Visual Impact Assessment

Introduction

This section describes visual impact assessment, firstly by examining what we mean by visual and environmental impact assessment (VIA/EIA). The historical background to VIA is discussed together with some of the politics involved and some issues concerning VIA. The second half of the section describes the VIA process - looking at methods of visibility analysis and examining the methodologies of zones of visual influence (ZVI) and viewpoint analysis.

Definition

There is no precise definition for Visual Impact Assessment (or VIA) as a single concept. But the term describes a systematic analysis of the possible impacts of environment resulting from a proposed development and the investigation of the means available to mitigate the effects of such proposals prior to implementation. Visual impact is defined as a change in the appearance of the landscape as a result of development which can be positive (improvement) or negative (detraction) (IEA and the Landscape Institute, 1995).

Visual and Environmental Impact Assessment

VIA forms part of the Environmental Impact Assessment. EIA was introduced in the USA in 1969, Europe in 1985 and formally in Britain in 1988. It also covers other environmental issues such as the effects of development on flora and fauna; noise, air and water pollution. Whilst a large proportion of these issues can be assessed in quantitative terms, visual impact is assessed largely by qualitative judgements, as it is concerned with the human appreciation of, and interaction with, the landscape.

In EIA reports the visual landscape is often described, based on its geomorphological features. Sometimes the visual impacts of the proposed development are evaluated but rarely is there a good visualisation of the project. There are three major problems in visual landscape evaluation: the technical problem of how to visualise possible changes in the landscape; the theoretical problem of how to evaluate scenic beauty; and the administrative problem of how to integrate visual aspects in the planning process (Lange, 1994).

In comparison with the other environmental issues, visual impact is perhaps one of the best publicised and most contentious issues, usually because of the adverse effect that a new development can have on natural or "unspoilt" landscape and the intense feelings of the public towards this subject (Fortlage, 1990).

At present, VIA studies are required to be carried out, as part of the application for planning approval, only for major developments such as power stations, wind farms, transmission lines in the countryside (Commission of the European Communities, 1985, cited by Flynn and Pratt, 1993). VIA studies may also be required for other types of development, but this is at the discretion of the local planning authority or the Secretary of State. While small-scale developments (such as houses and farm buildings) can also have dramatic effects on landscape quality, it is currently too costly and time consuming to conduct visual impact studies on such developments.

Historical background of VIA

There appears to be a misconception in the literature that the management of visual resource originated in the US with the legislation of National Environment Policy Act (NEPA) in 1969. In fact, visual resource management already had a long established tradition in the UK. It started with the publication of the Town and Country Act (1947), and the National Parks and Access to the Countryside Act (1949).

The Countryside (Scotland) Act (1967) and Countryside Act (1968) required every, "Minister, government department and public body to have regard to the desirability of conserving the natural beauty and amenity of the countryside in all their functions related to land" (Cullingworth, 1976 cited in Gillespie and Clark, 1979). These legislations empowered the local planning authorities to implement their policy on the preservation of visual amenity.

All developments must be approved by the local planning authority. Under the Development Control Process (DCP) system the authority had the discretion to insist on what documentation are required before planning approval will be granted. Britain persisted with the DCP system, despite arguments by many experts (Thorburn, 1978) that it can only be improved by using a more formal or systematic impact assessment process such as the VIA, which had already been introduced in the USA in 1969. These experts were concerned about the effects of new large scale developments on landscape quality and the ability of the design control process to take these effects into consideration. Unlike the DCP, there is a statutory requirement for VIA to be carried out on all large scale developments. Increased pressure for development in the countryside, finally prompted the central government to incorporate the formal methods and techniques of VIA into the development control process in 1988.

Laws and politics in scenic beauty

In Great Britain, political and economic policies are still the major influences on decisions about locations for new industrial developments. Already, much of its lowland landscape has been intruded upon by extractive industry and urbanisation, and its highland landscape by reservoirs for both water supply and hydroelectricity (Aylward and Turnbull, 1977).

