(20–21 October 1978)
This is a main centre for carp-rearing and the Training School has a large complex of ponds. Water flows by gravity from a feeder stream and then from pond to pond. The water in all ponds is opaque, often yellow and with a covering of red or green algae. Clean water is required for one basic purpose - to provide a medium free from the parasite Lernaea in which carp fry can be raised.
After a classroom session with the students, it was decided to install an SWS system in one of the ponds. The bottom had a layer of c. 10 cm mud. This was sealed by covering a circle of c. 1 m diameter with stones of up to 5 cm to a depth of c. 10 cm. This area was covered with c. 10 cm sand and the Unit placed in position. Sand was then tipped on until a large dome was formed and the Unit top was covered by at least 10 cm. The sand was ‘as dug’ with an excess of material below 1 mm, and much of the development consisted in removing this fraction. The system was pumped slowly for c. 5 hours on 20 Oct., with numerous stop/starts, each bringing out fine sand and clay. After one hour, the filtrate was cleaner than the raw water. The pump was restarted at 6 a.m. on 21 October and pumped for 2 hours; two further stop/starts were made and within 14 minutes, the filtrate was clear, with only a faint opalescence which improved over a further 30 minutes before pumping stopped.
Comment: Unfortunately, there were no facilities for measuring particle size or identifying suspended particles. However, it seems probable that the yellow opacity is largely due to organic debris and algae. The residual pale milkiness is caused by a clay particle, probably smaller than c. 1 micron, resulting from breakdown of felspar from the granite, and this is typical of much water in Java. The distribution between mineral and organic material is important, for the former should be broken down biologically leaving negligible solids, while the latter builds up. Most of the particles removed are probably organic. It is probably relevant that when some SWS filtrate taken in December '77 from a similar source was analysed, it showed a markedly reduced D.O. reading. There had been no time for a biological filter layer to form and the likely explanation is that there was aerobic breakdown in the Schmutzdecke.
The site chosen had been used for a successful trial/ demonstration in December 1977 (see Appendix 7). Since it had been shown that clean water could be obtained and since the carp pond trial had been in somewhat similar water, emphasis was placed on training teams in correct installation. The students were divided into two groups, each of which did a complete installation in running water, including stop/starts, release of suction and blowing back. A question session before dispersal showed that there had been wide comprehension of the system and its application.
General comment on water quality: With a dense population and with a large acreage of well-stocked carp ponds, all open water is of poor general quality and the average readings for ammoniacal nitrogen, BOD, nitrates and, sometimes, detergent are likely to be high. The effect of SWS sub-sand abstraction is expected to be as follows:
Complete exclusion of all algae, larvae, amoebae etc. down to c. 2 micron as soon as development is complete.
Removal of most of organic debris and thus reduction of BOD.
Removal of about 50% bacteria within 24 hours of stabilisation and perhaps above 90% after formulation of biological layer, which will take c. 7 days.
Breakdown of above 70% of ammoniacal nitrogen and BOD and at least 50% of detergent in the biological filter layer.
The action of b, c and d (above) all depend on oxidation and continuous pumping. The filtrate will thus have a low D.O. reading and high CO2: both are quickly corrected by simple cascading, injection, etc.
River and stream - These have several unfavourable factors:
Many sections are mountain torrents with beds of boulders and subject to rapid rise and fall.
Some rocks break down to fine clay and their beds consist of pockets of mud among boulders.
Some streams have cut deeply into soft earth beds: even if sand is present, the sites pose major pumping problems and are also subject to rapid rise and fall.
Stream beds are the main source of sand for concrete making and many accessible areas are regularly exploited.
However, there are a few sites, such as the stream used for the above trial, where installation is possible. The level seemed to be stable and occasional spates will serve to clean the bottom surface.
Fish ponds - It is obvious that the carp can survive and grow in the poor water normally available and there is little case for general filtration. However, the demonstration at the Training School shows that an artificial bed without a pond is effective and cheap. Sub-sand abstraction is, therefore, worth considering for that part of the water used for stages when eggs and fry are vulnerable to Lernaea and other parasites. Most farms are on slopes. In some cases, water could be drawn by gravity once the system has been fully developed by mobile petrol pump. Full instruction for installation and maintenance is given in Appendix 1.
