The use of alternative safe water options to mitigate the arsenic problem in Bangladesh: a community perspective

APPENDIX 1

Detailed descriptions of the alternative safe water options are presented below. I wish to point out that parts of the following section have been presented in an internal BRAC report drawn up by the present author (BRAC 1999).

A. Pond-sand-filter (PSF)

Technical details

The original design of the PSF consists of a tank containing the bed of filter material and a storage chamber. Water is pumped into the PSF using a tube well head connected to a pipe, which takes in water from a pond. It then flows vertically through the sand bed. At the bottom of the tank an underdrain system (the ‘filter bottom’) is placed to support the filter bed. The bed is composed of fine sand, usually ungraded, free from clay and loam, and containing as little organic matter as possible. The filter bed normally is 1 -1.5 m thick, and the water to be treated stands to a depth of 0.3-0.5 m above the filter bed. From the base of the filter bed the water is discharged into a storage chamber.

At a later stage it was found that due to high initial loads of bacteria in the pond water the PSFs were not performing satisfactorily. It was then necessary to add a roughing filter as a form of pre-treatment before the PSF sand bed. A chamber was added onto the side of the PSF containing brick chips. The pond water flows horizontally through the brick chips, over a weir and into the main PSF chamber.

To construct a PSF with a roughing filter or sedimentation chamber costs about Taka 41,000 and takes eleven days to complete. Keeping all the necessary structures functional the size of the PSF has been reduced to bring down the construction cost from 41,000 Taka to 25,000 Taka The efficiency of this modified PSF could fulfill the demand of the locality, as was the case with the larger one.

Strength

1. There is an abundance of surface water in Bangladesh, suggesting good potential for use of Pond Sand Filters.

2. PSFs can fulfill the water demands of a large number of people. The original design was intended to serve 200 families. In reality the distance which people are prepared travel to get water limited this number to approximately 40-60 families. However, this is still a large number in comparison to other options.

3. PSFs can be constructed with locally available materials and by local masons once trained.

4. There is no chemical treatment involved and therefore no risk of adverse health effects or damage to the environment.

5. It can operate continually throughout the day and throughout the year.

Weaknesses

1. The prevalence of fish culture in rural Bangladesh is the main obstacle to PSF use. Some fish farmers use highly toxic chemicals such as andrin/dieldrin to kill predatory fish before releasing fish fry. Most put different chemical fertilisers, cow dung, mustard cake, etc. into the pond as fish feed. Clearly ponds used in this way will not be suitable for use as a drinking water source even after treatment. Furthermore, where fish culture is being practiced farmers are reluctant to give it up as it provides a valuable supplement to their income.

2. The surface water available in Bangladesh is highly turbid in both dry and wet seasons. In the dry season, there is excessive growth of algae in pond and lake water. In the wet season, rainwater drainage from the catchment area brings a lot of suspended sediment and makes the surface water highly turbid. Slow Sand Filters (SSF) do not work properly for turbidities above 30 NTU. Pond-Sand-Filters operating on the principles of SSF in Bangladesh requires frequent washing for high turbidities.

3. Due to high initial loads of bacteria in the surface water of Bangladesh PSFs will rarely succeed in removing all bacteria.

4. The initial capital costs of construction are high (approx. 40,000Tk). It is possible to reduce this to 25,000Taka If the area of the PSF is reduced.

5. Pond-Sand-Filters are not maintenance free. Eventually the sand filter becomes saturated with solids and the upper sand layer must be removed, washed with filtered water and replaced again. For this the community needs to commit itself to maintaining the PSF and some people need to be trained in the cleaning and maintenance procedures.

6. This option requires perennial ponds.

7. Masons need initially to be trained in the construction of PSF.

8. It is recommended that bathing, use of soap, washing of cattle, etc. should be discouraged in ponds used for PSFs. In reality this is difficult to achieve and would require a lot of changes in people’s behavior.

