During the late 1970's a new area of brackishwater ponds was constructed by the Government of Malaysia in a mangrove area of Johore State. For three years, beginning in 1981, an intensive research effort was made at this 20 ha facility by the national staff of the station, by a number of fulltime FAO/UN personnel, and by consultants working in cooperation with the others. This research programme, which involved a broad spectrum of studies, included (i) diet development for several fish species and for penaeid shrimp, (ii) comparision of tidal and pumped water exchange economics, (iii) water circulation and aeration effects, (iv) bacteria contributions to pond food chains, (v) pyrite oxidation impacts on pond chemistry, (vi) laboratory incubation experiments, and a number of other complementary projects. The results have been recently compiled (Cook, Pongsuwana and Wechasitt, 1984). Throughout the period of research the yields of shrimp, including intensively fed culture batches as well as those grown following lime application, were relatively low (far below the yields in excess of 1 t/ha/yr which have been accomplished in some semi-intensively managed ponds with good soil conditions). Sufficient information was available to demonstrate that the presence of considerable amounts of pyrite in the dikes and pond sediments, derived from the sediments of the mangrove forest present prior to construction, was a major factor in limiting the success of shrimp culture operations. In extreme cases following heavy rains, stratification of the pond waters occurred with a deeper layer about 20 ppt salinity, near neutral hydrogen ion concentration, and relatively low dissolved oxygen. The surface layer had about 15 ppt salinity, abundant oxygen and almost one thousand times higher hydrogen ion concentrations as the result of acidic runoff of rain-water from the dikes. Shrimp stressed by low dissolved oxygen in the deep layer would swim to the surface and encounter a completely unacceptable chemical environment, resulting in the death of many of the culture organisms (Cook, Pongsuwana and Wechasitt, 1984).
Surveys of the natural buffer capacity in the pond waters at the research station showed that a significant fraction had been removed by reaction with acid during periods free of rain fall, even with half of the water renewed daily by tidal exchange. Thus substantial acid is supplied to the ponds even in the absence of direct runoff of rainwater from the soil dikes.
Based on laboratory measurements of total acidity potential of the dike soils, and extrapolation of the rates of acid leaching by normal water exchange practices at the research station, as well as acidic runoff of precipitation, the time scale for removal of the pyrite was estimated to be years to decades (Simpson et al., 1983).
Most of the detailed research on processes and effects of pyrite oxidation in acid sulfate soils has been related to land agriculture, since the problem represents a major barrier to high yield growth for nearly all of the cultivation options, including rice farming. Some of the most detailed work on acid sulfate soil distribution and development has been done in Thailand (Moorman and Rojanasoonthon, 1968; van Breeman, 1976). In the Bangkok Plain, the active tidal marsh zone contains some acid sulfate soils, but the main area of such soils, covering a total area of about 800,000 ha lies between 30 and 100 km of the coast. The parent materials have a total thickness of 5–10 m of pyrite-rich deposits (van Breeman, 1976) which were originally accumulated in a tidal marsh environment up until about 3000 years ago (van der Kevie, 1972), and have been subsequently uplifted to slightly above sea level. Soil containing minerals derived from pyrite oxidation products usually has thicknesses of 1–2 m, and has been used for a variety of agricultural purposes. Yields of rice in these soils have been considerably lower (~0.5 – 1.0 t/ha/yr) than for riverine deposits free of pyrite (1.75 – 2.5 t/ha/yr). In the tidal areas, which have less severe acid sulfate problems, the economic return in income for shrimp ponds, salt evaporation pans and coconut plantations is more than a factor of 10 greater than an equivalent area of paddy fields (Glopper and Poels, 1973).
Brackishwater aquaculture in Thailand has been applied fairly successfully for shrimp (annual production around 12,000 t in the early 1980's). Several factors have contributed to this success, despite the existence of major areas of acid sulfate soils in the country. In general the brackishwater ponds lie well outside the zone of highest pyrite abundances, and are situated in regions which have been used for salt production for more than a century. Thus even if the local soils initially had considerable pyrite, repeated drying and pond refilling cycles over many decades would have removed much of the problem of acid production. The potential acidity release measured on soils from a few Thailand aquaculture pond dikes was less than one percent of that from soils in Malaysia at recently constructed pond sites in former mangrove forests (Simpson et al., 1983). Secondly, the Thai shrimp ponds are relatively large and are managed semi-intensively, with natural detritus and primary production as the principal food source, and low-head pumping to accomplish relatively infrequent water renewal. With a very small proportion of exposed dike soils, less rainfall, higher fertility brackishwater, and far less soil pyrite, the shrimp yields are considerably greater than for more intensively managed small ponds with less favourable soil in other areas.
Under most conditions pyrite in mangrove forest sediments is inert because of relatively stable reducing conditions. However, if sediments become exposed to the air because of lack of rainfall for considerable periods, then substantial acid release can occur and produce unfavourable water conditions for large areas of natural fish habitat. Such occurrences have been well documented in peninsular Malaysia (Dunn, 1965).