(24 October 1978)
Most of the immediate hinterland of Semarang is flat land, heavily populated and farmed, with canalised streams tapped for irrigation of paddy fields. These areas are largely clay/silt and rarely offer any facility for in situ sub-sand working. No stream was found within reach where the straightforward use of SWS Unit in the bed could be taught.
A reconnaissance was made to a fishing village near the mouth of the Tumtang River and some 50 km from Semarang. The water is salt at this point. Deposits of sand at the water's edge were sampled. Although there was a small mud content, the specimen consisted basically of a fine black volcanic sand typical of the N. Java coast but with an adequate fraction of larger particles of assorted shell. If this deposit is 1 m deep, it could be exploited by either SWS Unit or SS screen well, and good clean water could be expected. Probing by pole was inconclusive and unfortunately, the SWS test equipment had not been taken. The site was too distant to justify a second visit - taking well over 3 hours - especially since no particular use for clean sea water was envisaged.
(25 October 1978)
The pond selected contained Tilapia and Gourami: the raw water was pale yellow in colour, due mostly to algae, and viewed in a 50 mm diameter beaker, it was opaque. To demonstrate a variation on the method used at Sukabumi, the Unit was stood on four large stones, one under each corner, holding it c. 10 cm above bottom. The whole was then covered with what was ordered as 2 m3 sand but was probably not very much over 1 m3. The only grade easily available had too high a fraction of -1 mm particles, most of which was evacuated during development and, therefore, wasted. This factor also delayed stabilisation. The water cleared after c. 90 m in situ pumping and continued to improve until it had only a faint milkiness, characteristic of the fine clay particles in many Java waters.
There were no facilities for monitoring but it was obvious that all algae were being excluded and, presumably, the assorted organic breakdown particles found in such fish ponds. All staff assisted in the work and the actual fish farm workers were especially helpful. All stages were explained and there seemed to be good understanding.
The Unit was then recovered by pumping back, showing how simply it can be moved and resited if necessary. The use of the SS screen well was then briefly demonstrated in the sand bed prepared for the Unit. Unfortunately, no deep sands were available for fuller demonstration.
Detailed procedure for establishing SWS system in fish ponds is given in Appendix 1.
(26 Oct. '78)
This seemed to be a useful session, for abstraction of clean sea water is causing serious problems which the staff were keen to discuss. The present system has gravity feed to a sump near the working area, from which raw water is pumped to an overhead reservoir/sand filter which is ineffective and wastes time. (It was assumed that the flow was continuous but this was not asked specifically. If, as is likely, this is an intermittent regime, the trouble may be the same as that diagnosed at Probolinggo, discussed below.) The suggestion was made by one of the scientists that the reservoir should house instead an SWS Unit in a well-prepared and developed bed. This is wholly practical and since it would involve little alteration of plumbing, we recommend it. Detailed suggestions for working are in Appendix 2.
In the long term, however, it would be simpler and more effective
to use the nearby beach which consists mainly of broken down coral with
a wide range of particle size that would make an excellent in situ medium.
The use of the yellow probe for bed assessment was demonstrated in detail
and then pumped for one hour, when the water was distinctly better than raw
sea. However, the pollution is so severe that top quality water is impossible
until the beach has been cleaned radically over a period of days.
Beneath a thin clean crust was a layer of up to 10 cm deep permeated with
black debris. Some measure of the condition is suggested by the following:
a small handful put in a beaker had three changes of water with vigorous
agitation before losing most of the fine black debris. Disturbance by the
probe left the sea black. Sea pollution of this degree had not been seen
previously but procedure for treating freshwater sites of this nature is
given in Appendix 11, and similar action is needed here, namely daily jetting
by probe for one week, covering approx. twice area needed for
abstraction and cleaning to depth of c. 50 cm. This has a double effect.
Physically, it will wash out the black debris and other unwanted fines, to
be scattered by wave action. At the same time, aerobic breakdown can
begin and at a temperature of c. 30°C this will be rapid. Once a sub-sand
system has been developed and is pumped continuously, the bed should remain
aerobic unless on-going pollution becomes overwhelming, when the raw water
might have too low D.O. content to maintain aerobic conditions.
