INTRODUCTION

The main report provides details of the species found within the Study Area of the Sheung Shui to Lok Ma Chau Spurline during the period April 2000 to May 2001 (survey dates shown in Appendix A1). Avifauna, insects and herpetofauna were specifically studied in areas along the alignment noted for their ecological value. In addition, mammals and mammal signs were recorded when these were noted during the course of surveys of other taxa groups. Results are presented graphically and with narrative to explain the implications of findings.

In addition, Section 4 of this Report covers a General Wildlife Survey which was undertaken along the section of the Spur Line route between Lok Ma Chau and Chau Tau during October 2001. Though of less ecological importance than either the Lok Ma Chau / San Tin and the Long Valley areas, it was recognised that this area might support habitats and species of ecological interest which might not have been detected by other surveys used in the preparation of the previous Spur Line Environmental Impact Assessment (EIA) Report (BBV 2000). This additional survey was, therefore, undertaken to ensure that the ecological assessment in the current (second) EIA Report was comprehensive.

1. AVIFAUNA

1.1 Introduction

1.1.1 At least 227 species were recorded within the Study Area during the period 1988 to 1998 (ERM 1999). A total of 107 species of birds was recorded during the baseline surveys. These are listed in Appendix A2.

1.1.2 ERM (1999) identified thirty Species of Conservation Importance as being regularly recorded within the Study Area. This report gives details of those species based upon the findings of Baseline Surveys at Long Valley and Lok Ma Chau, radio-telemetry results of work on Greater Painted-snipe Rostratula benghalensis, and a survey of two globally threatened eagles, Imperial Eagle Aquila heliaca and Greater Spotted Eagle A. clangula. The report covers the period April 2000 to May 2001. Baseline data on wetland dependent species at Lok Ma Cha and Long Valley are also reported. The status of Globally Threatened Species has been updated based on BirdLife International (2000, 2001).

1.2 Methodology

(a) Bird distribution and abundance at Lok Ma Chau, San Tin and Long Valley

• Locations

Monitoring was carried out in all habitats, other than village areas, within 500 m of the Spur Line in Long Valley (the area north of the Castle Peak Road, and between the drainage channels for the Rivers Beas and Sutlej, see Figure 1); the wetland mitigation area at the Lok Ma Chau terminus area; and the fish ponds at San Tin (see Figure 2).

• Timing / frequency

Four visits per month were made to Lok Ma Chau and Long Valley from late May 2000 to late May 2001. Four visits per month were made to the San Tin Fishponds from late November 2000 to mid May 2001 (exact dates are shown in Appendix A1). Monitoring commenced within one hour of dawn to coincide with peak bird activity and to standardise results.

• Survey Area

For Long Valley and Lok Ma Chau, surveys were undertaken within 500 m of the Spur Line. This distance was selected as it is the maximum referred to in the EIA at which it is predicted that there will be reduced densities of birds as a result of disturbance. At San Tin, all fishponds within 1000 m of the Spur Line were surveyed.

• Procedure

For the Long Valley, Lok Ma Chau and San Tin areas a base map was prepared with each agricultural plot or fishpond numbered (Figure 1). During each survey visit, the surveyor visited each pond or agricultural plot in the study area and identified to species level and counted all birds present.

Each fishpond or agricultural plot was surveyed from one point, that most accessible to the transect route. If required the surveyor then adjusted his position (if part of the fishpond was out of sight, or if closer views of a bird were required to confirm identification). When possible, the pond or agricultural plot was surveyed from a single viewpoint to reduce disturbance and lessen the possibility of double counting. If it was suspected that certain birds had already been counted on other ponds or agricultural plots, these were ignored. If areas contained large numbers of birds (e.g. foraging egrets on fishponds, or waders on flooded fields) these were surveyed at a distance to avoid flushing birds, and to further reduce the possibility of double counting. Where necessary, fields or fishponds that had been found to hold large numbers of birds earlier in the visit would be revisited if there was a suspicion that the birds had moved to elsewhere in the study area during the course of the survey. Flying birds were not recorded unless they were clearly foraging and associated with the habitat. The one exception to this was Black-faced Spoonbill Platalea minor, for which all birds were recorded, although efforts to avoid double counting were maintained.