Overall assessment

The availability of large numbers of perennial water bodies seemed initially to be a strong point in favor of constructing PSFs in the project areas. However, later it was found that many of these water bodies were in use for commercial fish culture which made them unsuitable for the reason that chemicals and nutrients are used in fish farming, and the fact that people were reluctant to lose their income from fish production by reserving their pond for drinking water only.

Generally it seems that PSFs can reduce the level of bacteria in the water by 2 orders of magnitude. The level of coliform bacteria in many of the ponds of Bangladesh is so high that even after treatment by a PSF the water will not be completely bacteria-free.

The PSF users often complain that PSF water tastes and smells bad. But practically no bad smell is found in PSF water. Many people have been found using this water only for cooking purposes and very few are using this for drinking purposes.

Besides, if any technical problem arises with the PSF they lose their interest in using the option and do not come forward to repair it at their own cost. Thus, without ensuring the people’s participation in cost sharing and commitment to maintaining further construction of PSFs would not be feasible.

Overall, PSFs are a good potential source of safe water for rural Bangladesh - however, sites for construction of this technology must be chosen carefully and local people must commit to proper use and maintenance of this option. It may be that if people commit to maintaining reserve ponds for PSF use over several wet seasons they will become cleaner and provide better quality water. Another possibility is that communal ponds such as those found beside Mosques could be used for the good of the entire community.

B. Rain Water Harvester (RWH)

Technical details

The 3200 litter RWH is constructed using pre-cast concrete blocks. These are set on a concrete base in a cylindrical arrangement and bound together with wire. The inner and outer surfaces are then covered with cement. Water is channeled from collection pipes on the roof into the RWH through a funnel with a mesh filter. The RWH is covered with a lid.

The cost of constructing one RWH including the material, transportation and labor cost is about Taka 8,000 and it takes about seven days to complete the whole construction process. After completing the construction the potential users are advised to fill the tank with rainwater and store it for at least seven days. At the end of the seven days they are advised to add ½ kg of bleaching powder to disinfect the tank and to wash the inside of the tank. At the same time the caretaker of the rain water harvester is given training on the operation and maintenance of the system.

The first rainwater collected may carry significant amounts of contaminant (debris, dirt, and dust) which have accumulated on the roof and in the gutters. It is therefore recommended not to collect the first flush of rainwater. A cover for the intake was provided and users were instructed to remove this 5/10 minutes after rain started to fall. The quality of the collected water can be improved by the proper maintenance of the roof and gutters and careful cleaning at the beginning of each wet season.

Strengths

1. In some areas the potential for rainwater harvesting in Bangladesh is good - however, rainfall is variable across Bangladesh.

2. Tests of bacteriological quality of rainwater stored in RWH show it to be of consistently high quality (BRAC, 2000).

3. Materials for construction of rainwater harvesters (concrete, PVC pipe, funnels, etc.) are easily available locally.

4. Generally people report finding the taste of rainwater very pleasing. There have been some reports of the water initially tasting of cement. This will decrease over time; furthermore, if the tank is washed thoroughly after construction then sterilized, it can be eliminated.

Weaknesses

1. In the areas (Jhikargacha and Sonargaon) where the designs described above were tested rainwater harvesting could only provide safe water for part of the year. A large amount of storage would be required to cover the entire dry season. It may be possible in the future to construct several RWHs in series (one beside another), or to use a different design e.g. underground storage, which would hold enough to cover the dry season.

2. Corrugated iron sheet roofs are best for collecting rainwater although we have also tested the system on houses with clay tiled roofs. This means that construction of RWHs is not accessible to the poorest of society, those with thatched roofs.

3. At 8,000 Taka the cost is out of the reach of many rural households, particularly considering that it will only provide a partial solution.

4. Casting the concrete slabs and the lid for the 3200-litre model requires some knowledge. Local masons need training to be able to undertake these activities.

