Brackish-water cultures are subject to environmental pesticide contamination from various sources, mainly through its water supplies from rivers and the coastal seawater. Also direct inputs through aerial transport might occur when, in the neighbourhood of the tambaks2, spraying for agricultural or anti-malaria purposes is carried out. Added to this are the residues which might remain in the tambaks from application of pesticides before or during the cultures of fish, in particular, from those which have a certain persistence.
Our objective has been to make a pilot study of various samples of fish, sediment and water mainly taken from tambaks and adjacent sea-areas (Figure 4). The results, taking into account the accuracy of the methods (Table 2), give a fairly good indication of the kind and the level of pesticides in the samples.
2 Brackish-water fish and shrimp ponds
Fish and shrimp were collected from tambaks of the Jepara centre and from those around Surabaya. The specimens of Surabaya were dissected on the spot, preserved and brought to Jepara as described in 2.3. For the conditions of the analysis and the gas-chromatograph (ECD) see Tables1 and 3 -B respectively.
For both locations, the spectrum of pesticides in fish and shrimp was almost identical: DDT and its metabolites TDE, DDE and DDMU occurred in the samples, mainly being the para, para (p,p') isomers. Polychlorinated biphenyls (PCB) which are in Europe and the USA often present in a 1:25 ratio to total DDT (DDT and its metabolites together) were absent, or at least below the level of detection of 0.1 ppb/wet weight (ppb = parts per billion 109, or nanogramme per gramme; nanogramme = 10-9 gramme). In addition to DDT and its metabolites, which mainly reflect the overall contamination of the environment with this insecticide, the residues of a few pesticides were found which have been applied in the ponds. This was mainly BHC (alpha and gamma BHC) and Thiodan (Endosulfan), where the presence of 2,4D-isopropyl ester was suspected in some samples. The results are presented in Table 6-A for flesh, stomach and content and liver, all expressed in ppb/wet weight.
It was not possible to establish some relationship between the feeding habits of the different species and concentrations of the pesticide, since relevant information is still too scarce. Only the ratio of DDT/total DDT gave some information on whether the intake of DDT is of recent origin or not. This DDT/total DDT ratio in the liver of several species was also found to vary with the application of Thiodan as a wild-fish killer. The results, plotted in Figure 5, illustrate this. Probably during killing by Thiodan, DDT has been rapidly accumulated within a day. As already mentioned before with Diazinon and Leptophos, this process might be related to the disturbance of the breakdown and exoretion processes, while uptake was continuing. The fact that this was rapid also suggests that the level of DDT in healthy organisms is a result of an active balance between uptake and loss, and not the result of a slow accumulation over long periods.
With help of the Directorate General of Fisheries at Jakarta, we obtained a number of sediment samples from several tambaks and adjacent estuaries of Indonesia brought to Jepara by students who participated in the training courses in fishculture at the centre. The locations are indicated in Figure 4 representing some major tambak areas in Indonesia. All sediments have been freeze-dried in Jepara, and analysed following the direct extraction method described in 2.3 (Table 1-B).
As can be seen from the results given in Table 6-B, the concentrations of the pesticides were rather low, in particular, of the “environmental” pesticides DDT and its metabolites. Thiodan and BHC were equally present, as was sometimes 2,4 D-isopropy ester. Dieldrin and Endrin in these samples were below detectable level, i.e. below 0.05 ppb/dry weight. PCB's (polychlorinated biphenyls) were absent, which indicate the absence of pollution from marine paints and industrial pollution, to which PCB's are frequently connected. One high value for BHC indicated a recent BHC application for irradication of chironomid larvae. Since the residue was mainly alpha BHC, which is 500-1 000 times less toxic than gamma BHC, this cannot be accounted as a major contamination.
Only a few water samples have been analysed and the results of these analyses are given in Table 6-C. In general the level of pesticides in tambak water is very low, except as for Leptophos shortly after an application of that pesticide. The low concentrations in normal tambak water are understandable, knowing the tendency of most pesticides to adsorb to sediments. The fact that the tambaks are very shallow explains why the water contains very little pesticide. Also the relatively high temperatures and active processes favour rapid degradation.
Marine fish and shrimp samples have been analysed, mainly from Surabaya and Jepara coastal waters. Marine fish were bought from the fish markets. These, like the tambak samples, were dissected, weighed and placed in hexane for later analysis.