Although local government has at its disposal sophisticated planning legislation, the community is still concerned that fundamental changes may occur in the physical and visual quality of their environment and often suspects that planning and consent may be given to a development without the full disclosure of effects on the community. Thus, local government in a rural area is often motivated by pressure groups and individuals to impose stringent planning conditions which ensure that both the developers and the community are aware of the effects of the development and the alternatives available. The presentation of the evidence must be in a form that can be clearly understood and assessed by all parties (Aylward and Turnbull, 1977).

In some countries the scenery is well protected. For example in the Swiss constitution it is stated that the scenery has to be taken care of and in the case of a great interest of the general public, it has to be preserved undiminished (Lange, 1994). In the state of Wisconsin, in the USA, scenic beauty has assumed an importance in the law that currently serves as a major consideration in many of the state's regulatory functions. In 1952, the State Supreme Court ruled that the "right of the citizens of the state to enjoy our navigable streams includes the enjoyment of scenic beauty". The court held that "the occupancy (by the public) is visual" and that indeed the enjoyment of the beauty of the land constitutes a legitimate public use of land whether or not the public is allowed to set foot on it (Bishop and Hull, 1991).

Issues in VIA

Regional and local issues

The visualization models must be reliable and generalisable, and must be able to convey their meaning to ordinary citizens. The systems should be able to accommodate detailed, fine-grained, data representations within coarse-grained data sets to support both regional and local aspects of modelling (Orland, 1992b).

Spatial, quantitative and qualitative issues

The visual impact assessment of a proposed development addresses three types of issues: spatial, quantitative and qualitative. Spatial issues include where the development is visible from or, more specifically, what or whom it is visible to; quantitative issues include how much of the development is visible, how much of the surrounding area is affected, and to what degree; and qualitative issues include the visual character of the development and its compatibility with its surroundings (Fels, 1992).

Basic functions of a VIA

Five basic functions are important in the VIA: clear identification of the various types of impacts; organisation of spatially and temporally dispersed inventory data; prediction of impacts based upon potential land use decisions; a usable interface between these functions and the planner/manager; and effective communication of potential impacts to the public and decision-makers (Bishop and Hull, 1991).

The Visual Impact Assessment Process

Types of Visibility Analysis

The principle of intervisibility states that visibility is determined in two ways either from the site or to the site, that is, if point A can be seen from point B then the reverse is true. Thus although a site's visibility is normally thought of in terms of it being viewed from outside its boundaries, the outward view from the site to adjacent areas can be adopted to simplify analysis (Aylward and Turnbull, 1977).

Different types of visibility analysis include intervisibility analysis to produce levels of visual impact, dead ground analysis, identification of the portions of the landscape forming a backcloth for the design object, situations where the design object appears above the landscape horizon and the identification of optimal location for vegetation screen placement (Kennie and McLaren, 1988). The ability to map all of the viewpoints may prompt detailed interactive investigations in portions of the landscape, and suggest which portions of a planned development are most problematic and which portions might be altered to reduce visual impact (Fels, 1992). Quantifying the area visible from any location is an objective measure of the extent to which a change in land cover will be visible to an observer (Miller, 1995). Critical portions of the development can be identified, critical viewpoints can be located (Fels, 1992).

Traditional techniques

Traditional manual techniques of visibility analysis have emphasised the zone of visual intrusion associated with site development. The most important variable determining this is the terrain or landforms surrounding the site. Frequently, it is the size of the area within which the installation can be seen that is important, in other cases the critical consideration is site visibility from individual locations, such as scenic drives or picnic sites (Selman et al, 1991).

Of course determining a site's visibility does not produce the design but it can go a long way to help the speedy evaluation of alternative ideas and solutions, something that does not normally happen with more laborious and graphic simulation methods (Aylward and Turnbull, 1977). GIS-visualisation goes beyond the simple ability to discuss single anticipated outcomes via traditional graphic tools or simple visibility analyses. It offers the opportunity to visualise relationships across time and space, and to explore more comprehensive ranges of possibility (Orland, 1994).

Local, wide area, analysis for viewpoints

An approach which uses GIS to assess resources from the perspective of recreation through evaluating scenery and visual impact has been described by Miller et al (1994). Three types of analysis are used: local analysis of scenery; wide area analysis; and analysis for viewpoints.