5. House roofs must be kept clean and free from dirt and debris otherwise water in the tank may be contaminated or the rain gutter may get blocked. Thus some commitment and effort is required from the caretaker.

Overall Assessment

Rainwater harvesting systems in these areas (Jhikargacha and Sonargaon) can only provide safe water for part of the year. A large amount of storage would be required to cover the entire dry season. It is thus unlikely that people would be willing to pay high costs for a partial solution.

Although construction of an RWH focuses people’s minds on rainwater it is not necessary to have a container for storage if rainwater is only to be used during the monsoon season. Any balti (water bucket), kolshi (pitcher) or other container can be used to collect rainwater from a rooftop. The Asian Arsenic Network is promoting a system in Samta village using a piece of old sari material strung from buildings or trees at the four corners with a weight in the centre and a collecting pot beneath the weight. Use of these techniques will at least provide an emergency source of safe water for part of the year.

C. SAFI FILTER

Technical details

The Safi filter comprises two concrete buckets of different sizes, one of which is placed inside the other. The upper bucket is filled with tube well water, which then flows through a permeable ‘candle’ and is collected in the lower bucket where it is stored. When needed it is drawn off with a tap.

The Safi filter candle is prepared from a chemical mixture of laterite soil, ferric oxide, manganese dioxide, aluminium hydroxide and mezo-porous silica. These materials absorb arsenic as the water passes through the candle and thus the contaminant is removed. It is also claimed that the candle eliminates pathogenic bacteria from the contaminated water. The manufacturers calculated that after two years of continuous use the candle would need to be replaced with a new one. Each new candle will cost 200 Taka.

Strengths

1. Due to the social status, which it confers, people still like to obtain their water from their own tube well. Home-based filters provide a means by which they can remove the arsenic and continue to use this bacteriologically good quality water.

2. The present cost of this filter is comparatively cheap at Taka 900. However, this is still costly for many rural households.

Weaknesses

1. Initially the Safi filter seemed to be one of the most promising options tested (BRAC, Interim Report, 1999) - however, time proved that this was not the case. By December 1999 almost half of the Safi filters distributed by BRAC in the field were experiencing mechanical problems. These ranged from serious problems such as disintegration of the filter candle to minor problems such as broken filter taps.

2. Clogging of the filter was one overall problem not related to mechanical breakdown of the filter. In many cases the flow rates from the filters became unacceptably low over time. This was also found by other organizations such as WaterAid (WaterAid, Preliminary Research Report, 2000).

Overall Assessment

The Safi filter was a promising technology originally; however, with time many problems became apparent. Professor Safiullah is currently in the process of re-designing the Safi filter to overcome these problems.

D. THREE-PITCHER OR THREE-KOLSHI METHOD

Technical Details

The Three-Kolshi filter consists of three 18-litre volume clay pitchers stacked one on top of the other in a frame. The first pitcher contains 2 kg of coarse sand with 3 kg of iron filings on top. The second pitcher contains 2 kg of coarse sand with 1 kg of charcoal on top. The third pitcher is for collecting the filtered water. Small holes are made in the bottom of the first two pitchers and a piece of synthetic cloth is placed over the holes to prevent the sand from spilling out.

Strengths

1. The Three-Kolshi filter is inexpensive. One filter can be constructed for around Taka 250. Of this 170 Taka is the cost of the metal frame, 30 Taka that of the clay pitchers and the remaining 50 Taka that of sand, iron filings and charcoal.

2. The filter is simple to construct. With a minimum of information any rural villagers should be able to obtain materials and construct such a device for themselves. In this case they may choose to construct the frame themselves with bamboo or wood.

3. The filter is made from locally available materials. Clay pitchers are available all over Bangladesh, sand and charcoal can be bought or collected easily, iron filings are produced in most small towns as a by-product of iron working (making doors, grills for windows, etc.), and the stand can either be welded together in a small town or made from wood or bamboo.

