8.1 INTRODUCTION
8.1.1 This Chapter presents the final Landscape Indicator developed during the Study. In particular, the Chapter addresses:
The background to the CASET indicators generally;
Alternative Indicators considered; and
The final Landscape Indicator proposed in the Study.
8.2 BACKGROUND TO SUSTAINABILITY INDICATORS
Development of Indicators in SUSDEV21
8.2.1 The Indicators developed under the SUSDEV21 study are aimed at providing decision makers with a means of capturing in a manageable form the full range of economic, environmental and social issues associated with a particular decision, and therefore contribute to defining and measuring progress towards sustainable development. For ease of use, and to encourage those responsible for developing projects and plans in the HKSAR Government to ‘think sustainably’, the indicators are incorporated within a decision support tool called CASET, which guides users through a detailed consideration of the effects their project or plan may have on life in Hong Kong.
8.2.2 In addition, by succinctly summarising the issues and concerns, the indicators provide an effective means of communicating and discussing the sustainability implications of different courses of action with stakeholders such as the public, the business sector and Non-governmental Organisations. This very much reflects the intention of Chapter 40 of Agenda 21, which identifies the development of indicators for sustainable development as a key activity for improving decision-making. Further background on Indicators is provided in Table 8.1.
8.3 POTENTIAL LANDSCAPE INDICATORS
Indicators for Landscape
8.3.1 As the UK Countryside Agency states, Landscape Indicators “need to provide a good indication of change in character…Key characteristics represent the essential character of individual areas but they are too general to act as indicators in their own right. Instead specific features or attributes need to be selected from the key characteristics. These need to be:
central to the distinctive character of the type or area;
liable to experience change either in extent or in their condition.
Indicators will need to be defined precisely. The desired direction of change of chosen indicators must be known.” (Countryside Agency, 1999, pp.82-83).
8.3.2 Two potential landscape indicators have been investigated and assessed. The two potential Landscape Indicators are explained in detail in this section, these being:
Indicator based on ‘Significant Landscape Features’ (with Advisory Mechanism);
Indicator based on ‘Areas of High Landscape Value’.
8.3.3 In the rest of this section, these two Indicators are described. A comparative assessment of the two Indicators is made and a recommendation provided as to the preferred Indicator.
Table 8.1 Background to Indicators
Indicators can be very simply described as quantified information which helps to explain how things change over time, and which can then used to assess how sustainable a society’s activities are over time. The role of sustainable development indicators in policy and project appraisal is particularly important since:
they provide a means of quantifying environmental, social and economic impacts of proposals;
they can help to provide clarity out of the mass of data available;
they can help to measure the extent to which policies aimed at sustainability objectives are being achieved; and
they can be useful in communicating impacts to a wider non-technical audience.
The indicators developed in the SUSDEV21 study represent the final, most specific, level in a process of focusing in from a broad definition of what sustainable development means for the HKSAR, through the elaboration of that definition in Guiding Principles, which express the sustainability aims for particular aspects of life in Hong Kong, such as the economy, or natural resource use, to the indicators themselves, which, as pointed out above, encapsulate a range of particular issues associated with each Guiding Principle, and quantitatively express them.
Current Operation of Sustainability Indicators
In the Spring of 2001, the Government inaugurated the Sustainable Development Unit (SDU) comprising a small team of sustainability specialists, reporting directly to the Chief Secretary via the Director of Administration. SDU promulgated a Guidance Note, “Guideline on Sustainability Assessment” which has been distributed to all Bureaux and Departments. This does not mandate the use of the CASET tool but the SDU has devoted considerable effort to providing training in the use of the tool to Department and Bureau representatives, and in explaining its benefits.
Criteria for Adoption of Indicators
There are a number of attributes that influence how successful an indicator will be at achieving the aims outlined above. In an ideal world, all indicators would have all of these attributes, but that is very rarely achievable, and the pragmatic approach usually adopted is to use them as a guide when selecting indicators with the aim of capturing as many of them as possible. The attributes are that indicators should be:
simple and robust;
policy relevant (representative);
sensitive to change;¨ easily understood;
measurable;
have a target level or guideline against which to compare it;
capable of showing trends over time;
scientifically or analytically valid;
capable of being updated at regular intervals;
measurable through existing or readily available data; and
based on data adequately documented and of known and acceptable quality.
