As indicated in TR2, in the final stages of the SUSDEV 21 Habitat Mapping Baseline Survey, ERM will use available existing data, including species lists derived from the HKU Biodiversity Survey, in addition to the field survey data collected by the SUSDEV 21 Study Team, to recommend an ecological value ranking for each habitat. Species data are accounted for in the ranking system agreed in TR1; areas of high biodiversity and areas which contain documented critical habitat for rare species will be given a "high" ranking. As analysis of the Biodiversity Survey dataset for high biodiversity area (ie "biodiversity hot-spots") by the HKU Study Team is still on-going and data on their location are unlikely to be made available to the SUSDEV 21 Study Team within the programme for the SUSDEV 21 study. As a fall-back alternative, data on "rare" species provided by the database were therefore considered for usage in the final ranking.

The HKU Biodiversity Survey GIS database (CD-ROM, version 1) contains a number of layers of information showing the locations of many species of plants and animals throughout Hong Kong. While the CD-ROM database has provided a wealth of interesting and important information on biodiversity, the authors of this database have stated that the database requires further validation by them before using it for any interpretative work (Dudgeon and Corlett 1999). Although the database is a useful source of information on species, there are several gaps and inconsistencies within the dataset which has impeded usage of the data for the purposes of upgrading the ecological ranking. For example, column names are not consistent between files, species information is incomplete and survey dates are not provided for many records. In addition, many of the data appear to have been generalised or recorded to a level of accuracy ranging from 50 m to 1,000 m, with some of the terrestrial species records sighted over the ocean. This generalisation of data has impeded usage of the dataset as the level of accuracy is outside the mapping range used for this study, ie 20 m (the accuracy level required by the Study Brief).

However, records on the Rare Plants, Sedges and Fung Shui Woods files were found to be comparatively more consistent and thus were considered as a data source for final ecological value ranking. Although there are a number (863 of the 954 records) of the "Rare Plants" records which lack information on date of sighting, all records were used to locate "rare" plant species.

The HKU Biodiversity CD-ROM contains a list of 348 rare plants and their respective locations in Hong Kong. A recent joint publication from HKU, Kadoorie Farm & Botanic Garden and South China Institute of Botany (Corlett et al 2000) has provided up-to-date information on the existing status of vascular plants which have been reliably recorded from Hong Kong. Among the 2130 listed plant species, 719 of them were regarded as "rare" or "very rare". The rare plant species list from the HKU Biodiversity Survey was combined with the list from Corlett et al (2000) (see Annex H). Among the 719 "rare" and "very rare" species identified by Corlett et al (2000), 200 of them are also listed in the HKU Biodiversity Survey CD-ROM Rare Plant file, 41 are found in the Fung Shui Woods file and 14 are in the Sedges file. Sighting records in the Sedges and Fung Shui Woods files were extracted and added to the locations of "rare" plants in the Rare Plants file.

As one of the criteria for high ecological value habitat is whether an area provides documented critical habitat for rare species (see Section 3 of TR1), the intention was to use the HKU Biodiversity Survey CD-ROM "rare" plant data to adjust (upgrade) ecological value of those habitat polygons which have one or more of the identified "rare" plant species located within it. However, use of the HKU CD-ROM data to perform the upgrade produced erroneous results, as the co-ordinate locations of the rare plant sightings are too general for this purpose. Many of the positions of sightings are recorded in an incompatible habitat type (see Figure 7.2a).

Use of the HKU CD-ROM data increases the area of High Ecological Value habitat by 16,821.9 ha, with 8,616.3 ha under the "Other" category. This is considered very unlikely to reflect the situation, and is believed to be a result of the aspects of the HKU data discussed above. Table 7.2a shows the area of each habitat type that was upgraded when the HKU CD-ROM data were used.

Thus, as the data location accuracies are so greatly incompatible, the HKU CD-ROM data has been excluded as a source for upgrading the ecological ranking of habitat. The rare plant information is however a useful reference and will be included with the map as a separate GIS layer.