4. The Three-Kolshi system is based on indigenous technology and uses materials familiar to the rural people; besides social acceptance is high.

5. Continued use of tube well water means less danger of increased diarrhoeal disease provided construction materials are sterile.

Weaknesses

1. There is a potential problem of clogging with iron, particularly if the filter is allowed to dry out between uses. Initial observations show that the iron filings may bond together into a solid mass over time making the materials difficult to replace or clean, and also increasing the potential for channeling of water and reduced removal of arsenic.

2. The system produces waste materials, which contain arsenic. It may be possible to convert the solid arsenic-laden iron oxide produced into iron, or to dispose of the material in bricks or in road construction.

3. There is a waiting period between pouring the water and receiving clean water. Whether this is acceptable and people continue to use the system over time has not been assessed.

4. The amount of water produced is small and can only be used for basic drinking and cooking purposes. It would not be enough for other uses such as consumption by livestock.

5. As the Three-Kolshis are stacked on top of each other and not protected from the open air there exists a chance of water contamination by bacteria in the atmosphere.

Overall assessment

The Three Kolshi or Three Pitcher system has enormous potential to provide an emergency drinking water source for the arsenic-affected areas of rural Bangladesh. It is based on an indigenous technology, is inexpensive and can be constructed with locally available materials. Social acceptance is high.

There are a number of unanswered questions about the Three-Kolshi filter, which need to be addressed before it is taken up on a larger scale.

Firstly it was reported that bacteriological contamination of the water is occurring before or during filtration. If this is during filtration then all filter materials must be sterilized before manufacture of the filter. It may be possible to do this by heating under the sun or boiling in water.

It has not been proven beyond doubt that water from the Three-Kolshi filters is completely free of chemical impurities. The potential for trace elements such as lead, chromium, zinc, tin, etc. to enter the water from the iron filings has not yet been conclusively discounted.

The final question about the Three-Kolshi filter is whether it is technically effective in the long term. It must be assessed after how long the filter becomes clogged with iron and must be cleaned and a cleaning or regeneration process designed. There are some initial reports that the layer of iron filings forms into a hard mass over time. This must be investigated and the potential for channeling of water past the iron filings, and thus leakage of arsenic through the system, determined.

E. Assessment of the effectiveness of the safe water options

A number of alternative safe water options are now in operation as demonstration units. The idea behind constructing these demonstration units is to raise the awareness level of the community people about this alternative safe water option which would later help in developing a system of involving community members in choosing, financing, and implementing safe water systems on their own.

Alternative safe water options which have been implemented in the field and assessed for this research project are: Pond Sand Filters (PSF), Rainwater Harvesters (RWH), three kolshi filters, and Safi filters. These have been assessed with reference to initial and running costs, ease of implementation, requirement for maintenance or ongoing supervision, provision of an intermittent or continuous supply, susceptibility to bacteriological contamination, and acceptability to the local community.

The matrix below shows ratings of each of these factors rated on a scale of 1 to 5. The maximum possible is 45 and a higher rating is better.

It can be seen from this that all of the options have their limitations. At present the Three-Kolshi filter is proving to be the best option for its ease of use, low cost and simplicity.

Initial reports implying that the layer of iron filings becomes a solid mass over time must be investigated further to ensure that the filter remains effective. If this is the case, the method of regeneration and the effectiveness of this process must be determined. The potential for leaching of trace metals from the iron filings is also uncertain. Finally, it may be necessary to sterilize the filter materials before use to avoid secondary contamination of the water with bacteria during filtration.

The RWHs and the PSF are both thought to be too costly to be taken up locally.

The Safi filter initially seemed to be a promising technology for treatment of arsenic-contaminated groundwater; however, over time the majority of the filters supplied ceased to be effective. Professor Safiullah is currently working to solve the problems of the Safi filter and to produce an alternative porous media column filter. If proven to be effective these filters may be useful in the future.