In addition to the ‘over-arching’ attributes above, the development of sustainability indicators for SUSDEV 21 was also guided by of the following study-specific criteria, which, again, would also be applicable to future indicators.
Capable of Prediction. Since the purpose of the indicators developed for the study was to provide the basis for the CASET decision support tool, it was important that they were formulated in such a way as to facilitate prediction of changes in their value resulting from the evaluation of strategic policies and projects.
Uni-Directional. Since the indicators are used to evaluate the implications of policy, it is important that changes in the indicator can be interpreted in a straightforward manner. For example the reason for a change in direction of some indicators, in particular those associated with resourcing, may be ambiguous. Taking an indicator on crime rate as an example, the cause of an increase in the indicator could be differently interpreted as either a worsening of the crime situation, or conversely an increase in crime detection as a result of greater police resourcing. Indicator formulation for the study therefore needed to be undertaken with care so that changes in the indicator can be linked to clear changes in underlying conditions.
Number of Indicators. In selecting indicators to represent the range of guiding principles, a significant hurdle was the requirement to have a sufficient number of indicators to be representative of the key sustainability issues, and at the same time to restrict the overall ‘pool’ of indicators to a workable total for use in the decision support tool, CASET. Where possible, the approach focused on identification of composite indicators which would be relevant to more than one guiding principle.
A further important point is that the indicators are intended to focus on “outputs” of environmental, social and economic change, rather than on drivers/causes of change or sectoral interests. This approach has a practical advantage in that it enables changes in different sectors (e.g. transport, waste management, natural resources etc) to be picked up using cross-sectoral indicators (e.g. air quality indicators, GDP or income differential) rather than developing less flexible sector-specific indicators.
More importantly, however, the “outputs” approach recognises that sustainability is concerned with taking a holistic view and that it is the impacts of change, rather than the changes themselves, that are important when establishing how the economy, community or environment is affected by a project or policy.
For example, a point raised frequently during the SUSDEV21 public consultation process was why population was not adopted as a sustainability indicator. The reason is that changes in population and demographic trends are drivers of other effects (e.g. increased natural resource use, strain on community facilities) rather than impacts themselves. A particular population figure is not inherently unsustainable - what is crucial to sustainability are the consumption patterns adopted by a population and their effect on sustainability issues such as resource efficiency, social provision and pollution (1000 people from Los Angeles exert a very different level of pressure on sustainability than do 1000 from Lhasa).
Another important consideration in the selection of indicators for use in the system developed for the HKSAR is whether evaluative criteria could be identified for them. Evaluative criteria provide benchmarks or guidelines to enable the users of CASET to judge whether the change in an indicator, illustrating the impacts of a proposal being assessed, is in a positive or negative direction, and whether the amount of that change is significant or not. In this way, while not seeking to imply whether a change in an indicator is acceptable or not in the overall consideration of the project (this is the role of the decision-maker), the criteria provide information as to the relative scale of the change to an indicator. From this, they provide the decision-maker with a form of reference framework against which the changes to the various indicators affected by a project or policy can be judged.
The next step from the evaluative criteria is the setting of a target value for the indicator concerned, which is a specific threshold (e.g. relating to a statutory guideline or standard) or target for achievement, against which the change in the indicator can be evaluated. The target value can thus be an aspirational figure, either reflecting a level of improvement, or a level of recovery or restoration, to be achieved by some point in the future.
Potential Landscape Indicator - Significant Landscape Features
8.3.4 This Indicator would consist of two Indicators working in parallel. Both Indicators would monitor change in features with a positive effect on the character of landscape - what are termed Significant Landscape Features (SLFs).
8.3.5 SLFs are a measure of positive aspects of landscape character and are linked to both proven consensual and professional judgments as to their contribution to character.
The first Indicator will monitor change in Hong Kong's Area-based SLFs. Area-based SLFs generally include ‘natural features, such as woodland, streams, fishponds etc which are not edifices or constructions;
The second Indicator will monitor change in Point-based SLFs. These include buildings, edifices, structures or built landscape features such as monuments, bridges, landmark geological or physiographic features and hydrological features with limited horizontal extent such as waterfalls etc.
