Total consumption of freshwater fish in Hong Kong increased from 11 419 tons in 1961 to 36 234 tons in 1973. The future demand is estimated to be about 43 000 tons in 1976 and 84 500 tons in 1986 (Nichols, 1974). While Hong Kong cannot expect to become completely self-sufficient, a major policy objective is to maximize local production and to minimize problems caused by an overdependence on external food supplies. As in other countries of the region, marine fisheries are the predominant source of fish. With the anticipated levelling off of the production from this source, however, freshwater fish production is gaining in importance. Land shortage poses a severe constraint to further expansion of freshwater aquaculture. Efficiency of fish culture, increasing productivity per unit area and the reduction of losses to the minimum, are the ways to deal with growing demand on the limited area. Proper water quality management is a fundamental prerequisite to prevent fish mortalities in farm ponds, which is becoming a serious problem and resulting in lower average annual yields.
There are about 1 400 hectares of freshwater ponds used for polyculture of commercially important Chinese cyprinids and grey mullet. These are concentrated in the alluvial plain of the northwestern sector of the New Territories. The culture system essentially conforms with traditional Chinese practices in fertilized earthen ponds ranging in size from about 0.3 to 1 ha each. Due to a limited supply of freshwater, the ponds rely on precipitation as the primary source of water. Ponds are flooded during the four-month rainy season and then are left to evaporate causing a gradual increase in salinity, which remains within the freshwater range.
The typical stock composition of cultivated species is as follows (Chan, personal communication, 1975).
Mugil cephalus (grey mullet)
Ctenopharynogodon idella (grass carp)
Aristichthys nobilis (bighead carp)
Hypophthalmichthys molitrix (silver carp)
Cyprinus carpio (common carp)
Cirrhina molitorella (mud carp)
Carassius auratus (edible goldfish)
Of these, the grey mullet is the most important species normally representin between 40 to 50 per cent of the total annual yield from inland fishponds. Average annual yield has been estimated to be 2.5 tons/ha. Maximum field is about 5.5 tons/ha which shows that there is ample room for further improvements in production.
The ponds are fertilized with rice bran, peanut cake and manure from domestic animals (cattle, poultry and ducks).
The Department of Agriculture and Fisheries operates a field research station in the fish farming area (Inland Fish Culture Sub-station at Au Tau, near the town of Yueng Long). Twenty ponds of different sizes totalling 3.3 ha are being used for research on methods of increasing fish production.
About 30–40 per cent of the fish in mullet-carp fish ponds is believed to be lost through mortalities caused by oxygen deficiency during the months of July-August. Fishkills are not total, however, and the larger part of the fish stock usually survives. A team from the Fisheries Research Division has been studying this phenomenon for some time and have been collecting data on basic water quality parameters at intervals of twice-a-month from Au Tau experimental ponds and some privately owned ponds. (surface samples only). The records show that the chlorophyll “a” levels in experimental fishpond are generally in the range of 200–400 ug/1, but occasionally rise to 700 ug/l. Commercial ponds, presumably due to heavier fertilization and use of manure, have chlorophyll “a” devels up to 800 ug/l, with the maximum levels occurring in August. Scenedesmus and Microcystis are the predominant algal species. Ammonia varied between 20–2 000 ug/l, and reactive phosphate between 10–140 ug/l. The lowest dissolved oxygen level recorded was 0.98 mg/l.
It seems quite obvious that oxygen deficiency is responsible for fish mortalities in Hong Kong freshwater fishponds. The cause is similar to that in Canada and the Philippines, i.e. overproduction of algae and their adverse effect on water quality. Although parameters selected by the research team were sufficient to characterize a summerkill, frequency of sampling (twice a month) was not adequate to cover the critical period, which lasts only several days as discussed earlier (Section 1/3.2.3., Fig. 5). It is quite possible that the actual fishkill periods were missed due to infrequent sampling.
