.9
Colour
152
78.8
Unstated
16
8.3
Total
193
100.0

Note: Surveys covered includes 8 using films or video. One of the studies used both b/w and colour photographs in its surveys.

Statistical Analysis

Virtually all the surveys involved statistical analysis of the results. In some cases this was rudimentary - means and standard deviations, but many cases involved sophisticated forms of statistical analysis, including factor analysis and regression analysis. Many surveys used a range of analytical methods. Table 17 summarises the main forms of statistical analysis.

Table17   Forms of Statistical Analysis Used in Surveys

Statistical Analysis Frequency %
Correlation analysis
68
22.2
Factor analysis
47
15.4
Regression analysis
58
19.0
Analysis of variance
48
15.7
Chi Square
19
6.2
Cluster analysis
13
4.2
Multidimensional scaling
10
3.3
Other
43
14.1
Total
306
100.0

Note: many surveys used several forms of analysis.

The ‘other’ category included: correspondence analysis, sensitivity analysis, preference rating, Wilcoxon Rank Sum Test, Mann-Whitney U test, Johnson’s hierarchical clustering algorithm and Tukey’s Honestly Significant Difference test. Factor analysis and/or regression analysis was used in 86 [44%] of the studies.

 
 

© 2009-14 Scenic Solutions   Privacy   Terms of Service   Site Map   Contact us