By: Dr. Dennis Lye, USEPA


Scientific studies worldwide are reviewed concerning both the collection of rainwater from rooftops and the quality of the water delivered from these systems. Recent reports reveal a better understanding of this water source and the specific areas of research needed. Throughout the world, rainwater runoff collected in urban areas has been documented to contain substantial amounts of containments. Depending upon the intended use of the rainwater, evidence is accumulating that various treatments and disinfections will be required prior to consumer usage. Proposed standards and guidelines regarding this type of water source vary widely worldwide, especially for microbial contamination. There is still a lack of clarity regarding water quality guidelines and health related standards for this type of water source.  Properly designed and maintained rainwater collection systems will be received more favorably in the future as a supplement to existing water supplies because of a number of benefits including reductions in demand on community water supplies/infrastructure costs, more effective management of storm water runoff, and increased restoration of underground reservoirs through controlled infiltration.

Keywords:  Rainwater, Roof Runoff, Water quality

1.1   Introduction:

The world is facing escalating demands for good quality water as current usage from surface and ground is outstripping supply.  Even in those areas of the world that appear to have adequate water supplies, there are constant needs to balance existing supplies with ever growing demands.  Human urbanization has dramatically increased the amount of impermeable areas which contribute to runoff flooding, runoff contamination and storm water management problems.  Cycles of droughts bring into sharp contrast the need to conserve, protect and supplement existing water supplies.  The collection and storage of rainwater to supplement existing water supplies might be used to address some of these problems.   Rainwater utilization may be one of the best available methods for recovering natural hydrological cycles and aiding in sustainable urban development (Kim et al. 2005a).

Rainwater collection systems can be categorized into two general types; 1) systems using surface/ground catchment areas and  2) those using above-ground rooftop catchment areas.  An important distinction is the intended use of the rainwater. Water intended for ingestion (potable water) requires much more stringent guidelines for the levels of allowable contamination than water intended for non-potable uses. A recent study by Signor et al. (2007) addresses what is known about microbial contamination risks associated with rainfall-induced surface/ground runoff. This review discusses only the above-ground rooftop catchment types of rainwater collection systems.

Numerous rooftop runoff rainwater collection systems are now being constructed around the world for non-potable uses such as toilet flushing and clothes washing. Countries such as Germany, Denmark, India, Japan, and Australia are leaders in the installation of systems for collection of rainwater from rooftops with storage areas and distribution lines within individual households. (Albrechtsen, 2002). The national legislative bodies in many of the above countries are now drafting legislation requiring all new construction to have rainwater harvesting systems for the purposes of flushing toilets and external water uses. The purpose of this legislation is twofold: 1) to reduce demand for treated water and the expansion of the water supply infrastructure; and 2) to collect and use rainwater instead of surcharging storm water management systems.

Despite the advantages of this historical water source, the widespread application of rainwater collection systems has been impeded by a lack of knowledge both within the general public and within local governing agencies concerning the chemical and microbiological quality of this type of water. In the absence of scientific literature to the contrary, rainwater collected from rooftops had been assumed to be a relatively safe source of water. Only in the last twenty years have scientific studies begun addressing the specific risks inherent to this water source in both highly developed and newly developing countries.

Studies have linked particular types of rainwater collection systems and the maintenance of these systems (or lack thereof) to a variety of potential human infections and chemical intoxications. This article reviews current scientific literature addressing the effects that rooftop collections have on 1) rainwater and its subsequent runoff quality, 2) the concerns of water quality issues, and 3) perceived health risks associated with this source of water.

2.1  Chemical contamination

The report by Chang et al. (2004) provides a good discussion concerning how rainfall onto rooftops may become contaminated.

▪     Airborne pollutants may be captured by the falling rain.

▪     Chemical compounds in the roofing material may leach into the acidic rainwater.

▪     Organic substances deposited on the rooftop (plant matter, insect matter, waste from living organisms such as birds) may be absorbed by the rainwater.

▪     Rooftops exposed to direct sunlight will exhibit significantly higher surface temperatures than surrounding materials.

▪     High rooftop temperatures may accelerate the chemical reactions and organic decompositions of materials deposited on the collection area.

Their results of how different types of roof materials contribute to rainwater runoff contamination are given in Table One (at the end of article).

Polkowska et al. (2002) analyzed volatile organohalogen compounds, petroleum hydrocarbons, various ions, as well as various pesticides in rooftop runoff from buildings along major transportation routes in the city of Gdansk, Poland. More than 50% of the samples were found to have elevated levels of these contaminants.

