Local warming?


DEB News





Diversity at a glance


"In the News"

Book Review

Wild Corner

Recent Publications

Information for Contributors



Miscellany (pdf)

CITES in Santiago
Explorations of two underground water channels in Hong Kong!
Ting Kok: a conservation issue
The Virtual School of Biodiversity: an update
Aeromonas and Vibrio spp. detected in Mai Po Nature Reserve and Inner Deep Bay
Trawling experience and treasures in Hong Kong waters

Explorations of two underground water channels in Hong Kong!

by Rita Yam, Sze-man Cheung and Benny Chan

The environment and life inside underground water channels often gives people a mysterious impression due to the darkness, reduced temperature, difficulties of access and the stories of monsters inside. Ecological studies concerning underground water channels have tended to focus on the diversity of bats which are the major occupants on the ceiling of these channels (Ades, 1994, 1999). A bat surveying team encountered a water monitor (Varanus spp.: Class Reptilia, Oder Squamata) during their investigation of an underground water pipe (L.K. Lin pers. comm.). Inside the channels, however, there is a lotic environment and there are, so far, no formal records of what species are living inside. One month ago, a piece of news (Oriental Daily 20 November, 2002) reported that a man got lost in an underground water channel in his great journey of fishing in the darkness. This led us to wonder whether there are assemblages of aquatic organisms living in these water draining channels.

Fig. 1. A map of Hong Kong showing the locations of the two underground water pipes visited.

Out of our own interest, we visited two underground water channels to conduct a preliminary survey on the fauna living inside. On 25 November, we visited an underground water channel located in Tseung Kwan O, where the man got lost when catching shrimps and crabs. This channel drains stream water from Tseng Lan Shue and diverts the water out to the harbour in Junk Bay (Fig. 1; total length 5 km). A gradient of salinity along the pipe is therefore established. Try to guess what we first saw when reaching the pipe entrance–a dead dog! (which was possibly disposed by the villagers). The channel is rectangular in shape and about 3 meters high with smooth walls. There are lots of turnings and junctions making people get lost easily, thus we needed to attach light sticks on the wall to mark our tracks. After a 20-minute walk, nothing was observed except annoying tiny flies. At the moment we decided to leave, there was something suddenly moving in the water under the light of our torches. Oh! It was a big mitten crab swimming slowly. We picked it up immediately and confirmed it to be an Eriocheir japonica (carapace width = 9.8cm). Having this as a ‘hint’ that we might encounter more organisms when going further, we carried on. Unlike the previous section of the channel, we found patches of boulders and sand accumulated on the bottom. We thought that there should be more organisms associated with such substrates. Searching the boulder patches intensively, we discovered a number of unidentified Macrobrachium spp. (Fig. 2) which will be sent to the Raffles Museum of the National University of Singapore for confirmation of species identity, mitten crab Eriocheir and estuarine crab Varuna sp. The crab Varuna is reported to have a distinctive seaward migration in water channels for breeding (Lee and Leung, 1999). In general, there were high abundances of crabs and shrimps in this completely dark environment. The salinity of the water where we got the crustaceans was 1 o/oo suggesting that the aquatic lives were experiencing a gradual increase of salinity from the entrance of the channel, at which the salinity was 0.02 o/oo. Going further into the channel, we also collected two more mangrove crabs Scylla paramamosain. We continued to walk about 3 km down into the channel. Since only several cockroaches and moths greeted us, we decided to go back.

Two weeks later (6 December 2002), we visited another underground water channel at Nam Chung. The channel collects waters from hill streams going into Plover Clove Reservoir. It is a straight circular pipe about 6 km long. Having walked down into the first 100 meters, we found a group of bats and a long-legged centipede (Order Chilopoda, Family Scutigeridae) on the wall. The bottom of the channel was quite smooth, neither boulder patches nor leaf packs were present. As we had expected, there were not many aquatic organisms observed in this area. Passing that "desert", we encountered a number of Eriocheir japonica and the unidentified Macrobrachium spp. Besides shrimps and crabs, we also found gobies (Oder Perciformes, Family Gobiidae). Strange enough, some dead "headless" fishes were lying on the stream bottom. Since fish may probably be one of the major food sources for the bats living there, we guessed those fishes might have fallen onto the ground when they were being eaten by the bats hanging on the top of the water channel. According to Ades (1999), fish eating bats, Myotis ricketti, are found in Hong Kong and 59% of their gut contents by volume consist of fish (Cyprinidae) scales. After confirmation of the bat species, we were able to make the preliminary conclusions about our fish observation. On going further down into the channel, the wall became very rough with stalactites growing down from the ceiling, suggesting that this section might be constructed from a natural cave. Here, another group of bats was found and the water had become much deeper (0.5m) than that at the beginning area (0.3m). Also, we found two more species of fish (Parazacco spilurus and Osteochilus vittatus, fork length = 21.6 cm). We continued to explore this water channel for about 1.5 hours more and then returned.

Fig. 2. Unidentified Macrobrachium spp. collected at the underground water pipe at Tseng Kwan O.

We believe that most of the animals found in the water channels originate from feeder streams, interconnected reservoirs and, in the case of Tseung Kwan O site, from coastal shore near the seaward channel mouth. Compared to the feeder stream, at least for some taxa such as fish, the abundance was much reduced in the channel. For example, a school of Parazacco spilurus was observed in a pool directly outside the entrance of Nam Chung channel, but only a single individual of this species was recorded inside. The food chains in the channel also appeared to be incomplete. Although no systematic observation data were obtained, the channel served as a bat roosting rather than feeding site. We suspected that the bats have to disperse to nearby sites for feeding. In addition, water marks up to about two metres were seen in both channels and this suggested that water may rise to that level during the wet season. This means most aquatic life might be washed away during peak flow and animals from outside have to "colonize" the channels after the peak wet season. As a result, we believe underground water channels cannot be considered as a self-contained ecosystem, but rather as a special semi-natural microhabitat interconnected to other water bodies, depending on the imports of both faunal colonizers and nutrients.

One thing worth pointing out is the nature of these water channels in conducting water between different aquatic habitats. Gene flow of aquatic species such as Macrobranchium spp. among freshwater habitats may be promoted by the linkage of these pipes. Channels connecting hill streams to the sea may provide pathways for mass migration of anadromous species such as Eriocheir japonica and Macrobranchium nipponensis. However the spread of introduced species, especially large mobile species such as predatory exotic fish to nearby natural water bodies may threaten the survival of native resident species.

For faunal inventory purposes, future studies should cover water channels with different engineering designs as we feel that preference for "natural cave like" channels by different taxa of animals may exist. As both water channels were investigated during the daytime, further explorations should be extended to include night visits to recover nocturnal species.

Whether the gene flow of aquatic species among fragmented freshwater habitats and mass spawning migration of anadromous species through these underground water channels really exist is a potential research area for further studies.


Ades, G.W.J. (1994). A comparative ecological study of insectivores bats (Hipposideridae, Vespertilionidae and Rhinolophidae) in Hong Kong, with special reference to dietary seasonality. Ph.D. Thesis. The University of Hong Kong.

Ades, G.W.J. (1999). The species composition, distribution and population size of Hong Kong Bats. Memoirs of the Hong Kong Natural History Society. 22: 183-209.

Lee, S.Y. & Leung, V. (1999). The Brachyran fauna of the Mai Po Marshes nature reserve and Deep Bay, Hong Kong. In: S.Y. Lee (ed). The Mangrove Ecosystem of Deep Bay and the Mau Po Marshes. Proceedings of the international workshop on the mangrove ecosystem of Deep Bay and Mai Po




For more information, contact

Copyright © 2000