There is also another possible mechanism of summerkill than that described earlier. The oxygen deficiency may take place only for a short time in the early morning hours. This is a result of respiration by a heavy algal bloom during the night causing a temporary exhaustion of available oxygen, but not an actual collapse of the algal bloom. This can be repeated each morning for some time, or only on some days. Only a few fish die, probably the less resistant species or individuals, and the bulk of the stock manage to survive this critical period. This type of fishkill pattern appears to be quite probable as complete fish mortalities have not been reported and the typical disappearance of the algal bloom which takes place during a total summerkill, has also not been observed.
As in the Philippines, only systematic water quality monitoring can explain the cause of fish mortalities in the mixed mullet-carp onds in Hong Kong and, what is even more important from the economic point of view, the possible corrective measures in order to eliminate this phenomenon. Some additional information will also be needed to assess the existing situation. These include: the amount and kind of fertilizer used in the ponds, stocking rates in individual ponds, water level fluctuations during the growing season and corresponding changes in salinity and ionic composition of water, fish biomass at the time of the summerkill, and the final fish yields and recoveries. Monitoring should cover not only experimental ponds operated by the Department, but also private ponds where summerkills are reported to be more frequent and severe.
4.2.1 The water quality parameters and methodology selected by the Fisheries Research Division are necessary and should be continued (DO, ammonia, chlorophyll “a”, soluble reactive phosphate, pH, primary production by oxygen method, BOD). Basic information on major ion chemistry is desired also for an overall characterization of ponds and possible correlation with survival rates (twice a year, to be taken at the maximum and minimum water levels).
4.2.2 Selection of ponds for monitoring: This should include 2–3 experimental ponds (Au Tau Sub-station), 3–4 high fishkill risk commercial ponds (based on the summerkill severity in 1974) plus some additional suspected ponds which should be tested occasionally upon report by the fish farmers.
4.2.3 Frequency of sampling should be increased from the present twice-a-month to at least once-a-week during the critical summerkill period of June to August or September. When the actual summerkills occur, it will be necessary to monitor selected ponds even more frequently (2–3 times a week). In addition to the surface sample, an additional water sample from the sediment zone (10–20 cm above the bottom) should be obtained at the same time intervals by appropriate sampling equipment and techniques (Van Dorn, Ruttner or comparable bottles, adjusted preferably for horizontal, i.e., parallel to the bottom intake). The bottom sample will characterize the decomposition processes due to accumulation of excess food and sedimenting plankton better than the surface sample.
4.2.4 Due to the remoteness of the experimental area from the main laboratory at Aberdeen on Hong Kong Island (about 2–3 hour drive depending on the traffic conditions), and to the probability of early morning fishkills, it will be advisable to stay at the station overnight to sample the selected ponds early in the morning (5–6 a.m.) and deliver the samples quickly to Aberdeen for analysis. Another possibility would be to fix the ammonia (the color is stable for about 24 hours), filter and freeze chlorophyll samples, and deliver them to Aberdeen later. DO determination does not pose any problem as the titration can be performed directly in the field.
4.2.5 Additional diurnal measurements of DO and ammonia should also be made in the selected ponds before and during the critical periods at less frequent intervals due to their time and manpower consuming character. This can be timed with the primary productivity measurements.
It would be useful to initiate experiments with artificial aeration and/or forced water circulation, which appears to be feasible in small stagnant ponds and which could help overcome short lasting DO-deficiency. It would also be useful to set up experiments to compare the effect of the organic fertilizers presently used with inorganic fertilizers to determine whether or not a reduction in the input of bio-degradable and oxygen consuming material will improve water quality conditions in the ponds without adversely affecting fish production.
The summerkill problem in the Hong Kong freshwater fishponds, although not as severe as in Canada or the Philippines, causes heavy losses of fish, amounting to 30–40 per cent of the expected total production. Water quality monitoring, if performed with adequate frequency, may provide background data needed for maximization of aquaculture production and for finding efficient and economical corrective measures to avoid summerkills.