The broad physical characteristics of each agricultural plot or fishpond were recorded, including information on water levels, whether it was actively managed or not, and, for fishponds, details of drainage activity. Agricultural plots were identified as either ‘active’ or ‘inactive’. This was based on farming activity (presence of farmer working in the plot, presence of farming equipment, and dominant crop/vegetation type). Whether the plot was wet or dry was also recorded. The condition of each plot was recorded once every two weeks. Drainage activity of fishponds was recorded (not drained/drained/being drained), and details on fish harvesting where apparent were also noted. Partially drained fishponds that had not been harvested were identified on the basis of the obvious numbers of larger commercial fish. For partially to fully drained ponds the extent of the exposed bottom was recorded to the nearest 25% of the total area of the fish pond, and also whether the exposed bottom was wet or dry.

1.3 Species Specific Bird Surveys

Radio Telemetry of Greater Painted-snipe

1.3.1 Greater Painted-snipe a key species at Long Valley, and is the only Species of Conservation Importance that breeds there; it breeds at only two sites in Hong Kong (Leader 1999). In Hong Kong, radio telemetry (radio tracking) of this species has been successfully been used in the Kam Tin environs to identify differences in diurnal roosting sites and nocturnal foraging areas. However, prior to this present study, nothing was known of the nocturnal foraging areas of this species in Long Valley. Accordingly, as a component of the baseline monitoring programme, radio telemetry was employed to identify the preferred foraging areas of this species, and provide data on movements and distribution at Long Valley. This was then related to broad characteristics of land use, vegetation typeand structure and hydrology of the areas utilised.

• Location

The radio telemetry survey work of the Great Painted-snipe focused on the area of Long Valley covered by general bird studies (give number of figure here-map showing numbered fields at Long Valley). However, due to the mobility of this species, some observations were conducted in a broader area, notably north of the River Beas (Figure 71).

• Frequency

Radio telemetry was undertaken on a seasonal basis. The seasons were defined as follows:

Spring: (April-May)

Summer: (June-August)

Autumn: (September-October)

Winter: (December-January)

Prior to the commencement of each season's monitoring exercise, daytime visits were made to prime habitat frequented by this species in the Long Valley area. Prime habitats were identified as any area of semi natural marsh that was neither devoid of surface water, nor lacked areas of water that were less than 15 cm in depth. Up to six 12 m long two-panel nets were used, depending on the size of the trapping area. These were erected on bamboo poles that were 4 m long, with the bottom of the net placed about 0.5 – 1 m above the level of the surrounding vegetation, and the top about 2 – 3 m above the level of the surrounding vegetation. Nets were placed as close to the trapping area as was possible without disturbing the birds. Following this, the area was slowly walked by two to five persons with the aim of flushing the birds into the nets. Each bird trapped was aged, primarily on the colour of the tips of the secondaries (white on adults, buff on juveniles), sexed, weighed, and information on biometrics and moult were also collected.

In addition to the radio-transmitters, each trapped bird was fitted with permanent individually numbered metal rings in accordance with the Hong Kong ringing scheme. A 2.5g gauze-mounted TW-4 single button celled tag (radio-transmitter) (Biotrack Ltd., Dorset, UK) was attached to the upper mantle with epoxy glue (Hirons and Johnson 1987, Kenward 2001). The birds were located using an M57 receiver (Mariner Radar Ltd, Suffolk, UK) and a hand-held flexible 3-element yagi antenna (Lintec, Dorset, UK), and could be detected up to a maximum (line of sight) of 2.5 km.

• Methodology

Telemetry was undertaken over two ten-day periods during each season. During each ten-day period all tagged individuals were searched for twice during each 24-hour period, once during daytime and once during nighttime. The area of search initially focused on the area of trapping and, if individuals were not located, this was slowly expanded. If necessary, higher vantage points were sought (such as the hills above Ho Sheung Heung) in order to increase reception range. Each bird was located through triangulation (Heezen and Tester 1967, Kenward 2001), and where possibly to the nearest field but always taking into account the requirement to ensure that birds were not disturbed or flushed by the observer. This was particularly important at night when individuals were often foraging in more exposed habitats. Where possible, the distance between the receiver and tag was kept low (generally <50 m), in order to reduce triangulation error (White and Garrott 1986). If a bird was accidentally flushed, this information was recorded. Once a bird was located, a sketch map showing the location of the bird was drawn, or sufficient text detailing the location was recorded. Exact co-ordinates of the observation point were then determined using a handheld GPS. This was particularly important for nighttime radio-tracking when it could be difficult to determine the exact location of a field and/or it’s physical characteristics. In such cases, the GPS information was used to confirm these data during daytime.