It is interesting to note that except for those pesticides generally applied in the tambaks (like BHC and Thiodan), marine fish samples showed higher values of pesticide residues than the tambak samples (Table 7-A). Secondly, around Surabaya, the sea fish extracts contained a peak in the chromatogrammes from which it was expected to be 2,4 D-isopropyl ester. Tests were set up using different columns in the gas-chromatograph, and acid and alkaline treatment of the extracts were carried out which verified the identity of this compound.
Another observation was that in one large milkfish, caught from the sea at 5 miles from Jepara, relatively high amounts of DDT were found, and for the first time also PCB. This PCB matched Arochlor - 1260, a product used as insulator liquid in large transformers. Where this milkfish might have been contaminated is so far unknown since only one sediment sample of our collection contained slightly the same PCB, namely that collected near the landing centre at Surabaya. Whether there are dumping sites of industrial wastes is also unknown, and no information was available at the Directorate General of Fisheries.
Marine shrimp contained noticeably less pesticides than fish, a result which was also found for shrimp cultivated in tambaks. This is probably related to the very low lipid (fat) content of the flesh of shrimp which is approximately only between 0.03 and 0.07%. For the fish species we investigated, the lipid content of flesh varied from species to species between 0.1 and 7%, the highest being for milkfish.
Since most of our marine sediments were estuarine sediments, their concentrations of pesticides need not necessarily be representative for the open sea sediments, but are of direct interest for the tambaks. Analysis were made by similar method as was carried out for the tambak sediments (Table 1-B). In Table 7-B the results of the measurements are given.
In general the impression was obtained that the estuarine sediments contained slightly more pesticides than the sediments of adjacent tambaks. Probably this is caused by land run-off of pesticides, which is marked, in the samples of Medan and Jakarta, with the presence of Dieldrin and Endrin. Only for one sediment sample the chromatograph reflected a spectrum similar to the PCB Arochlor 1260, although the spectrum was less clear than when measured with the one milkfish mentioned in section 5.3.1. The sediment containing the PCB, was sampled from Surabaya, the second largest city in Indonesia.
A few water samples have been taken from some obviously polluted waters from the cities Jakarta, Semarang and Surabaya. The objective was to investigate the content of organic load and concentration of pesticides and eventually PCB's dissolved in the water connected to the suspended matter.
Water, filtrated over a GF/C Whatman glass fibre filter, has been analysed for BOD (Biological Oxygen Demand), COD (Chemical Oxygen Demand), Phosphate and Oxygen. Pesticides have been determined from small volume hexane extracts, 20 ml for 200 ml water. (Table 1-A.)
The polluted waters of black colour contained organic matter loads between 2 and almost 60 ppm COD (as mg 0/1), with high BOD values as well, which could hardly be determined since oxygen was already lacking (therefore we tried with 1:1 mixtures with distilled water, of. Table VIII-A). A striking factor is that in some waters PO4 was completely lacking, which implies that potentially natural phytoplankton growth and thus the natural oxygen production by photosynthesis is blocked. A conclusion could be that fertilization of the heavy anoxic waters might improve the situation. Tambaks are equally shallow waters exposed to a high stress of organic matter, but being fertilized, the photosynthesis compensates the oxygen demands. Tests with additions of phosphate in quantities of 20–100 mg P/1, at the points where the aerobic system converts in anoxic system, might reveal the possibility of controling the aerobic/anoxic system.
The pesticides in solution show similar spectra as found in sediments. However, the concentrations are below the ppb level. PCB's could not be detected, which means they did not occur above the level of detection of 0.005 ppb.
The flass-filters containing the suspended matter from filtration were extracted with hexane, after being dried in a desiccator for the determination of the weight. Other filters were used for the determination of the ignition loss by heating in a porcelain orucible to red hot temperatures.
The hexane extracts contained approximately the same pesticides which were found in the extracts from the related waters (Table 8-B), except in higher concentrations. The percentages being in suspension are equally calculated, and gave values from about 6% for alpha BHC, to 91% for DDE being in suspended form. Endrin was found to be present in suspension in Jakarta waters, indicating that this pesticide is still probably used for domestic reasons.