8.3.6 The reason for having two Indicators is that it is not possible to find a basis of measurement which will adequately apply to all types of SLFs. In particular, if one were to measure all SLFs on an area basis, this would not adequately capture the landscape contribution of small buildings or historic structures which may make a big contribution to the visual landscape, but which are very small in terms of area (especially when compared to natural SLFs). Therefore, those SLFs which need to be measured on an area basis and those which need to be measured on a numeric basis are separated out into different categories.
8.3.7 The Indicators would operate by measuring the net loss/gain in the area/number of SLFs of each type. (It is important to understand that there could potentially be a gain of SLFs). The proposed Indicators would therefore be:
"Percentage change to area of Area-based Significant Landscape Features" and
"Percentage change to number of Point-based Significant Landscape Features”.
8.3.8 A long-list of the two types of Significant Landscape Features is set out in Table 8.2 below. Included in the table are possible sources for the SLFs required to create the indicator layers. This is intended as a guide only for the party responsible for developing the Indicator in the future.
Table 8.2 Long-List of Significant Landscape Features (SLFs)
Significant Landscape Feature Unit of Measurement
Possible Source Comment
Area-based SLFs Physiography Natural upland topography (over 40mPD) Area Land Dept LIC data Will require some manipulation of source data Rocky shore (measured as being a horizontal distance of 20m from the coast) Area SDU Habitat Map layer‘Rocky Shore’ Will require some manipulation of source data Sandy beach Area SDU Habitat Map layer‘Sandy Shore’ Use-as-is Vegetation Woodland (defined as a group of trees no less than 100 sq.m.) Area SDU Habitat Map layers:“Fung Shui Forest’‘Lowland Forest’‘Montane Forest’ In order to operate Indicator, different woodland layers needs to be extracted. Scrubland Area SDU Habitat Map layers:‘Baekia Shrubland’‘Mixed shrubland’‘Shrubby Grassland’ As above Grassland Area SDU Habitat Map layers:‘Grassland’ Use-as-is Plantation Area SDU Habitat Map layers:‘Plantation or Plantation / Mixed Forest’ Use-as-is Urban parks and public open spaces Area PlanD OZP and/or LIC Parks Layer from 1:20,000 data In order to operate Indicator, data needs to be compiled Golf Courses Area Landscape Value Mapping Study Use-as-is Cultivation Area SDU Habitat Map layer:‘Cultivation’ Use-as-is Hydrology Natural stream (permanent) Area SDU Habitat Map layer‘Natural Watercourse’ Use-as-is Wetland (including freshwater marsh, salt marsh and mangrove Area SDU Habitat Map layer:‘Freshwater/brackish Wetland’‘Mangrove’‘Seagrass’ In order to operate Indicator, data needs to be compiled Modified Watercourse Area SDU Habitat Map layer‘Modified Watercourse’ Use-as-is Fish pond and Gei wai Area SDU Habitat Map layer‘Fishpond/Gei Wai’ Use-as-is Open coastal water Area Landscape Value Mapping Study (LCM Land/Water boundary) Use-as-is Inter-tidal mud flat Area SDU Habitat Map layer”‘Inter-tidal Mudflat’ Use-as-is Point-based SLFs Historic monuments, buildings, and other features with a visual manifestation; Number Compiled under Landscape Value Mapping Study Landmark buildings (modern buildings which have won HKIA or international awards) Number Compiled under Landscape Value Mapping Study Landmark structures (bridges, structures, etc which have been recognised as making a significant visual contribution to the landscape) Number Compiled under Landscape Value Mapping Study Landmark hydrological features (e.g. waterfalls) Number Compiled under Landscape Value Mapping Study Landmark geological/ physiographic feature Number Compiled under Landscape Value Mapping Study 8.3.9 The two Indicators will be entirely separate and no attempt will be made to aggregate them. Each Indicator expresses landscape change as a positive or negative change in the total area/number of each type of SLF.
