Article
Roof water collection systems in some Southeast Asian countries: status and water quality levels A Dr
Appan
Adhityan Appan
BSc, BE (Hons), DIC, MSC, PhD, FICE, FCIWEM, FIES, CEng,
EurIng School of Civil
&
Structural
Engineering Nanyang Technological University Nanyang Avenue Singapore 639798
Key words Contamination; roof water collection; Southeast Asia;
water
quality
Abstract There is extensive use of simple and inexpensive roof water collection systems in Southeast Asian countries. The collected runoff which flows off different types of roofs is affected not only by the inherent quality of the roofing material but also by the contamination of roofs by rodents, birds, etc. Consequently, bacteriological quality levels are excessive though the collected rainwater is still used extensively for potable
some
purposes. From samples collected in most locations, there were positive Total and Faecal Coliform counts though, in terms of physico-chemical parameters, roof water appears to be of a higher quality. Various causes have been attributed to the frequent presence of Faecal Coliform but, mostly, pollution is of animal origin as the Faecal Coliform/Faecal Streptococci ratio is less than unity. It is proposed that the collected roof water be boiled, disinfected with household bleach or be subjected to radiation from sunlight which appears to have good potential to be an effective bactericide. It is recommended that simple testing methods be developed and health education imparted to the various aspects of utilising roof water collection systems.
Introduction Roof water collection
its origin to as long ago as 2000 BC in the Middle East (Reid, 1982). In essence, it involves the tapping of all the rainwater running off roofs of structures, transporting the runoff through gutters to a storage tank and using the water primarily for domestic purposes. The primary input in such roof water collection systems (RWCS) is rainfall. In the past the quality of rainfall had not always
been
can trace
until about a decade ago noted that acid rain was
questioned
when it
was
becoming
a
reality
in in
some
developed developing
countries. However, countries rainwater quality appears to be one of the purest in nature though the quality of the water which has been most
collected off roofs can be affected detrimentally. The main objectives of this paper are to briefly discuss the existing water quality, draw conclusions and make some recommendations for maintaining good RWCS practice.
Status of RWCS in Southeast Asia The types of RWCS being adopted at present are largely dependent on the location of the country, the available rainfall, the traditional methods of using rainwater and a host of other socio-cultural and economic
factors. Thailand Thailand has annum.
rainfall of 1,150 mm per Clay jars have been used a
319
traditionally
to
collect rainwater
as
Table 1
bamboo reinforced tanks, reinforced concrete tanks and quite a few ferro-cement tanks have been built in the Khon Kaen area which has a much lower rainfall (Vadhanavikkit, 1983). In most of these systems, the collected water is not disinfected. One of the few precautions taken to safeguard the quality of water appears to be diversion of first flushes. well
as
WHO Guidelines for
Drinking Water (1984)
Indonesia The collection of rainwater for domestic purposes has been prevalent for several centuries. A systematic approach to building cisterns was pursued in 1978 using different materials such as brick, bamboo, and
finally ferro-cement. The RWCS were containers in which roof water was collected but the water had no specific treatment or long-term maintenance and operation programmes. In some collection
tanks, fish were being reared mosquito breeding.
to
prevent
T’he
Philippines
Rainwater collection has been practised for a long time particularly in the rural areas using receptacles of
brick,
cement,
earth, heavy plastic
or
I
E
*
Note: All
units, except pH,
in
galvanised iron. The first attempt at systematically propagating 30 units of RWCS was undertaken in the Province of Capiz. Construction of such tanks was progressively extended to a further 540 units in the same region (Lee and Appan, 1991).
Singapore
affected
largely by quantity and quality of rainfall, roofing material,
There is abundant rainfall of about 2,500 mm and the country is wellsupplied with drinking water although 40% of the water is
type and condition of gutters, type and location of storage containers and mode of collection. Collected roof water is mainly used for potable
from Malaysia. Feasibility studies have been carried out in urban high-rise buildings for dual-mode supply systems using rainwater for
purposes.
