Two representative (i.e., replicate) sites in each of the four sections (A, B, C and D) were chosen which allowed spatial comparisons within and between sections (i.e., 2 × 4 = 8 sites in total). These sampling sites were:
Section A - Ting Kok West and Ting Kok
Section B - Island House and Tolo Pond
Section C - Sai Keng and Yung Shue O
Section D - Fung Wong Wat and Lai Chi Chong
To encompass seasonal variation, surveys were be conducted in both the dry (December to March) and the wet (June to September) seasons, as seasonal variation in these fauna is expected to be important. Two temporal replicates (i.e., twice per season) were conducted at each site. The date of each temporal replicate was randomly chosen within the season. We follow the established protocols for sampling mangrove biota (Skov et al., 2002; Cannicci et al., 2009). Apart from studying the organisms inhabiting mangrove plants (i.e., arboreal organisms), we also studied the organisms inhibiting the substratum of the mangrove area (i.e., epifauna) in order to capture as many species inhabiting the mangrove as possible.
A 30 m transect was laid along the shore and three 5 m × 5 m quadrats were deployed on the landward side at 10 m intervals along the transect. To accommodate spatial variation within each site, replicate transects were established at different shore heights (i.e., high and low shore) where possible, depending on the extent and complexity of the mangrove stand (e.g., having mixture of layers of different mangrove species, or having a large mangrove area).
Three mangrove trees in each 5 m × 5 m quadrat were randomly chosen and all the animals present on the leaves, branches, trunk and roots were identified and counted. An estimate of the density of organisms was made by multiplying the mean number of animals on the sampled trees by the number of trees in the quadrat and then dividing this value by the quadrat area (after Walthew, 1995). Algae present on the trees were only scored as present or absent and were collected and identified later in the laboratory.
In each 5 m × 5 m quadrat, three 0.5 m × 0.5 m quadrats were randomly deployed on the surface of the mud to sample the epifauna. The abundance of mobile animals was counted, and the coverage of sessile organisms was estimated using a 100-point double strung 0.5 m × 0.5 m quadrat.
Visual survey of crabs
During the surveys in the wet season, we added visual counts of epifaunal crabs by using binoculars. There were two experienced persons standing outside the 5 m x 5 m quadrat, who observed any epifaunal crabs inside the quadrat for 10 mins. A 5-min resting period was added before the counting so as to allow the fauna within the quadrat to recover from being disturbed. To avoid double-counting, one person was responsible for half of the 5 m x 5 m quadrat only.
We measured the height of mangrove seedlings, and counted the number of leaves on each of them during the wet season, as a proxy for measuring their growth. In each 5 m x 5 m quadrat, we tagged three seedlings (< 50 cm of height), measured their height and counted the number of leaves.
All adult plants (height > 0.5 m) were counted within the 5 m x 5 m quadrat. At 0 m, 10 m, 20 m and 30 m along each transect, a centre point was set at a position of 2.5 m away from the quadrat. The centre point was used to divide the adjacent mangrove area intro four portions. In each portion, the tree closest to this centre point was measured for its height, stem basal diameter (calculated from the circumference at the base of each stem), and distance between the tree and the centre point.