By Thomas P. Warwick
Summary: When the goal is bringing potable water to developing countries, a point-of-use device’s technology is only a small part of the solution. Through its own trials, one organization has found that a good deal of monitoring must also take place to better the odds of effective results.
Children today in the rural areas of the Caribbean Basin continue to suffer terribly because of waterborne disease—the biggest killer of those age five and under. Recently, point-of-use (POU) water treatment devices have regained popularity in humanitarian efforts to solve contaminated water problems in developing countries. The drinking water treatment units (DWTUs) range from ceramic, two-bucket filters to simple containers using only chlorination to disinfect the water. The low cost of the water treatment systems has resulted in a proliferation of programs, many times without proper evaluation of whether or not the program makes a difference. This article explores both the necessary technologies behind developing country DWTU needs and the required monitoring to ensure that the program is effective.
The goals of a developing country’s DWTU program are simple—save the lives of children younger than 5 years of age, improve the health of older children and adults, and prevent infectious diseases. Therefore, the objectives of a POU technology design for a developing country are:
- The systems must be easy to use and repair,
- They should be inexpensive,
- They should be rugged,
- They should keep stored water safe,
- They should provide good tasting water filtered in two hours or less, and
- They should be effective against key sources of contamination (i.e., protozoa, bacteria and common pollutants).
In evaluating any POU water treatment system, one must give very careful consideration to its effectiveness. For young children, protozoan and bacterial infections are the greatest concerns. Much attention is rightfully given to bacterial infection, but protozoan infection shouldn’t be overlooked. About 200 million cases of giardiasis (acute infection with Giardia) are reported every year.1 Giardia is also highly infectious—one cyst has a 2 percent probability of causing giardiasis.4
The design objectives seem mutually exclusive at times. For example, the oldest style of a two-bucket system (see Figure 1) was invented by the British Army in India during the 1850s. Today, it uses a porous ceramic filter with a hypochlorite solution in the storage bucket for disinfection.
The ceramic filters are effective against protozoan infection, and the chlorine source (when properly implemented) can kill most bacteria. The ceramic filters, however, process water very slowly—often 24 hours for five gallons. Increasing the number of ceramic cylinders makes the replacement cost of the system unaffordable. A further consideration is the fragile nature of the ceramic cylinders, which requires frequent scrubbing to abrade them to refresh filtering capabilities. The author’s experience with the system is that people don’t use them because of the slow filtration.
The other extreme of DWTU technology uses a hypochlorite solution (via a hypochlorite generator) and no filters. This approach, however, doesn’t meet the goal of killing protozoa consistently. Chlorine levels to kill Giardia, for example, require a Ct (contact time) × concentration of disinfectant of 65-150 milligrams per minute per liter of water (mg-min/L).6 One gallon of water sitting for 30 minutes, for example, would require a minimum free chlorine (Cl) concentration of 2.17 parts per million (ppm) or 4.84 mg/L of sodium hypochlorite (NaOCl) to meet the 65 mg-min/L required. Chemical/mineral chlorine demand would further increase hypochlorite levels to be greater than 6.5 or 7 mg/L.2 The water would simply be unpalatable.
The smallest protozoan cysts of concern are Giardia lamblia cysts—7-12 microns (µm)—and Cryptosporidium parvum (3-5 µm).1 Newer technologies, such as low cost string-wound filters, with 1-5-µm ratings and fast flow rates, were thought to offer some hope. Recent testing at the Massachusetts Institute of Technology, however, suggests that the string-wound filter may not be as effective as their micron rating suggests. Figure 2 shows the ratio of effluent to influent for water containing suspended 5.7-µm test beads with similar density and electrostatic properties to a cyst. The influent passed through a 5-µm “nominal” sediment filter that, according to the manufacturer, should trap a minimum of 85 percent of all 5-µm particles. On average, the 5-µm nominal filter trapped 20 percent of the 5.7-µm cysts. This is a far cry from the World Health Organization and U.S. Environmental Protection Agency (USEPA) goals of zero cysts per liter.7,8
A system developed by Gift of Water, a charitable organization launched in 1995 and based in Brevard County, Fla., uses a derivative of the two-bucket water treatment system (see Figure 3). The organization’s approach has been to over-chlorinate the water in the top bucket and then filter the chlorine and any disinfection by-products (DBPs), such as trihalomethanes (THMs), that may be formed when water contains certain organic constituents such as remnants of decayed leaves via a granular activated carbon (GAC) filter. Funding for projects is accrued through three main sources—foundations and corporations, private donors, and churches.
A 5-µm nominal sediment filter is used that usually keeps the GAC from plugging. Yet, the GAC hasn’t been tested for sediment effectivity. Preliminary results from a Massachusetts Institute of Technology (MIT) study suggests it’s more effective than the sediment filter1; however, it has been tested for adsorptive properties.3
The top bucket detaches, which allows for the critical contact time needed for bacterial and protozoan disinfection/removal (Ct = 290 mg-min/L, on average.) A user places a small amount of chlorine (.693 mg/l of NaOCl or .31 ppm Cl) in the bottom bucket for residual disinfection to prevent bacterial regrowth. As long as the GAC filter is changed on a regular basis, THM concerns are minimized3. The finished water exceeds the 4 log mark for Giardia removal6. Meanwhile, the search for a fast-flow, 1-3-µm filter continues.