(Appan et al, 1987). In addition, existing systems using rainwater exist for washing of quarry sand (Appan, 1991) and for fire-
As most of these countries do not have well-defined standards for drinking water, the World Health Organization (WHO) Guideline (1984) values have been used (see Table 1) for comparison purposes. Bacteriological parameters such as Faecal Coliform (F Coli) and Total
imported
Malaysia Having an abundant annual rainfall of 2,420 mm and relatively good conventional large-scale water supply systems, the activity on RWCS has been very limited. However, in rural areas water has been collected in jars, tubs, etc from very early times. Experimentation on the use of RWCS commenced in East Malaysia in 19711 and by 1982 about 2,700 tanks were built in Sarawak and Sabah and a few tanks constructed in Perlis and Kedah in West Malaysia (Appan, 1983).
320
unless stated otherwise
mg/L
non-potable
uses
fighting and toilet flushing, airport (Appan et al, 1995).
etc
in
an
Collected roof water quality Major factors affecting water quality The water quality in cisterns will be
Criteria
for comparing
water
quality
Coliform (T Coli) well
as
were
Faecal
Streptococci (FC/FS)
determined
as
Coliform/Faecal ratios have been
Table 2
Roof water
quality - Thailand (1)
determined. These FC/FS ratios have been shown (Duktar, 1977) to indicate whether the sources of pollution are of human (FC/FS> 1 to 4) or animal (FC/FS < 1 ) origin. Water quality in Southeast Asian RWCS
Thailand In
a
study covering the whole country,
1,292 samples
were
collected
over a
period of four years mainly from clay jars which are used in abundance (Nantana, 1987). The results
*
Note: Values are % of total
samples
Table 3
Roof water
quality - Thailand (1I)
Table 4
Roof water
quality (averages) - Malaysia
are
shown in Table 2. In terms of physicochemical parameters, more than 83% of the samples were satisfactory except for about 40% of the samples exceeding the allowable limits of lead. In terms of Coli, more than about 76% of the samples had values exceeding the WHO Guideline standards. In another series of bacteriological tests conducted in three locations in Khon Kaen (Wanpen, 1992), 709 water samples were collected from tiled roofs and gutters, containers located in homes, jars, and the point of consumption. A summary of the results is shown in Table 3. In terms of the three Coli groups, only 10% to 67% of the samples were within the WHO Guideline values. Samples other than those collected from the container showed that, due to an FC/FS ratio of less than one, 79% to 82% of the contamination could have emanated from animal droppings. In the same study, only manganese (in 2% to 20% of the samples) and zinc (in 4% to 26% of samples) did not meet the Guideline levels.
Indonesia
Although RWCS seem to have been practiced with extensive experimentation using different types of materials, there is practically no available information on the quality
*
Note: Parameter
measurements are
mg/L unless otherwise stated
of collected water. It has been observed that some measures such as fish rearing within the tanks have been practised so as to keep the water clean.
321
Roof water
Table 5
of T Coli and F Coli
counts. In the FC/FS ratios indicated that the source of the pollution was largely of animal origin such as the droppings of birds, rodents, etc. The few cases of human contamination can be attributed to poor handling practices. There is evidence that bacteriological contamination terms
addition,
quality - Singapore
(salmonella aechevalata), presumably from animal
droppings, has led to gastrointestinal problems (Koplan et al, 1976). It has also been well-established that animals and birds, both domestic and wild, are frequent carriers of salmonella (Quevado et al, 1973; Osborne, 1976). The presence of different species of salmonella in roof water has also been reported by Philippines Discussion and conclusions many researchers (Wanpen, 1987; In a study spread over a period of one and Fuj ioka et al, 1991). Hence, with the Sampling testing year in three villages in the Province It is evident that available data on frequent presence of F Coli and of Capiz involving 25 ferro-cement water collected from cisterns in the particularly salmonella, there is the T’he
tanks,
it
shown that
not less than T Coli had samples the WHO Guideline values
24% of the
region are grossly inadequate. There is no systematic collection and testing
exceeding
which is
was
(Personal Communication, 1986). z
Malaysia The quality
of rainwater and roof runoff has been monitored and 72 samples were collected from two types of roofs in West Malaysia (Yaziz et al, 1989) (see Table 4). The range of turbidity, lead and F Coli values far exceeded the WHO Guideline values. The pH value of rainwater also has a tendency to lie towards the lower range of the Guideline values.