Apart from the above, the site is ideal for abstraction by stainless steel screen well. Detailed procedure is given in Appendix 7. With a tidal amplitude of c. 1 m on a fairly steep upper beach and only c. 20 m run to the perimeter of the Institute, installation would be simple. However, it must again be stressed that vigorous cleaning is essential, followed by full development and continuous pumping.
Since at certain seasons it could be useful to draw water from under or near the ponds, the use of the yellow probe for exploration and testing was demonstrated, with the possibility of finding a stratum of sand below the mud such as is characteristic of the Tegal area and elsewhere. In fact the thick mud, with numerous shells, was shown to extend to at least 2 m. This result was expected but the main objective was to show how such facts could be established within a few minutes and with little labour.
(28 October 1978)
This Research Station works on giant prawns (Macrobrachium) and is sited on a section of coast with foreshore wholly of mud and mangrove. The tide has an amplitude of c. 1 m but recedes at least 400 m. The potential value of sub-sand working was discussed, but it was clear that with a continuous demand for only 6 m3/h in the immediate future, a new pipeline would be totally uneconomical, even if a suitable intake site could be found. In fact, it was stated that the nearest known sand foreshore was one hour distant and that the coral sand used for filters, etc. is brought from an offshore island.
The existing intake and filtration system is somewhat as follows: A series of stone-lined canals allow water to flow at high tide to settling chambers from which it is pumped to an overhead gravity sand filter which is run on approximately alternate days, feeding into storage. The filter gives much trouble and water quality is poor. Gravity sand filters, though apparently simple, are often unsatisfactory but here the basic trouble may be the intermittent regime. The most important role of such a filter is its biological action, which depends on continuous flow to maintain its aerobic state. This is especially important when temperature is around 30°C, for biological activity is high and saturated D.O. low. In this regime, the biological filter is not complete until about the eighth day, so that the filter has hardly started to develop before the flow stops: the D.O. is quickly exhausted and any organisms die, except perhaps for those very close to the surface. Full biological filtration is never achieved and, even worse, the dead protozoans etc. and breakdown products are drawn through. Further, D.O. is probably far below saturation and BOD high. In Appendix 2, detailed procedure is suggested for using an SWS Unit to abstract water from an enclosed filter bed.
Meanwhile, two suggestions have been made which could simply and quickly improve water quality from the existing circuit -
When filter is started, run to waste for up to an hour to evacuate dead organisms and metabolites.
Raise D.O. level by casdading, injection, etc. of water.
The test probe was demonstrated for exploring beds at depth, but mud was found to extend to at least 2 m.
(30 October 1978)
This is a large centre for breeding Carp, Puntius and Gourami, with a large complex of ponds and sluices with typical carp water quality. Excellent sand is available within 3 or 4 Km at 1,000 Rupiah/m3 delivered. Unfortunately, the truck broke down with punctures and over 2 hours were lost. The Unit was placed in a prepared bed after excavating mud over an area of 2 × 2 m to depth of 30 cm in base of pond: this resulted in a fairly hard bed, which was then covered with round granite, etc., pebbles average size of 8–10 cm. Digging up the mud left the bottom and sides very dirty, and as water flowed in, it became opaque at 1 cm thickness. After pumping for 1 hour, good quality water was obtained at about 4 m3/h. This was steadily improving and volume increasing when operation had to cease. Removal of Unit by pumping was then shown and also the use of the stainless steel screen well in a bed of sand. With good medium at low cost available, there is no need to use minimum beds.
(31 October 1978)
The small fishing harbour near the town of Sampang was visited with the object of demonstrating the use of well screens for abstracting salt or brackish water needed for fish-washing, etc. in what was said to be a region of coral sand. In fact, the shore consists of soft clay and silt, with perhaps 20% stones and shell debris, and is wholly unsuitable for this system.
In the hope of finding a more permeable lower stratum, the beach was jet-probed to 2.5 m in several places but without success. The tidal amplitude is 3 m, with recession of 400 m. If a supply of clean sea water was essential, it should be possible to establish an artificial bed near L.W., but it would be expensive in pipe line alone, while the long line would make it difficult to run, especially on the intermittent regime imposed by electricity being available only between dusk and midnight, or earlier. An easier solution would be a bed excavated and filled with coral sand at a point to allow pumping and storage for 2 or 3 hours over high tide, but this also poses many problems. Ample coral sand can be brought from a few kilometers inland.
