2.1        INTRODUCTION

2.1.1     Para 6.6(1) of the brief requires that the Consultants carry out "a review of the initial landscape categories, classification...in the light of public comments/views received..." 

2.1.2     This Chapter identifies key comments received on the Initial Landscape Classification (presented in TR1) during the First Stage Public Consultation and describes the consultant's proposed response to these comments and how they were incorporated (or otherwise) into the Revised Landscape Classification. (Please note that comments received are discussed more fully in the First Stage Public Consultation Report).

2.1.3     It also describes other changes which have been made to the Initial Landscape Classification, as a result of comments received on TR1 and also as a result of the development of the Preliminary Landscape Character Map.

2.1.4     Finally, a full Revised Landscape Classification is set out, for use in development of the Preliminary Landscape Character Map.

2.2        CHANGES TO LCTS ARISING AS RESULT OF THE FIRST STAGE PUBLIC CONSULTATION                                  

2.2.1     During the First Stage Public Consultation, a number of consultees raised issues relevant to the Initial Landscape Classification. The key comments made in this regard are summarised below and a commentary provided on them.

Comments on Theoretical Approach to Classification

The Landscape Classification

2.2.2     Grouping and classifying landscape into LCAs and LCTs is central to the Study approach. A consultee commented that since every place was unique, the need and rationale for such landscape classification should be provided. 

Response/Action

2.2.3     Although classifying landscape into LCAs/LCTs is not the only approach to landscape assessment and mapping, the current landscape classification recognises that concurrences of geology, landform, topography, vegetation and land use give rise to landscape types which have a great deal in common. This approach establishes a practical framework for assessing a diversity of landscapes. It does not deny the unique characteristics of each landscape, as every landscape is assessed individually within this framework.

Landscape Character Types

2.2.4     During the consultation, a query was raised as to whether the LCTs identified were specific to Hong Kong. Another consultee suggested that LCT titles should be self-explanatory as far as possible, and that the names of LCTs in Chinese and English might have subtly different meanings. It was also pointed out that it is important to distinguish between 'land use' and 'landscape', as these are not the same thing.

Response/Action

2.2.5     Because of the unique landscape characteristics of Hong Kong (in particular in the urban area) all LCTs have been developed specifically for this Study and for the Hong Kong landscape. For ease of reference, LCT titles need to be relatively concise and at the same time need to describe the essence of an LCT, which is often only subtly different from another LCT. We shall keep the titles under review so as to make them more intelligible. In addition, care will be taken in the translations of LCTs in different languages to avoid confusion and misunderstanding. The landscape classification has been produced on the basis that land use is only one of a number of factors which give a landscape its character.

The Boundaries between Different Landscapes

2.2.6     The transition from one landscape area to another is usually gradual. During the consultation there was an enquiry on how boundaries between LCAs would be determined. The enquirer suggested that rather than using solid lines, 'gradients' or blurred lines should be used where appropriate to reflect the gradual change.

Response/Action

2.2.7     It is well recognised that landscape is a continuum and that character does not in general change abruptly. However, standard practice in the case studies examined is to use a solid line to define the boundaries between LCAs. Usually, for the sake of certainty, this boundary follows a definable feature on the ground, such as a road, woodland, coastal feature, fence or drainage channel etc. 

The Scale of Landscape Character Areas

2.2.8     The Study proposes 5 hectares as the size of the basic unit of LCAs. There were however some general concerns on the size of LCAs.. One consultee enquired if the less homogeneous urban area would be divided into smaller sub-units. Another considered there might be a need to break down landscape features into even smaller components so as to identify individual characteristics and to generate pattern. Other consultees agreed that the 5-hectare size might be too big to capture sufficient detail. Conversely, another consultee felt that if LCAs were too small, then wider patterns and connections in the landscape would be lost. It was also mentioned that the choice of the size would affect the applicability and effectiveness of the findings for use in landscape impact assessment (as part of the EIA process).