Singapore Roof
monitored from a high-rise building in the Nanyang Technological University for six years from May 1989 and the range of mean values is shown in Table 5. The values appear to be acceptable in all the water
was
physio-chemical parameters except pH which is quite low. T Coli and F Coli values also exceed the Guideline values. Earlier field investigations have also shown that, during January 1974 to July 1983, the range of pH in 11 monitoring stations distributed throughout Singapore was 4.8 to 5.5
integral part of good water quality management practice. It is suggested that wherever RWCS have been established, or are being planned to be established, the sampling programme should be an integral part of the whole project and water quality an
should be monitored basis.
on
a
routine
Physico-chemical parameters Generally, collected water appears to exhibit quality levels that are comparable to the WHO Guideline values and it is understandable as to why, in most cases, rural folk prefer rainwater over river water for drinking purposes. The pH range is on the acidic side in Malaysia and more so in Singapore where it could reflect the level of industrial activity at the sampling locations. In both Thailand and Malaysia the high lead values can be attributed to the roofing material and hence care should be taken to avoid the use of unsuitable roofing material.
Bacteriological parameters There were quite a few samples which
’
(Tan, 1984). 322
,
exceeded WHO Guideline values in
need
to research its impact on human systems. As infectious doses of most salmonella species have to be greater than 1,000 (Fujioka, 1993), there is the need to develop simple and inexpensive methods of identifying salmonella during routine sampling.
Disinfection Currently it
appears that quality control is limited in very few cases to diverting of first flushes and rearing of fish within the container. Boiling, despite its limitations, is the easiest and surest way to achieve disinfection, but there is a reluctance to accept this practice as taste is affected. Alternatively, simple methods of adding any one of the halogen compounds could be practiced. Chlorine, in the form of household bleach, has been successfully applied to collected roof water, retaining a residual level of 0.2 mg/L (Krishna, 1991). Such applications would call for action by the individual owners of RWCS in which case the periodic addition of household bleach could be recommended. The cheapest UV system could still be prohibitive in many developing countries. It is also recommended that, as most of the countries in the region have an
abundant amount of sunlight, research be undertaken to study the possibility of using solar radiation as T Coli, F Coli and Faecal Streptococci have effectively been removed when exposed to sunlight in Honolulu (Fujioka and Chinn, 19 8 7 ) and Thailand (Wanpen,
1992). Health education A basic difference between conventional large water supply systems and RWCS is the fact that the onus of treatment and disinfection is transferred from the central water authority to the individual householder
(Appan, 1984).
erable effort has the individual
fection.
to
Hence consid-
be put
to
to
convey the need for disinmethods should be
users
Simple proposed, implementation
overseen
and results monitored. Roof water which is generally of a higher order quality level than that of surface runoff is readily imbibed although it is apparent that bacteriological contamination could render it below the allowable level for drinking. Such information should be systematically conveyed to the users and good collection practices introduced. The roofs, gutters, pipes and cisterns should be frequently inspected. All these measures should form the basis of an integrated health education programme that should go hand in hand with the RWCS building
project. Water
quality study Currently most of the information on water quality appears to be carried out on an ad hoc basis, data is limited and the results not applicable to all locations. A systematic study should be undertaken on a national or regionalI basiss taking into consideration rainfall patterns, types of roofing materials, rain and roof water
qualities, guttering
systems, materials used for storage tanks, maintenance practices, sequence and
frequency of collection, testing schedules and any other salient factors
specific
to
the country
or
region.
Standards for roof water quality With quality levels of drinking water being progressively upgraded and made more stringent in developing countries, it would be prudent to lower the acceptable level of bacteriological quality of roof water, unless it can be established that roof water that is contaminated only with animal faeces only can adhere to a different set of standardss based on research as outlined earlier.
J Hyg Camb; 81,
KRISHNA H (1991). Improving cistern
303water
quality. Proceedings of the 5th International Conference on RWCS, Keelung, Taiwan, 4-10 August LEE K W and APPAN A (1991). Proposal for disseminating ferro-cement rainwater tank technology in the Province of Capiz, The
Philippines. Proceedings of the 5th International on RWCS, Keelung, Taiwan, 4-10 August
Conference
NANTANA, water
quality
SANTATIWUT (1987). Rain in Thailand (supplementary
paper). Proceedings of the 5th International Conference on RWCS, Khon Kaen, Thailand,
References APPAN A (1983). Design and development aspects of rainwater cistern systems in South East Asia. Proceedings of Regional Seminar and Workshop, Khon Kaen, Thailand, 29 Nov - 3 Dec APPAN A (1984).