The problem causing most concern to Fisheries officials was the very poor quality of fresh water, which is always slightly brackish (though usable) and contaminated with fine silt. This is described and a solution suggested in Appendix 3.
(2 November 1978)
The new fishing harbour of Muara Angke at Jakarta was inspected briefly, where a water source is needed to provide c. 20 m3/day for fishwashing etc. About 50% of this is needed during the morning. Such sand as was visible was the uniformly small black grain typical of N. Java beaches. Since it extended to c. 30 cm, the deepest that could be dug under water, it seemed to justify a jet-probing trial. This was, therefore, planned for morning of 3 November and a group of officers from Fisheries H.Q. came to assist. It was also felt that even in the probability that the site would be useless, training in the use of jetting would be valuable.
The immediate area of the Harbour was tested first. Part was underlain by concrete at under 1 m. Just within the harbour itself, silt and/or mud extended to at least c. 2 m and although it contained some coarse gravel, it was useless for abstraction. The equipment was then moved c. 200 m to a point between mangrove clumps where surface sand included larger particles and seemed more promising. In fact, this sand extended to a maximum depth of 40 cm, below which was dence mud/clay.
The sand quality and, particularly, its shallow nature, made direct abstraction impossible but another overwhelming factor that prohibits its use is the massive pollution. The inshore water is full of urban refuse including plastic, tins and assorted containers, and organic waste. The shore is lined to a variable depth while jetting showed that it extended to below 30 cm in the sand. In addition, the sea itself is grossly polluted and no method of abstraction through this beach could make it usable. This beach was incomparably the most polluted on which I have ever worked and, looking back, I doubt if I was justified in asking anybody to work in such filthy conditions without full protective clothing. In fact, everybody was most helpful and the exercise was, after all, worthwhile for the training given in various aspects of jet-probing, both for survey, abstraction and cleaning.
My first reaction is to reject totally any thought of using such a source. However there is a great need and the benefits of success would be big, while the effectiveness of the system can be simply tested by using equipment already in hand. The tidal range is only 0.5 m and is, therefore, not a problem. A suggested programme is in Appendix 4.
Advice was also asked about freshwater supplies for the new fishing village close to Muara Angke. A possible system is detailed in Appendix 4.
(2 November 1978)
The fish farm at Ciracas was inspected briefly after planning work at Cianjur, mainly to consider running a second fishpond demonstration on 4 November. This was found impossible because it was a much longer journey than expected and there would be no time. Further, the basic problem is industrial pollution, for the stream supplying the ponds receives effluent from 2 nearby factories, one a milk-processing plant and the other making packaging materials.
Only Gourami gourami seems to survive: this is a labyrinth fish able to obtain most of its oxygen directly from the surface air and a low D.O. figure is suspected. There was little suspended material in the stream and it appeared less coloured than average. The low D.O. is presumably caused by soluble pollutants, especially from the milk plant, and any sub-sand abstraction trial would be meaningless without monitoring such parameters. It was, therefore, suggested that the water was analysed. These facts would decide if such a trial would be useful. Meanwhile, it would also help to know what volume needs to be treated and the final quality that would be required. The stream seemed to be fast flowing and, therefore, capable of being aerated cheaply.
(4 November 1978)
A Unit was installed in an artificial bed in the presence of a number of senior officials, and pumped for c. 1 hour, by which time the filtrate was distinctly better than the raw water and steadily improving, even though the sand available was only half of what had been expected. Progress was entirely typical of ponds worked at other centres, where the water is rich in algae and organic debris and has a constant clay suspension. Recommendations for detailed action are in Appendix 1.
(5 November 1978)
At the request of Mr. Rustami, Director of Research for Inland Fisheries, a visit was paid to the Research Laboratories of the Brackishwater Centre, Jakarta, to discuss problems of closed circuit filtration in hatcheries, laboratories, etc. Suggestions for 2 simple and economic filter systems are given in Appendix 5.