Local analysis of scenery uses visual impacts at particular locations and for particular scenes as well as three further factors: distance depth cueing and size perspective; angle of intersection of the view with the terrain; and land cover. The use of this method to visualise and measure the impact of forest land use change in an example scene shows that the visual impact of afforestation is generally low, being below the horizon and blending with the dark texture background of the mountains.

Wide area analysis uses two approaches. Viewsheds for selected visitor viewpoints are calculated to identify priority zones of visual importance to tourists. A census of the total area visible from all locations is also calculated. Analysis for viewpoints uses viewpoints to measure scenic potential by determining the number of viewpoints from which any terrain location is visible.

At present these GIS tools for the analysis of scenery are used for either historical evaluation of land use change (Gauld et al, 1991) or prediction of the impacts of future changes (Aspinall, 1990). The approach can also be linked to more traditional methods of landscape analysis, such as the use of questionnaires to assess perception of place (Miller et al, 1994).

Distant, close, panoramic, and corridor views

Four types of view can be analysed using using digital data: distant, close, panoramic and corridor (Miller, 1995).

Distant and close are terms describing a concept. They have been coined to enable scoring of the landscape with respect to observer value judgements and predict scenic value for units in which only the data on the landscape elements are known. They are views characterised by the distance of the horizon and the land immediately below the horizon, from the observer. A working definition of close may be one in which the observer can discriminate between features of interest; for example, distinguish between coniferous and deciduous trees or between woodland and heather moorland. The implications of this are that a view from the same locality, in a particular direction, may be categorised differently if the atmospheric conditions are dramatically different, and therefore accompanying a categorisation of view must be a statement of prevailing modelled conditions.

Corridors are characterised by the existence of lateral terrain features such as valley sides or woodland either side of the observer, constraining their view in a narrow field. Photographic panoramas are usually made up of a number of laterally overlapping photographs taken in a sequence around 360 degrees with the disjoins evident. In a digitally produced panorama the number of views generated can be significantly larger than those taken on the ground.

Mapping techniques

Intervisibility analysis is accomplished through many applications of viewshed mapping procedures: projective and reflective, individual and composite. The effects of temporary alterations to the topographic model can be seen and evaluated (Fels, 1992).

Projective and reflective mappings

Projective mappings are initiated from viewpoints within the development (inside looking out) while reflective mappings are initiated from viewpoints in the surrounding landscape (outside looking in). The objective of projective mapping is to reveal the extent of visibility of the development to its surroundings. The objective of reflective mapping is to determine whether, and to what extent, the development is visible from its surroundings. Reflective mapping is more similar to conventional viewshed mapping procedures than is projective mapping (Fels, 1992).

Single and cumulative mappings

Single viewpoints are useful in evaluating the effects of a specific component of the development. The use of multiple viewpoints produces composite intervisibility maps; true cumulative mapping is more useful than mapping from a predetermined set of points.

Cumulative maps portray the visibility of every point in the development with respect to every point in the landscape. If the object of study is a form of development, such as a landfill or an electrical transmission line, the study must include every model point within that landfill or along that transmission line. If the object of study is a visual resource, such as scenic river, the study must include every model point associated with that river. These maps are the most effective means for producing comprehensive appraisals of spatial and quantitative impact issues (Fels, 1992).

Colour coded impact maps

One version of such an evaluation can yield a qualitative impact, ranging, for example, from beneficial and benign to harmful and disastrous. You can produce an impact map using an evocative colour coding scheme, e.g. red for bad, green for good, yellow for neutral. (This colour scheme is by no means universal: no colour scheme is. But, as Ervin (1993) observes, within a culture of viewers used to traffic lights, it is easily understood.)

Patterns of impacts that have an element of geographic or spatial correlation (such as along drainage courses or ridge tops) became obvious in this approach and seem much more immediate when seen in pseudo-3D than when simply presented as a coloured impact map.

Zones of Visual Influence

Zones of visual influence (ZVI) and Viewpoint Analysis are two of the most common techniques or visual tools used by experts for assessing visual impact. The two techniques provide factual data about the visual appearance of the development in the context of the existing landscape. ZVI is a map-based method used to define the area of visibility (also known as the view envelope) of a proposed development within the landscape. Originally developed by Tandy (1971), it can be carried out manually.