Existing rain water catchmethodologies in Southeast Asia and their future development. Proceedings of the 2nd International Conference on RWCS, US Virgin Islands, June APPAN A (1991). Unit process operations in quarry sand washing using rainwater and secondary sewage effluent. Proceedings of the International Conference and Exhibition on Environment, Enviro Asia ’91, Singapore, 7-10 ment
November APPAN A, LIM K L and LOH S K (1987). The utilisation of high-rise building rooftops for development of a dual-mode water supply in Singapore. Proceedings of the 3rd International Conference on RWCS, Khon Kaen, Thailand, 1416 January APPAN A, JAYAPRAKASAM T and PUNITHAN S (1995). A total approach towards the design of RWCS in airports subjected to tidal effects. Proceedings of the 7th International Rainwater Catchment Systems
Conference, Beijing, 21-25 June DUKTAR A P (1977). Escherichia Coli: The Faecal Coliform. In: Baterial Indicators/Health Hazards Associated with Water, pp 48-58 (A W Hoadley and B J Duktar, eds). Philadelphia: American Society for Testing Materials R S (1993). Guidelines and standards for cistern waters. Proceedings of the 6th International Conference on RWCS, Nairobi, Kenya, 1-6 August
FUJIOKA microbial
FUJIOKA R S and CHINN R S (1987). The microbiological quality of cistern waters in Hawaii. Proceedings of the 3rd International Conference on RWCS, Khon Kaen, Thailand, 1416 January FUJIOKA R S, INSERRA S G and CHINN R D (1991). The bacterial content of cistern waters in the Tantalus area of Honolulu, Hawaii. Proceedings of the 5th International
Conference on RWCS, Keelung, Taiwan, 4-10 August KOPLAN J P, DEEN R D, SWANTON W H and TOTA B (1976). Contaminated roofcollected rainwater
outbreak of Salmonellosis. 309
as a
possible
cause
of
14th-16th January OSBORNE A D (1976). Salmonella infections in animals and birds. J Roy Soc Health; 106(1), 30-33 PERSONAL COMMUNICATION (1986). Letter from HE Governor of Capiz, The Philippines on 14th March to Dr A Appan, Consultant to International Research Center, Canada
Development
LORD R D, DOBOSCH D, GRANIER I and MICHANIE S C (1973). Isolation of salmonella from sparrows captured in horse corrals. Am J Trop Med & Hyg; 22,
QUEVADO F,
672-674 REID (1982). Lessons of history in the design and acceptance of rainwater cistern systems. Proceedings of the International Conference on
RWCS, Honolulu, Hawaii, June TAN G C I (1984). Study of rainfall acidity in Singapore. Singapore: Dept of Geography, National University of Singapore VADHANAVIKKIT C (1983). Ferrocement water tank. Proceedings of Regional Seminar and Workshop, KHon Kaen, Thailand, 29 Nov - 3 Dec
WANPEN W (1987). Rainwater contamination. Proceedings of the 3rd International Conference on RWCS, Khon Kaen,
Thailand, 14-16 January WANPEN W (1992). Rainwater contamination and control. Proceedings of the 1992 Regional Conference of the International Rainwater Catchment Systems Association, Kyoto, Japan, 4-10 October WANPEN W, MUNG KARNDEE P and HOVITCITR P (1989). Evaluation of rainwater quality: Heavy metals and pathogens. Proceedings of the 3rd International Conference on RWCS, Khon Kaen, Thailand, 14-
16 January WORLD HEALTH ORGANIZATION (1984). Guidelines for drinking water quality. Vol 1 Recommendations. Geneva: World Health
Organization YAZIZ M T, GUNTING H, SAPARNI N and GHAZALI A W (1989). Variations in rainwater quality from roof catchments. Water Res;
23(6), 761-765
an
323