Simmons et al. (2001) investigated one-hundred and twenty-five domestic rooftop rainwater systems in four rural Auckland districts for levels of heavy metals (zinc, copper, and lead). Their study found 14% of the systems exceeded New Zealand levels for lead in drinking water, 2% for copper levels and 1.0% exceeded zinc guidelines.

The study by Magyar et al. (2007) of the chemical composition of water and sediment in urban rainwater tanks in Melbourne, Australia revealed high metal concentrations in sediments that could be potential sources of pollution. Their preliminary results suggested that carefully designed tanks had a major influence on water quality delivered to the end user. Tank design solutions included mechanisms to reduce sediment resuspension, facilitating removal of sediment, and regular specialized service for tank maintenance.

Melidis et al. (2007) reported on the chemical characteristics of rainwater in the city of Xanthi, Greece. Ten sites within the city were sampled representing distinct residential densities, road traffic volumes and industrial activities. A total of 130 samples taken over two years revealed that chemical pollutant levels were generally higher in roof drainage samples than those found in the rainwater itself, but both were lower than current Greece drinking water guidelines.

Berndtsson et al. (2006) reported on the chemical composition of runoff from vegetated rooftops, a type of rooftop now experiencing a renaissance as a new trend in urban design. Although there are no standards for the water quality released from this type of rooftop runoff, they monitored vegetated rooftops in four different regions of Sweden for metals and other nutrients.   Their results suggest that vegetated rooftops behave as a source of contaminants. Although the level of contaminants from these rooftops was lower in concentration than those found in normal urban runoff, the levels were high enough to still be classified as moderately polluted. They suggested that maintenance of these rooftops greatly affected the water quality of runoff and that easily dissolvable fertilizers were to be avoided. Kim et al. (2005a) also reported on newly developed technologies that reduce contaminants from “green roof systems” and decrease the impact of runoff from these systems on local water environments.

3.1  Microbial contamination

A clear consensus on the quality and health risks associated with rooftop collection systems depends on the use of the water and the maintenance of the systems. Previous reviews have reported that roof-harvested and tank-stored rainwater was of acceptable quality for drinking and cooking purposes (Dillaha and Zolan, 1985; Heyworth, 2006). Since 1978, there have been only six disease outbreaks worldwide documented to be associated with rainwater consumption (Heyworth, 2006). In contrast, a number of reports have also indentified the common occurrence of various pathogenic microorganisms in samples taken from rainwater systems (Lye, 2002).

Standards for microbiological contamination levels of various water types vary greatly throughout the world. This is in stark contrast to the recognition for levels of chemical contamination which have reached almost universal agreement worldwide. Methods of chemical analyses, chemical detection limits, and chemical water quality standards are similar for all parts of the world. For microbiological standards however, there is a lack of global agreement regarding water quality guidelines and health related standards. Table Two presents data about widely varying standards from one country to another and a general lack of clarity regarding water recycling schemes used for toilet flushing (Birks et al, 2004).

Microbiological water quality standards range from relatively simple to relatively complex. If the water is intended for human contact or consumption, it must be treated to the fully potable standards required by the individual country.  However, there are often exceptions where a lesser standard for human contact is used rather than for human consumption. For example, the water quality standards required for swimming in environmental waters are deemed adequate for water used for bathing, showering and laundry purposes within a household. In more complex water reuse systems, a variety of quality standards and recommended treatments are applied for 1) human consumption/contact, 2) for non-contact situations, and for 3) ecological uses such as irrigation waters, etc.

Simmons et al. (2001) investigated one-hundred and twenty-five domestic rooftop rainwater systems in four rural Auckland districts.  Samples from cold water faucets  were analyzed for chemical and microbiological contaminants. Their study corroborated previous studies suggesting that rooftop rainwater was of relatively poor quality.  Potential microbial pathogens such as SalmonellaAeromonas and Cryptosporidium were identified in some of the rooftop collected rainwater.  Their survey also suggested a significant association between the presence of Aeromonas and increased gastroenteric symptoms among household users (See Table Three).

In 1996, Hollander et al. reported on a study involving 102 German rainwater storage tanks in systems used for toilet flushing, garden irrigation and clothes washing.  Although seven different specific potential microbial pathogens were surveyed, only 12% of the samples contained one of the survey pathogens, Pseudomonas.