Advisory Mechanism - Areas of High Landscape Value (AHLVs)
8.3.10 As well as the Landscape Indicators described above, a GIS layer should be created and included within the CASET system to allow CASET operators to check if development is proposed within one of the areas of landscape identified in the Study as being of 'high' landscape value, termed Areas of High Landscape Value (or AHLVs).
8.3.11 Areas of High Landscape Value (AHLVs) are those areas of Hong Kong's landscape which meet the highest standards of scenic quality and which are assessed as being 'High' value under the criteria set out earlier in this Report. It is important to realise that these represent the very best landscapes in Hong Kong. It is equally important to realise that AHLVs can equally be urban or rural landscapes. Therefore, there is no immediate correlation between naturalness and landscape value.
8.3.12 The AHLV would not form part of the Indicator per se, but would be included as a supporting layer of information visible when the Indicator is triggered as with other indictors in the CASET system such as the Area of High Ecological Value indicator. In this way, if development is proposed in an AHLV, the operator of CASET would be provided with a mechanism to check if the proposal falls within one of Hong Kong's best landscapes. It is important to note that the development would not necessarily be precluded should a proposal fall in an AHLV, but that a decision would have to be made as to whether to proceed, based on all other relevant factors (as well as the net loss/gain of SLFs).
Evaluation Criteria and Target
8.3.13 The evaluation criteria for the Landscape Indicators would therefore be:
“Area of change to Area-based SLFs” and
“Numeric change to Point-based SLFs”.
8.3.14 The Indicator Targets would be:
"No net loss to area of Area-based SLFs" and
"No net loss to number of Point-based SLFs."
8.3.15 It is important to note that these targets do not preclude loss of certain landscape features, nor do they preclude construction within AHLVs, merely that decision-makers should be aware of developments which affect landscape features which contribute positively to landscape character or which are to take place in Hong Kong's best landscapes.
Discussion of Landscape Indicator Based on SLFs (with Advisory Mechanism)
8.3.16 The following are the merits (or otherwise) of the SLF-based Indicator against the SUSDEV21 and UK Countryside Commission criteria noted above:
Simplicity – Conceptually the Indicator is fairly straightforward. Although having two Landscape Indicators is not an ideal situation, it is not unacceptable in CASET terms. Moreover, it is really the only way in which a representative value can be assigned to SLFs of different kinds.
Robustness – the SLFs which are objectively based are very robust. Those SLFs which contain a value component (e.g. Landmark buildings) are less so, although these are defined by reference to fairly well documented criteria.
Policy Relevant (representative) – the SLFs are representative of landscape character and value at a national/regional level as opposed to at a local level (the precise contribution to local landscape character/value is not specified). However, the fact that SLFs are comprehensive (and apply to landscapes of all values) mean that they are likely to capture policy change well. It can however be argued that no distinction is made between the quality of different SLFs.
Sensitivity to Change – the fact the SLFs are disaggregated makes them very sensitive to change. It can be argued that the area based character of the SLFs means that they do not adequately take account of the impacts of tall structures with a relatively small plan area on scenic or landscape value. Equally, it can be argued the removal of a point based SLF will not reflect the true change to the character or value landscape.
Easily Understood – as noted above, the concept of features which make a significant contribution to landscape value and character is a fairly straightforward one that is likely to be easily understood by the layman.
Measurable – SLFs are in principle easily measurable. Because all SLFs of each type need to be comparable, it is essential that they all be measured using the same units of measurement. It is acknowledged that this might lead to a certain degree of approximation in quantification of some SLFs, but because the Indicators (and CASET) are for use at a strategic and not a detailed level, this is not considered objectionable.
Have a Target Level or Guideline against which to Compare it – it is possible to establish targets and guidelines (by reference to e.g. existing area/number of SLFs).
Capable of Showing Trends over Time – again, it is possible to fulfil this criterion by reference to e.g. an area measurement of SLFs.
Scientifically or Analytically Valid – the methods by which area-based SLFs are defined or measured are fairly well recognised. More value-laden SLFs (such as Landmark Buildings, etc,) whilst rationally identified by reference to defined criteria, cannot be said to be wholly ‘scientific’ but rather based on well-documented consensual judgments by experts.