The manual method has now been largely superseded by view area analysis software such as VIEW suite (Aylward and Turnbull, 1977; Turnbull Jeffrey Partnership, 1983), VIEWIT (Tlusty, 1979), GROUSE (Evans, 1984) and MAP - Map Analysis Package (Tomlin and Tomlin, 1981 cited by Hadrian et al., 1988). Some researchers have incorporated the digital terrain mapping software into GIS (Geographic Information System) for analysing landscape modification and mapping predicted visual impact levels of proposed development from surrounding areas of visibility.

As the visibility of the development and natural landscape features are illustrated on plan (i.e. map), sections and wire-line perspectives, the ZVI technique only allows the user to develop a limited understanding of the visual properties of the landscape and man-made structure. It is difficult for the expert to judge the impact of the development on actual existing landscape quality, when wire-frame perspectives only indicate whether the development is seen or not.

Viewpoint Analysis

In the viewpoint analysis technique, two or three dimensional representations of the landscape before and after proposed developments, are often produced. The common simulation tools used by experts for VIA are models, perspectives and photomontages as viewed from specific points in the landscape. The key viewpoints are identified or must be agreed between experts and the local planning authorities.

There are limitations to the use of these simulation tools in viewpoint analysis (MacAulay, 1988; Hershberger and Cass in Nasar, 1988). Scale models of proposed developments and surrounding landscapes, are not only expensive and time-consuming to build, but are also limited by the area they can cover, and how well they represent the actual settings. There is also the additional problem of simulating the human viewpoint, even with the use of a modelscope. Given that details of the landscape and structure are drawn by artists, perspectives may not be totally realistic in representing the actual environment. Therefore, even the best perspectives can suffer from a degree of bias.

Photomontages, on the other hand, can provide a much higher degree of realism provided the proposed development can be simulated and superimposed accurately onto photographs or slides of the existing landscape. This technique can, however, be slow and tedious when done manually. Fortunately, the limitations of manual photomontage techniques have largely been resolved, using computer simulation techniques. As a result, digital photomontages are now commonly used in viewpoint analysis in the VIA process. The conventional approach adopted by experts in VIA process typically consist of:

• a factual description of the existing site, based on surveys conducted on site and desk studies;
• a description of the proposed development;
• an analysis of the impacts of the development using zone of visual influence study (ZVI) and/or Viewpoint Analysis; and
• predicting the level of impacts. Impacts are classified as having "major", "some", "minor" or "no" significance.
• suggestions on the scope of mitigative measures (if necessary) to ameliorate the potential impact.

Example of a VIA study using ZVI and viewpoint analysis

A typical example of a comprehensive VIA study is the Nicholas Pearson Associates' (1993) assessment of the Bryn Titli windfarm for National Wind Power in Wales. The VIA statement contained a description of the physical features (such as topography and landuse), landscape characteristics, and visual policies (i.e. statutory and non-statutory landscape designations as described in Task Two) of the existing landscape in and surrounding the development site. The written description was supported with a visual presentation which included maps and photographs.

The mitigative measures involved configuring the layout and spacing of the wind turbines, to reduce the number of visible units; and the use of recessive colours against the skyline. The impacts of the proposed windfarm development were considered, in terms of its construction period and operational life. These were analysed using ZVI and Viewpoint Analysis.

In their ZVI study, a digital terrain mapping computer software was used to produce two sets of visibility maps of the development site. The first map defined the area of visibility of turbines up to rotor height; and the second, up to tip of the blade at its highest point. The areas of visibility were divided into 4 classes, depending on the number of turbines seen.

In viewpoint analysis, the impact of the wind farm development from key viewpoints were analysed using wire-line perspectives and photomontage (i.e. computer simulated model of turbines superimposed onto digitised images of the actual site). The expert's judgement (that the impact of the development was of minor significance) was based largely on the consideration of the number of visible turbines, their distance from key viewpoints, and the complexity, diversity and variety of the surrounding landscape.