Two main needs exist:
Research and Breeding Institutes, etc. - High quality water for laboratory work, egg and fry stages of fish, prawns, etc.
Fishing Harbours and markets - General purpose water for fish-cleaning, washing down, etc.
Much of the north Java coastline is fringed with a broad band of mangroves and mud, now largely made into brackishwater fish ponds, and from most of this zone direct abstraction of clean water is impossible. However, there may be sites where the mud is underlain by permeable sand/ gravel, as in the river mouth at Tegal. If this is suspected, the use of test probe will quickly give an answer. (See Appendix 7.) Normally, the only practical solution is to bring in raw water by artificial channel and filter on shore after settlement, as at Probolinggo.
Pollution is already affecting many miles of otherwise exploitable sand beaches. At some sites, as at Jepara, radical cleaning can make a beach usable, but around Jakarta, the shore is lined with a broad belt of refuse, which extends to a depth of at least 50 cm, and direct use cannot be made of the sea bed even if sand were present. Where there is no reasonable alternative, the use of an inshore artificial bed is worth considering, but this is not a simple solution while gross pollution persists. Suspended solids will be removed and if the system is run continuously, it should serve as a biological filter, reducing BOD, bacteria, etc. However, only actual monitoring will show how effective it is in reducing such a level of pollution; we cannot speak from experience for no SWS system has been run in such adverse conditions. The possible benefits are such that we believe it could be worthwhile to run a month's trial along the lines suggested in Appendix 4.
Massive soil erosion, resulting in large volumes of soil, mostly fine particles, being carried down to the sea and deposited on beaches is bound to reduce further the usable reaches.
At all station, the need was stressed for parasite-free water for the ova and fry stages of Carp. Demonstrations at four centres showed that local or semi-local sand could be used for an artificial bed for the SWS Unit from which clean water could be abstracted. In hilly areas, once the system has been developed fully by pumping, it may be possible to draw small volumes by gravity, with no cost for power. Detailed instructions are in Appendix 1.
The opportunities for direct abstraction from rivers are limited by factors discussed under ‘Geology’ but there was certainly areas where this can be used to provide high quality water and it is worth considering low volume gravity abstraction to supply small hill communities with drinking water.
Experience at Jakarta, Semarang and elsewhere pointed the problem of being based in large cities, but it is hard to see how this can be avoided, since it is essential to meet senior officials. Several problems arise. Most potential sites are rather distant and several hours may be spent in travel. Some otherwise possible demonstration sites within or close to city boundaries are so grossly polluted that no staff could be asked to work in the water. One such site was inspected and rejected at Semarang, on the Gurang River. At one point, it was covered with detergent foam, while bubbles of H2S and Methane were rising from the bed. A trial would have been justified only to check whether sub-sand abstraction would lower the pollution parameters enough to allow discharge, as is being done in U.K. and elsewhere.
Some degree of pollution was known to exist in Java but the extent and severity was unexpected. Experience in more or less worldwide sites has shown that many systems can be prepared and developed within an hour or so and this had been true of the Java sites worked in December 1977. With beds saturated with dirty water and pollution, it is seldom possible to obtain crystal clarity within a few hours, so that work normally had to stop before completion. This applied particularly to the beach alongside the Research Station at Jepara. The coral breakdown material was potentially a perfect medium from which the water would normally be crystal clear within fifteen minutes, but it was permeated with black debris.
It seems probable that in many streams the D.O. level is minimal, with massive BOD counts. This is particularly true where streams receive raw sewage every 10 or 20 m and also large volumes of refuse. Working in such areas is hardly water abstraction: it is better described as treatment to bring effluent within acceptable limits for discharge. Removal of suspended particles is basically a physical process and numerous tests in many different sites have shown that within a few hours of establishment, an SWS system removes matter down to c. 2 micron, as well as up to 50% of all bacteria. In water such as described above, the exclusion of organic debris causes-massive reduction of BOD but further improvement of the filtrate (reduction of bacteria, ammoniacal nitrogen, soluble BOD, etc.) is dependent on formation of an aerobic system, where a limiting factor is likely to be the low D.O. It may be worth considering increasing this by aeration of raw water, or whole or partial recirculation through the SWS system. This principle is being applied successfully to industrial and farm effluents in Europe.