Capable of being Updated at Regular Intervals – data on most SLFs can be very easily updated by reference to data received in EIAs, or by updating the data base whenever comprehensive baseline data is updated by other Government Departments (e.g. Wetland Compensation Study). Data on point-based SLFs would be harder to track and update and would require review at regular intervals.
Measurable through Existing or Readily Available Data – All SLFs are already defined and measurable through existing data.
Based on Data Adequately Documented and of Known and Acceptable Quality – SLFs would be based on recognised Government data sources or data gathered under the Landscape Value Mapping Study. This is all Government data of known provenance and accuracy.
Multi-Directionality – The Indicators are in principle, dual-directional (i.e. they measure whether SLFs are increasing or decreasing). It must however be acknowledged that for Point-based SLFs, it is not possible to predict positive change (i.e. whether a proposed development will result in a net increase in landmark buildings/ structures/features). With regard to geological or physiographic features, these are to all intents and purposes, un-re-creatable. Positive change is therefore not realistically possible. With regard to landmark buildings/structures, it is not possible to predict positive change because the identification of such buildings/structures relies on their being recognised as of value (through the receipt of an award) which is something that cannot be predicted. However, this is not thought to be an objection to the Indicator, as the value of any building or structure is inevitably linked to the age in which it exists and its perceived value can only be recognised contemporaneously or retrospectively. That is to say, buildings and other structures do not have value until people/society decide that they do. Clearly the landmark status of such structures is to a large extent a product of prevailing current aesthetic perspectives and an Indicator based on retrospective recognition only, is probably as good as one can achieve.
Specific Features Central to the Distinctive Character of the Type or Area – SLFs are specific features which represent the basic building blocks of all landscapes. Specific SLFs are not identified for each LCA or LCT simply because it proved too difficult to define thresholds at which an LCA would pass from one LCT due to changes to its set of SLFs. However, because the Indicator is a strategic level one, this level of aggregation is deemed to be acceptable.
Potential Landscape Indicator – Areas of High Landscape Value (AHLVs)
8.3.17 The Indicator would operate by simply measuring the net loss/gain in the area of High Landscape Value (or High and High(Q) Landscape value, as may be determined/required). The proposed Indicator would therefore be "Percentage change to area of High Landscape Value”.
Evaluation Criteria and Target
8.3.18 The evaluation criteria for the Landscape Indicators would be “Area of change to areas of High Landscape Value”. The Indicator Targets would be "No net loss to area of High Landscape Value".
Discussion of Landscape Indicator Based on AHLVs
8.3.19 The following presents a discussion of the merits (or otherwise) of the AHLV-based Indicator against the SUSDEV and UK Countryside Commission criteria noted above:
Simplicity - Though superficially attractive as being simple to understand, the concept of ‘value’ underpinning the AHLV Indicator is based on numerous unstated complex assumptions and concepts.
Robustness – the concept of landscape value is a complex one. Whilst there is considerable consensus as to the usefulness of the concept and as to the existence of the concept in popular and professional thought and discourse, there is equally, considerable disagreement as to the rational or theoretical underpinnings of the concept. Therefore, any indicator based solely on the concept of value will inevitably be subject to question to some degree.
Policy Relevant (representative) – the concept of AHLV is not representative of the regional landscape as a whole (as it considers only certain areas or values of landscape as being worthy of analysis). The fact that this indictor fails to consider the very landscapes typically under the most pressure from change (e.g. urban fringe landscapes already of moderate value) is one of its most objectionable aspects.
Sensitivity to Change – the Indicator is problematic in so far as it cannot be assumed that the quantum of development within a given LCA can be equated with an equivalent quantum reduction in landscape value. It rests on the incorrect assumption that development has an intrinsically negative effect on landscape value. This is of course is not the case, as the impact of development on landscape value is a far more sophisticated question of specific design, location, etc. Certain ‘development’ might in fact enhance or reinforce landscape value, so there would probably be problematic issues of identifying which types of development could be deemed to be positive or negative. Equally a very small development may have a very large impact on landscape value not captured adequately by the indicator (if it were a tall building seen for example over a wide area). In addition, an AHLV Indicator will capture landscape change only in a very limited way: that is, only those landscapes which change from High(Q) to High value or vice versa and will ignore all those landscapes changing from Low to Moderate value or Moderate to High(Q), etc.