The conventional method of assessing visual impact, exemplified by Nicholas Pearson Associates' VIA study is referred to as the "description and analysis approach" (Landscape Research Group, 1988). The prevalence of this approach in the VIA process can be attributed to the fact that these studies are carried out by members of the design profession, who have been trained traditionally and are, therefore, familiar with this approach.

Limitations of the Conventional VIA Process

There are several potential limitations of the conventional VIA process. The implementation of VIA using the descriptive and analytical approach has provided a systematic and structured process towards the assessment of visual impact. Whether its implementation has succeeded in improving, or preserving the visual quality, or ameliorating the potential intrusion of new developments in the countryside, remains questionable. Some critics have argued that, in reality, VIA studies have been inconsistent with prevailing public perceptions and reactions, to the impacts of development in the countryside (Brun, 1995).

The Decision Maker's View

It is interesting to look at some of the comments made by those who make the decisions on planning applications about VIAs. Some admit that the final decision "must be subjective even if informed by other material" and that their conclusions "are based principally on my site visits" and not on any evidence from the imapct assessment; another notes that "even with the aid of various techniques in presenting the evidence, it is very much a matter of judgement as to how much of the proposed development is likely to be seen". Others question how to interpret the data they receive - "(there) remains the question of calculating the significance of such information". Not all inspectors fully trust the VIA findings, such as "the photomontages tended to exaggerate the visual prominence" of the windfarms. However, the majority do agree that "surveys are a reasonable starting point from which to calculate the effects of the proposal on the landscape" (Jones, 1996).

Landscape Simulation Methods

A wide range of landscape simulation methods have been developed and implemented as tools for impact assessment. These include: plans, diagrams, elevations, perspective sketches, renderings, modified photographs (photo renderings and photomontages), slide projections, scale models, movies, videotapes and computer graphics (Oh, 1994). Photomontages are often used to get a subjective impression. However, it is a relatively expensive method and is restricted to fixed observer locations (Zewe and Koglin, 1995). Simulation techniques are further discussed in the section on Visualisation Tools.

Computer-based simulation

Computer-based simulation methods include: two-dimensional drafting and painting; three-dimensional wire frame models; surface and solid modelling; image processing; and animation techniques. Several reports have shown successful applications of advanced computer technologies such as three-dimensional solid modelling, coloured and dynamic functions (Oh, 1994).

Another possibility is the calculation of computer graphic images by projecting aerial photographs as a texture onto landscape polygons. This method is currently being developed in France by the EDF (Électricité de France). It delivers very realistic images particularly for distant areas, but not nearby the observer (Zewe and Koglin, 1995).

Criteria for good simulations

Sheppard (1989) states the following as criteria which good simulations should fulfil:

• Representativeness - a simulation should represent important and typical views of a project.
• Accuracy - the similarity between a simulation and the reality after the project has been realised. Of course, judging this is a little difficult beforehand.
• Visual clarity - detail, parts and overall contents have to be clearly recognisable.
• Interest - a simulation should hold the attention of the viewer.
• Legitimacy - a simulation is defensible if it can be shown how it was produced and to what degree it is accurate.

Visual simulation is only descriptive; it does not release the planner from the difficult task of evaluation nor does it provide an evaluation in itself in a publicly based evaluation approach. However, visual simulation is, or at least should be, the prerequisite to predict and to evaluate the visual consequences of planned alterations (Lange, 1994).

Changes due to the seasons

There is a need to visualise change over time, using colour cues and animation. To create different simulations for the three different seasons symbols can be substituted where appropriate (bare branched deciduous trees in the winter, sparsely foliated in the spring and autumn). Fields can appear tan in the autumn and white in the winter, and bright green/yellow/blue (depending on crop) in the spring. Other dimensions of the landscape - such as visibility distances, sun and shade pockets, or "sense of enclosure" or "view of water" also change with the seasons (Ervin, 1993).

References

Aspinall, R.J. (1990) An Integrated Approach to Land Evaluation: Grampian Region. In Evaluation of Land Resources in Scotland (eds. Bibby, J.S. and Thomas, M.F.), p45 - 56, Macaulay Land Use Research Institute and the Royal Scottish Geographical Society.