Distribution and Habitat Requirements of Imperial and Greater Spotted Eagles

1.3.2 Imperial and Greater Spotted Eagles are globally threatened species that are regular but scarce winter visitors to Hong Kong. Less than 10 Imperial Eagles and 5 Greater Spotted Eagles occur each winter, mainly in the Deep Bay area (Carey et al. 2001). The survey was undertaken to establish the distribution and habitat requirements of the two species to inform habitat enhancement measures for them as part of the design of the wetland compensation areas at Lok Ma Chau.

• Location

The survey area included those areas impacted by the Spur Line in the Deep Bay area, but given the extremely mobile nature of large eagles, also extended well beyond that area.

The Inner Deep Area between Tsim Bei Tsui and Ma Tso Lung was surveyed during weekly coordinated transects in the period from mid-November 2000 to the end of March 2001 (Figure 3). The survey area comprises most of the range of Greater Spotted and Imperial Eagles in Hong Kong, and all of the brackish wetland range (some birds also rarely use the Kam Tin area). This large study area was divided into six sub-units. Each sub-unit was surveyed weekly, with three alternate sub-units being covered by three observers during one day (i.e. units 1, 3 and 5 were covered simultaneously on one date, and 2, 4 and 6 on another date). Units, selected on the basis of size, suitable habitat and physical barriers to observers are detailed in Table 1.

Table 1

Survey Units Used During the Greater Spotted and

Imperial Eagle Surveys

Unit No.	Name
1	Tsim Bei Tsui – Fung Lok Wai
2	Nam Sang Wai
3	Lut Chau
4	Mai Po
5	San Tin – Lok Ma Chau
6	Ma Tso Lung

• Timing and Frequency

Greater Spotted and Imperial Eagles only frequent Hong Kong during the winter months. The first birds typically arrive during the last week of October and the latest individuals depart during early April (Carey et al. 2001). Accordingly, the study period for these species covered the main period of mid-November 2000 to late-March 2001. Dates are given in Appendix A1.

• Methodology

Co-coordinated transects

Observers followed a standard transect through each area (Figure 2). Each transect was covered by bicycle (except Lut Chau, which was covered on foot) and each transect was repeated three times on each survey date. Each transect occupied one to two hours (typically one and a half hours) spaced through a five and a half hour survey period. The observers remained on the transect, but actively searched for flying or perching eagles along it. When an eagle was observed the location of the eagle was estimated as accurately as possible by using the base maps.

The observer then kept the eagle in view for 10 minutes or until it moved out of sight (whichever was the shorter period), during which time the behaviour of the eagle was recorded. Where appropriate, activity was recorded in more than one category. The following activity categories were used:

· Attacking (with reference to the species being attacked)

· Displaying (breeding or territorial)

· Feeding (with reference to the food items where possible)

· Flying

· Soaring

· Hunting

· Mobbing (with reference to the species involved)

· Perching

· Preening

· Roosting

Whenever possible, eagles flying overhead were photographed with an auto-focus camera using a minimum lens size of 300mm. From this it was possible to identify individual birds through a combination of moult, feather damage/loss, and plumage characteristics. This was useful in determining the population size and mobility of individuals within the study area, and in identifying several migrant eagles.

The primary aim of the fieldwork was to determine which areas are used by the two eagle species, with particular reference to those areas to be impacted by the Lok Ma Chau Spur Line. The data were analysed to determine different densities of occurrence throughout the study area.

The density data were also compared to habitat type to determine any relationship.

Continue

Kowloon-Canton Railway Corporation Capital Projects Planning Department No. 9 Lok King Street Fo Tan, Shatin New Territories	Asia Ecological Consultants Limited 127 Commercial Centre Palm Springs Yuen Long New Territories
	Binnie Black & Veatch Hong Kong Limited 11/F, New Town Tower Pak Hok Ting Street Shatin New Territories

	January 2002

Environmental Support Services

1. AVIFAUNA

1.1 Introduction