Response/Action

2.2.9     5 hectares is perhaps the smallest size which can be reasonably classified as a "landscape" and used as a yardstick for delineating LCAs. If one goes to smaller units than this, the effect of land use on classifications tends to become unduly predominant. However, it is agreed that larger LCAs would be appropriate in rural areas where landscape diversity is less, whilst smaller LCAs (in some rare cases slightly smaller than 5ha) would be more appropriate in urban areas where landscape diversity is greater. In addition, the Study will include an overview of landscape at the sub-regional scale, to ensure the wider patterns and connections in the landscape are adequately captured. With regard to the use of results in EIAs, the landscape/visual component of an EIA normally adopts a scale of classification appropriate to the particular EIA Study. However, there is no reason why the LCAs/LCTs identified in this Study could not be employed in certain EIAs.

Comment - The Need for a Landscape Classification System

2.2.10   One consultee questioned the need for a classification system at all, suggesting that every place can be regarded as unique.

Response/Action

2.2.11    As set out in the Inception Report and Technical Report 1 of the Study, classification is not the only approach to landscape assessment and mapping. However, it is central to the character-based approach, which is one of the most widely accepted, if not the most widely-accepted approach.

Comments on Specific LCTs in the Initial Landscape Classification

Comment - The Need for Additional LCTs

2.2.12    A number of consultees suggested that the preliminary list of LCTs provided in the Digest might usefully be reviewed in order to better explain or reflect landscape variation in:

• the urban fringe; 

• undeveloped countryside and 

• LCTs reflecting historic features areas. 

Response/Action

2.2.13   The Consultants accept that it will be necessary to review the LCTs in various areas to ensure that they accurately reflect landscape character. 

2.2.14    In the natural countryside, a number of additional LCTs have been created to reflect the diversity of natural landscape types. Changes to LCTs include splitting the former 'Upland Peak' District Level LCT into 'Peaks', 'Undulating Uplands and Hillsides' and 'Miscellaneous Uplands' LCTs at District Level. This has resulted in the following new Local LCTs:

• Scrub-covered peaks and ridges;

• Grassy undulating uplands and hillsides;

• Scrub-covered undulating uplands and hillsides;

• Wooded undulating uplands and hillsides.

2.2.15    The 'Upland Valley' Local LCT has been split into:

• 'Scrub-covered upland valley' and 

• 'Wooded upland valley'.

2.2.16    Additional Local LCTs have been created within the 'Lowland Valley' District LCT. These include:

• Lowland valley farmland;

• Lowland valley farmland with other uses;

• Scrub-covered lowland valley and

• Wooded lowland valley.

2.2.17    With regard to LCTs in the urban fringe and in historic areas, as no obvious refinements are immediately evident, it is suggested that this refinement be best carried out during Field Survey. At this stage, the ability of the Initial Classification to reflect landscape character can be tested and if necessary, refined by the creation of new LCTs.

Comment - Suggested Additional LCTs

2.2.18    A number of consultees suggested additional Landscape Character Types, in order to define landscapes which they felt were not adequately reflected in the Initial Landscape Classification. The suggested additional LCTs were:

• 'Specialised Street market';

• 'Pedestrian precinct';

• 'Public square';

• 'Small house policy area' in the New Territories;

• 'Fresh market'

• 'Natural streams and waterfalls' and

• Open space / view corridor.

Response/Action

2.2.19    'Specialised street markets' are certainly features within the urban landscape, but because of their relatively limited extent or intermittent nature, it is difficult to call them 'landscapes' in the conventional sense of the word. It is suggested therefore that these be recorded as landscape features, rather than landscape types.

2.2.20    'Pedestrian precincts' like street markets are significant features within the urban realm. However, because of their relatively limited extent in Hong Kong, it is difficult to call them landscapes in their own right. Again, it is suggested that these be recorded as landscape features, rather than landscape types.

2.2.21    The comments above also apply to the suggested 'Public square' LCT. Given the limited number and small size of such features in Hong Kong, they are probably best recorded as features in the landscape rather than LCTs in their own right.