Aylward, G. and Turnbull, M. (1977) Visual analysis: a computer-aided approach to determine visibility. Computer-Aided Design, 9, 103-108.

Bishop, I.D. and Hull, R.B. (1991) Integrating technologies for visual resource management. Journal of Environmental Management, 32, 295-312.

Brun (1995) pers. comm.

Commission of the European Communities (1985) Council directive on the effects of certain public and private projects on the environment. Official Journal of the European Communities, No. L175.

Cullingworth, 1976 cited in Gillespie and Clark, 1979

Ervin, S.M. (1993) Landscape visualization with Emaps. IEEE Computer Graphics and Applications, 13, 28-33.

Evans, B. (1984) The development of a ground modeling system for William Gillespie and Partners. in Evans B (Ed), Computer aided landscape design: principles and practice, Proceedings of the Greenchips Symposium, 1984.

Fels, J.E. (1992) Viewshed simulation and analysis: an interactive approach. GIS World, July, 54-59.

Flynn, A. and Pratt, A.C. (1993) Costing the countryside. Journal of Environmental Planning and Management, 36, 3-13.

Fortlage, C.A. (1990) Environmental Assessment: a practical guide. England: Gower Publishing Co. Ltd.

Gauld, J.H., Bell, J.S., Towers, W. and Miller, D.R. (1991) The measurement and analysis of land cover changes in the Cairngorms. Report to the Scottish Office Environment Department and the Scottish Office Agriculture and Fisheries Department.

Hershberger, R. and Cass, R. (1974) Predicting user response to buildings. Environmental Design Research Association, 5, 117-134.

IEA and the Landscape Institute (1995) Guidelines for landscape and visual impact assessment. Londdon: E and FN Spon.

Jones, C. (1996) Chapter 8: The Decision Maker's View. MA Thesis, University of Salford.

Kennie, T.J.M. and McLaren, R.A. (1988) Modelling for digital terrain and landscape visualisation. Photogrammetric Record, 12, 711 - 745.

Landscape Research Group, 1988

Lange, E. (1994) Integration of computerized visual simulation and visual assessment in environmental planning. Landscape and Urban Planning, 30, 99-112.

MacAulay, 1988

Miller, D.R. (1995) Categorization of terrain view. In Innovation in GIS2 (ed. Fisher, P.), chap 17, p215 - 221, Taylor and Francis.

Miller, D.R., Morrice, J.G., Horne, P.L. and Aspinall, R.J. (1994) The use of geographic information systems for analysis of scenery in the Cairngorm Mountains, Scotland. In Mountain Environments and Geographic Information Systems (eds. Price, M.F. and Heywood, D.I.), Chap 7, p119 - 131, Taylor and Francis.

Nasar, J.L. (1988) Environmental aesthetics: theory, research and applications. Cambridge: Cambridge University Press.

Nicholas Pearson Associates, 1993

Oh, K. (1994) A perceptual evaluation of computer-based landscape simulations. Landscape and Urban Planning, 28, 201-216.

Orland, B. (1992b) Evaluating regional changes on the basis of local expectations: a visualization dilemma. Landscape and Urban Planning, 21, 257-259.

Orland, B. (1994) Visualization techniques for incorporation in forest planning geographic information systems. Landscape and Urban Planning, 30, 83-97.

Selman, P., Davidson, D., Watson, A. and Winterbottom, S. (1991) GIS in rural environmental planning. Town Planning Review, 62, 215-223.

Sheppard, S.R.J. (1989) Visual simulation: a users guide for architects, engineers and planners. New York: van Nostrand Reinhold.

Tandy, C. (1971) Landscape evaluation technique. Working Paper, Croydon, Land Use Consultants.

Thorburn, A. (1978) EIA - the role of the planning authority. Built Environment, 4(2), 122-128.

Tlusty, 1979

Tomlin and Tomlin, 1981 cited by Hadrian et al, 1988

Turnbull Jeffrey Partnership, 1983

Zewe, R. and Koglin, H.-J. (1995) A method for the visual assessment of overhead lines. Computers and Graphics, 19, 97-108.