The extent of oxidation of pyrite varies considerably in exposed acid sulfate soils. Deep in the soils, the pyrite is largely unaltered and the soil water has relatively low acidity, while within the zone from 2 m below the surface up to perhaps 10 cm of the surface the occurrence of oxidation product iron minerals and high soil moisture acidity is readily observed and in situ conditions are quite unfavourable for both plant and animal growth. In some acid sulfate areas the upper soil layers (0–10 cm) have less acidity due to downward leaching of pyrite oxidation products. In peninsular Malaysia, where the total extent of acid sulfate soils is in the order of 100,000 ha, the upper 20 cm of soil in some areas has relatively low oxidizable sulfur (≤ 0.1 percent), even though the deeper layers contain abundant pyrite (Tai, 1969). Thus the variation of acidity potential with depth in the soil can be considerable in an acid sulfate soil area. Minimizing soil disturbance for aquaculture pond construction can, in such environments, lessen considerably the production of acid by pyrite oxidation during subsequent culturing operations (Pedini, 1981).
Another country in which extensive brackishwater shrimp aquaculture has been established in recent years is Ecuador. Total annual shrimp landings in Ecuador increased from about 7,000 t to more than 20,000 t between 1976 and 1981, primarily due to growth of aquaculture. Most of the ponds were constructed in alluvial flat salty soils which were free or almost free of vegetation, thus reducing considerably the cost of clearing land. These soils appear to have considerably less pyrite abundance than the adjacent mangrove forests. By making very large ponds (over 10 ha in many cases) within dikes constructed from materials scraped from the top few centimeters of the entire surface of the pond, or dredged from the peripheral canal, there was little disturbance of the soil which became the pond bottom. Thus two of the initial choices, (1) the pond siting, and (2) construction technique, were probably critical in permitting reasonably good yields of shrimp from these new aquaculture facilities. This approach for site selection also limited the damage caused by the reduction of nursery grounds for valuable marine and estuarine species which results when ponds are built at the expense of mangrove forest. This topic, however, is not totally relevant to this paper and has been dealt with in many other papers, e.g., Ong, Gong and Wong, 1980; Saenger, Hegerl and Davie, 1983; Gedney, Kapetsky and Kuhnhold, 1982; Kapetsky, 1982.
It also has to be said that extreme soil acidity does not seem to be a serious problem in Ecuador, even in the case of ponds built inside mangrove forests, as frequently happens in the provinces of El Oro and Machala, where problems of acid leaching have not been reported. The reasons for this are surely complex but may be linked to the relatively high freshwater runoff into the Gulf of Guayaquil, and perhaps also to difficulties for the establishment of sulfate-reducing bacteria, due to either dilution of SO2-4 by freshwater or drastic variations in its availability, which may have limited formation of pyrite.
In contrast to the experience in Ecuador, in other Latinamerican countries, where acid sulfate soils existed, as in Costa Rica, the excavation of shrimp ponds inside the mangrove in 1973–74 led to such serious problems of acidity, in spite of the large size of the ponds, that the work in this area was finally abandoned in 1983 with considerable economic losses.
One of the two countries with the largest brackishwater aquaculture development is the Philippines, which has a total pond area of about 196,000 ha. Since the early 1950's, the total area of ponds has approximately doubled, and the total production of milkfish, the predominant culture organism, has increased by about a factor of four (Table 7). The yields have increased from 350 kg/ha/yr to a little less than 800 kg/ha/yr over about 30 years. The aggregate yields are still far below those from the most productive fish farmers (in excess of 2 t/ha/yr) in the Philippines (Chong et al., 1984) for a number of reasons. Those operations which use intensive fertilizer application have substantially greater production (~1.7 t/ha/yr) than those which use no fertilizer (~0.3 t/ha/yr). Acidity levels in the intensively managed, more successful ponds are also about a factor of 5 lower than for those in the lowest yield category. Many factors contribute to culture organism yields (Smith and Peterson, 1982), including the quantities and types of fertilizer application, experience and expertise of the farmer, and a host of environmental factors (Chong et al., 1984). The experience of milkfish farmers in the Philippines indicates that time-scales of several decades are probably necessary to accomplish major improvements (factor of two) in aggregate yields in brackishwater systems. Some of the increased yields have clearly derived from intensive fertilizer applications, but it also seems plausible that some of the yield increases may have resulted from gradual removal of acidity potential from the pond complexes. There are some indications that this may have really happened. In previous reports (Tang, 1973) it was estimated that 60 percent of the total area of milkfish ponds had problems derived from acid sulfate soils. It must be noted that at the time of Tang's report many of the ponds considered were relatively “young”. In a recent survey (Chong et al., 1984) a total of 322 farms were analysed for soil pH by the Bureau of Fisheries and Aquatic Resources in six different provinces, and the soil pH was never lower (on average) than pH 5.3, which, although acidic, is not as serious as the levels found in newly built ponds in Malaysia. Age of the ponds (averages for provinces) was less in provinces with more acidic soils (see Table 8). Although derived from three different sets of data, these trends seem to suggest a progressive leaching of the dike and sediment acidity due to tidal water exchanges over several decades.
This point is more important than it appears at first sight. It has to be considered in the light of several large-scale investment projects promoted by national governments in Southeast Asia and financed by bank loans. These schemes normally require considerable alteration of water supply and drainage for the areas considered for renovation, which have normally developed in a rather chaotic way over the years. In several cases, a complete new design of pond distribution, including substantial deepending of the ponds for shrimp culture, is planned for the pojects.
The undesired effect of such reconstruction in areas which have experienced a rise in soil pH over the years may be an abrupt return to acidic conditions when deeper layers of acid sulfate soils are exposed to air and oxidized. The economic implications of such a “surprise”, in terms of delays of a number of years to reach expected outputs, can be easily imagined.