Two-bucket systems are, for the most part, manufactured in the project country while some parts arrive from the United States. The system flows through within 30-45 minutes at +/-2.5 sigma. Combined with the contact period, the total usage time is about 1 to 1.5 hour.
Just because a technology—no matter how easy to use—is put into place doesn’t mean it will be used correctly. The recipients of the POU water treatment system must value the clean water and be educated in its use as well as water hygiene. Monitoring provides the feedback necessary for continuous improvement of the program.
When Gift of Water first implemented a program in the community of Barassa, Haiti, for example, the results immediately following family training were 74 percent correct use, i.e., the DWTUs were producing good water. Three months later, however, the results dropped to 30 percent. Without monitoring, this would have never been caught. The organization took corrective action. Today, Barassa families enjoy clean water, and test results are typically greater than 95 percent effective at producing good water, which is defined as the top bucket meets or exceeds the Ct requirement and the bottom bucket has no more than .2 ppm Cl.
The monitoring schedule the organization attempts to follow is:
For understanding chlorine demand, see Table 1.
For monitoring the use of the purifier, see Table 2.
Gift of Water trains local people to be “community technicians.” Technicians are responsible for family education, DWTU maintenance, and frequent presence/absence (P/A) monitoring for chlorine levels. A simple chlorine test is implemented by using a colorimeter. The technicians are the heartbeat of community success. For more complex monitoring, the program recruits, trains and sends volunteer teams to the field and works with different universities to ensure effective usage of the systems.
The final aspect of monitoring is measuring health improvement. The most reliable measure in a health survey is the incidence of diarrhea during the prior month. Families in the two-bucket DWTU program must be measured against families not in the program (e.g. a differential study). Further, other influential factors, such as the relative wealth of a family, education levels of the parents, toilet facilities, hygiene, etc., must be taken into account statistically. Finally, the study must correlate health improvement to the correct use of the purifier.
Last January, the program in Dumay, Haiti, underwent an exhaustive, independent study conducted by MIT. Gift of
Water learned that the purifiers accounted for a 50 percent reduction in diarrheal incidence for children under the age of 5 while accounting for other influences, such as relative wealth and education.5 The program is now also being pursued in Jamaica and Honduras.
Based on technology, monitoring and effective results, point-of-use water treatment devices can be successful in preventing diarrheal disease in children. The technology, however, must address all forms of contamination, especially potential protozoan and bacterial constituents in drinking water. Ceramic filter/chlorination drinking water treatment units and the heavy chlorination/activated carbon DWTUs meet these criteria. The latter is particularly effective at significantly improving the quality of drinking water. It also produces it at a faster, more practical rate for family use while reducing incidence of waterborne disease.
Technology, however, isn’t enough in a program. Monitoring is critical to continuously improve and reach the ultimate goal of 100 percent clean water for children in developing countries such as those found in the Caribbean Basin.
The author would like to thank Susan Murcott and Daniele Lantagne, of the Massachusetts Institute of Technology, for their assistance on preparing this article; and the author’s wife, Barbara, for editing it.
- Borucke, M.J., Filtration of Giardia Cysts from Haitian Drinking Water, Massachusetts Institute of Technology, Cambridge, Mass., June 2002.
- Elice, S.J., Chlorine Demand in Haitian Water Supplies, Massachusetts Institute of Technology, Cambridge, Mass., June 2002.
- Lantagne, D., Evaluation of Trihalomethane Generation and Removal from Point-of-Use Purifiers, Massachusetts Institute of Technology, Cambridge, Mass., May 2001.
- Rose, J.B., J.T. Lisle and C.N. Haas, “Risk Assessment Methods or Cryptosporidium and Giardia in Contaminated Water,” Protozoan Parasites and Water, Cambridge: The Royal Society of Chemistry, pp. 238-242, 1995.
- Varghese, A., Point-Of-Use Water Treatment Systems In Rural Haiti: Human Health And Water Quality Impact Assessment, Massachusetts Institute of Technology, Cambridge, Mass., May 2002.
- Vressman, W., and M.J. Hammer, Water Supply and Pollution Control, Harper Collins, New York, 1993.
- U.S. Environmental Protection Agency, “Interim Enhanced Surface Water Treatment Rule,” Federal Register, USEPA Office of Ground Water and Drinking Water, December 1998, website: www.ega.gov/safewater/mdbp/ieswtrfr.html
- World Health Organization, Water and Sanitation: Guidelines for Drinking Water Quality, WHO, Geneva, 1993, website: www.who.int/water_sanitation_health/GDWQ/Microbiology GWDWQMicro-biologal2.html
About the author
Thomas P. (Phil) Warwick, a Purdue University engineering graduate, started traveling to developing countries as a volunteer in 1991. Experiences there motivated him and his wife to start Gift of Water, a tax-exempt organization dedicated to bringing clean water to children, in 1995. Gift of Water, of Melbourne, Fla., produces and distributes the Gift of Water® two-bucket chlorination/granular activated carbon drinking water treatment system. Warwick and his wife, Barbara, live in Melbourne with their three daughters. He can be reached via email: firstname.lastname@example.org or website: www.giftofwater.org