In such situations, the border line between pollution and natural breakdown, with maintenance of aerobic conditions, is a narrow one. Once the former begins, it is cumulative, for anaerobic decay is slow and has noxious by-products such as H2S. Vigorous action is needed to reverse this process, but after correction, it can be maintained by the regime resulting from sub-sand abstraction with passage of oxygenated water. This principle applies equally to closed circuits such as aquaria, and in all cases, pumping should be continuous if high quality water is needed, especially in high temperatures of the Tropics, where the D.O. saturation is low and biological activity high. This problem is commented on more specifically when discussing problems at Jepara, Probolinggo, etc.
The predominantly volcanic nature of Java, with many rocks of the granite family, has several bearings on the application of SWS Units.
Large volumes of decomposed rock are carried down the rivers and sorted by hydraulic action, with the larger particles of quartz and mica dropping out first and the finest, mostly from felspar, being carried far or retained in suspension. These deposits are removed manually as they form and appear to be the major source of fine aggregates for the building industry, and for road-mending, throughout Java. An advantage of this is that sand/gravel is usually available at prices from 1,000 Ra/m3 delivered: if extra freight cost is acceptable, a range of grades may be bought. Samples of such sands were collected and these are commented on in Appendix 8. A disadvantage is that constant removal of sand, sometimes to more or less bed rock, may destroy potential in situ filter beds.
While quartz and mica are stable, and broken only by physical action, felspar is the origin of the clay forming many cultivable areas. This clay is especially typical of the lower ground, where many river beds are poor in coarse sand, which has dropped out upstream and been removed. There are exceptions, where coarse estuarine beds have a shallow covering of mud and it is always worth probing for such.
Some felspar decomposition particles are very small -of the order of 0.1 micron - and there is no suggestion that these can all be removed by SWS filtration. (In fact, only special flocculation can take them out.) Such particles never settle out and water originating in clay zones will almost always have a slight milkiness.
(7 November 1978)
The Central Hatchery for Central Kerasaan is a main supplier of several species of Carp, Puntius and Tilapia to a large area, issuing them either as small fry or, in the case of goldfish, at c. 12 cm. Annual output up to 1 m. The complex of ponds, some 17 hectares in an estate of 37 ha had larger ponds and a greater total area than average, but the water was typical of much of Java with its permanent slight milkiness due to fine clay particles. Since these ponds are largely for fry, the water generally was poorer in organic matter and algae, and the muddy bottoms less disturbed.
The need for clean water free from Lernaea was stressed by the Director, Mr. A. M. Tarigan, who had invited a dozen officers from surrounding districts for a day's workshop. After a classroom session, a site was chosen at a point where it could become a permanent supply to a series of concrete tanks. Gravity sand filters have been tried from time to time but are found difficult to maintain.
Once again, the truck bringing sand broke down but it was possible to use an old pile of large stones for the base and sand from the old filter tanks to make an artificial bed. This was pumped for c. 2 hours, until the filtrate was cleaner than the raw water and steadily improving. All types of procedure for development and maintenance was then demonstrated, after which the Unit was taken up and the whole operation was carried out again by the local officers. Although the jet test probe and stainless steel screen well are not likely to be used here, their application was briefly demonstrated.
Since the SWS Unit was not wanted in Medan, this was left at the Central Hatchery, where it was to be installed permanently.
A visit to only one major Hatchery is not enough to draw any general conclusion, but it would appear that the comments on inland ponds for Java are equally applicable here. Lernaea and other parasites cause serious casualties to the fry and each hatchery would benefit from having access to enough clean water for the early and vulnerable stages.
The largely muddy nature of the north coast, which is being aggravated by massive silting from the rivers, makes direct abstraction of sea water either difficult or, more usually, impossible in most places. Even so, as has been found in parts of Java and the Philippines, mud/silt deposits are sometimes only shallow and if clean sea water is urgently needed, it is always worth exploring down to a depth of several meters, as described in Appendix 6. With suitable pump and hose and using the yellow test probe, a large area of beach can be surveyed within a morning. This technique was shown at several places in Java, and officials should be available to carry out the search.