2.2.22    'Small house policy area' is a type of landscape which it was envisaged that would be captured in the 'Expanded Village' LCT presented in TR1. As it is usually difficult or impossible to coherently separate the traditional housing in villages from village housing policy areas, it is suggested that the 'Expanded Village' LCT be used to cover what is usually a fairly mixed village landscape.

2.2.23    'Fresh markets' are within the natural landscape, streams and waterfalls are really not sufficiently extensive to be classed as landscapes in their own right, but are best described as landscape features and recorded as such. With regard to the proposed 'open space / view corridor' LCT, an 'urban park' LCT is already proposed, where these are a scale to comprise a complete LCA. As view corridors are defined by reference to only one characteristic (i.e. visual connectivity), they are not really LCTs in the sense that this Study uses the term. In particular, a view corridor does not possess a consistent set of landscape attributes, but may include landscapes of many types. It is therefore, probably not appropriate for inclusion as an LCT. 

2.3        CHANGES TO LCTS ARISING DURING PREPARATION OF THE PRELIMINARY LANDSCAPE CHARACTER MAP

2.3.1     During the course of preparing the PLCM, it became evident that a number of LCTs additional to those new LCTs identified above, were required in order adequately reflect the full range of landscape character in Hong Kong. These additional LCTs include the following landscapes.

Salt pans

2.3.2     These are open, large-scale expansive and low lying landscapes located next to coastal areas. They comprise large regularly shaped pools where salt would be extracted from sea-water as an industrial practice. With the decline of this industrial salt extraction, these landscapes are now generally derelict and have a rather degraded character. e.g. Tai O.

Disused coastal plain farmland

2.3.3     These are former agricultural landscapes found on coastal plains, below 40mPD. Knolls and blocks of woodland are scattered across the coastal plain, but villages, farms and former agricultural land has often been abandoned with fields having become overgrown. Other uses have not yet begun to encroach, resulting in a somewhat neglected character, e,g, Sai Sha area.

Transportation corridor

2.3.4     These are linear landscapes associated with major highways which attain a certain critical width, often through being associated with other transportation routes, such as railways. They comprise roads, railways, elevated structures, noise barriers signage, gantries, clover leafs, traffic islands, toll plazas, intermediate strips of landscape as well as associated miscellaneous roadside land uses, resulting in diverse, incoherent, linear landscapes, e.g. Tolo Highway, West Kowloon Expressway.

Rail depotyard

2.3.5     These are landscapes comprising extensive areas of railway tracks, sidings, sheds, rolling stock, gantries and rail infrastructure. They include sidings and marshalling yards and form landscapes which are expansive and incoherent. e.g. Fanling marshalling yards.

Waterfront village

2.3.6     These landscapes comprise former fishing villages which have been built right on the coast, and which retain many original structures. They form a coherent, compact village landscape characterized by its small-scale, enclosure, but also by its proximity to the sea, e.g. Yung Shue Wan.

Fish farm

2.3.7     These are marine landscapes lying just off the coast comprising open water, where regular patterns of booms and nets have been set out for the farming of fish. Open and expansive, these landscapes have a certain formal quality as a result of the regular man-made features within them, e.g. waters off east Lamma coast.

2.4        INTERIM LIST OF LANDSCAPE CHARACTER TYPES

2.4.1     In the light of the above, a comprehensive review of the Initial Landscape Character Types has been carried out. This has resulted in the Interim List of Landscape Character Types. The Interim List of LCTs is set out in Table 2.1 and in Appendix 4. This list forms the basis for the production of the PLCM.

2.4.2     However, it is important to recognise that the Interim List of LCTs cannot necessarily be regarded as complete, and that it may well need to be modified during field survey (when the urban fringe and historic LCTs in particular will be kept under review). In any case, a Final List of LCTs will be presented in TR4, reflecting any changes made to the list during Field Survey.