Easily Understood – as noted above, the concept of ‘areas of high landscape value’ is one that is capable of being easily understood at a superficial and intuitive level. However, the precise process by which a measure of value is reach is in fact a very complex one, far less well understood by the layman.
Measurable – the concept of AHLVs is inherently measurable (by increase or reduction in area), but only if one accepts the incorrect precept (noted above) that there is a direct correlation between the plan area of a development and a reduction in landscape value, which is not necessarily the case.
Have a Target Level or Guideline against which to Compare it – it is possible to establish this by reference to e.g. existing area of HLVs.
Capable of Showing Trends over Time – again, it is possible to fulfil this criterion by reference to e.g. an area measurement of AHLV, although as noted above, a correlation between the change in areas of AHLVs and a change in the value of the landscape cannot necessarily be made. In addition, the Indicator will fail to reflect trends prevalent in areas other than those of High Landscape Value.
Scientifically or Analytically Valid – whilst there is broad agreement as to the conceptual usefulness of the idea of ‘landscape value’ as a mode of discourse, there is far less agreement as to its analytical validity and even less as to its scientific underpinnings. This represents a major flaw in the idea of a landscape indicator based on AHLV.
Capable of being Updated at Regular Intervals – whilst it is possible that data on landscape value for the HKSAR could be updated at regular intervals, in practice, the process of repeating a comprehensive landscape assessment for the HKSAR on a regular basis is likely to be a fairly expensive and time-consuming exercise.
Measurable through Existing or Readily Available Data – the data provided under the Study is adequate for the purposes of running an Indicator based on AHLV.
Based on Data Adequately Documented and of Known and Acceptable Quality – as above.
Multi-Directionality – whilst an AHLV Indicator would be at least in theory multi-directional, in practice is very unlikely to be so. This is because it is exceptionally difficult to improve areas of Moderate or High(Q) value landscape to High value without dramatic reversals of the processes underpinning landscape change, which are usually beyond the scope of practicable Government control. These include reversing changing agricultural practice, destruction of landscape features (such as wetlands) the effects of permitted development rights in the NT, changing rural fringe land uses such as container storage, and declining agricultural practice. In practice therefore, an AHLV indicator is likely to measure only loss of landscape value. (However, given that almost all areas of high landscape value lie in Country Park, it is even then in practice highly unlikely that there will be any change to such areas, given the statutory protection from development afforded to them).
Specific Features Central to the Distinctive Character of the Type or Area – the Indicator does not identify features central to the character of the LCA or LCT.
8.4 RECOMMENDED LANDSCAPE INDICATOR
8.4.1 After consideration of the two potential Landscape Indicators discussed above, it was determined that the Indicator based on SLFs achieved the highest levels of compliance with CASET, as well as internationally accepted, criteria. In particular, the better ability of the SLF-based Indicator to sensitively reflect the effects of policy changes as well as the poor ability of the AHLV-based Indicator to reflect slight changes to Low and Moderate value landscapes, were seen as being crucial in reaching the recommendation.
8.4.2 A summary of the relative strengths and weaknesses of the two potential Indicators in provided in Table 8.3 below.
Table 8.3 Compliance of Possible Landscape Indicators Against SUSDEV21 / UK Countryside Commission Criteria
SUSDEV21 / UK Countryside Agency Criteria Conformity with Criteria SLF-based Indicator AHLV-based Indicator Simplicity * *** Robustness ** * Policy relevant (representative) *** * Sensitivity to change *** * Easily understood ** ** Measurable ** *** Have a target level or guideline against which to compare it *** *** Capable of showing trends over time *** * Scientifically or analytically valid ** * Capable of being updated at regular intervals ** * Measurable through existing or readily available data * * * * * * Based on data adequately documented and of known and acceptable quality * * * * * * Multi-directionality * * * * Specific features central to the distinctive character of the type or area * * * Metrics:
*** = Good Compliance
** = Moderate Compliance
* = Poor Compliance