In a study of Danish rainwater systems, Albrechtsen (2002) reported that these systems introduce different microorganisms into households than those organisms typically associated with water from community water systems. Different collection surfaces influenced the microbial populations found. Since no disinfection residual is present in Danish drinking water networks, their study suggested that improperly designed rainwater systems could increase risks of infection to household water supplies especially of cross contamination were to occur.

Evans, et al. (2006) studied direct roof run-off from urban housing in Newcastle, Australia. They monitored a total of 77 samples collected during 11 separate rainfall events for microbial levels and several ionic chemical contaminants. Their report shed light on the microbial compositions of roof-harvested rainwater and revealed the following conclusions concerning microbial loads commonly found in rainwater collection systems.

  1. Local weather conditions have a profound influence on microbial loads.
  2. The microbial profile found in rainwater systems was dependent on local environmental conditions and wind speeds/directions.
  3. Commonly used traditional microbial indicators for community treated drinking water supply systems (coliform counts) were not useful for determining the diversity of bacterial contaminants found in rainwater systems.
  4. The influence of weather patterns on microbial compositions may prove useful in predicting and monitoring potential risks associated with rainwater systems.

Evans et al. (2007) also reported that the bulk of microbial contamination for rainwater catchment systems at an urban housing development in Newcastle, Australia was not due to random animal activity. Studies such as these contribute to the development of effective monitoring and management of rainwater collection and storage systems.

In a study concerning the effects of local environmental disturbances on rainwater systems, Spinks et al. (2006) compared microbiological and chemical contamination levels of rainwater collected from rooftops containing residues of smoke and ash (from nearby extensive bushfires) to several other studies conducted under normal conditions in Victoria and South Australia. They report that even in the case of extreme ecological disturbance such as significant air pollution due to fire particulates in the atmosphere, water quality of rooftop runoff could be maintained by simple precautionary measures such as diversion of the first flush of runoff.

4.1  Conclusions

Rainwater utilization occurs using a variety of impermeable collection areas. Rainwater runoff from streets, parking lots, sidewalks and other ground-surface areas are known to contain substantial amounts of contaminants. These types of rainwater runoff require retention and treatment methods that are not covered in this review.  Rainwater collected from rooftops located many meters above ground has been shown to contain much lower concentrations of contaminants than ground-surface systems.

Recent reports have identified the influence that the storage of rainwater has upon overall water quality.  Han and Mun (2008) recognized that the quality of water released from rainwater systems will not only determine the mode of water usage and treatment but will also affect the degree to which people accept this water source. Their study of rainwater systems on the campus of Seoul National University in Korea revealed how particles could be more efficiently removed and how this affected the ultimate water quality. Efficient particle removal was seen with longer distances between inlet and outlets in the storage tanks.  They suggest that more effective particle removal can be attained with tanks designed to maintain a minimum of at least 3 meters of water depth at all times as well as floating outlet suction devices that would draw from near the water surface.

A specific area of research not addressed by current literature is the effect of diverting water from a collection area during the initial phases of each rainwater event. The highest levels of contamination occur during the initial periods of rainwater runoff from any collection area (Kim et al. 2005b).    The diversion of this initial “first flush” away from the rest of the collected rainwater results in a dramatic increase in water quality. However, not enough is known about geographical parameters, the effects of weather patterns, the volumes required, and properties of catchment surfaces to identify exactly what constitutes a “first flush”. More studies are needed in this area.

Untreated rainwater collection systems may constitute a risk for contamination of public drinking water systems if proper guidelines concerning back-siphonage, leakage, and incorrect installation with cross-connections are not designed into each system. This is especially pertinent to community systems worldwide that do not maintain disinfection residuals throughout the water system.

Most national health legislation in countries worldwide presumes that all water entering a community building from a central water supply meets each countries definition of potable quality. Any other water supply is presumed to be non-potable and not fit for human contact/consumption and there is currently no leeway for the use of these other non-potable water sources. Interestingly, most national legislation guidelines concerning potable water do not apply to individual homes that are not served from a central system (ex. rural residences); and some community buildings, such as schools, are often routinely approved for exemption if not connected to an existing community water system.

Countries with substantial portions of the population turning to rainwater collection are beginning to permit the use of this water source for certain uses under certain conditions. In many countries, untreated, collected rainwater can now be used for external water uses (such as irrigation, automobile washing, or toilet flushing) but only where design and construction of these systems prevents cross contamination or cross-connections.

A number of experimental buildings that incorporate rainwater harvesting systems have been constructed throughout the world and studies have demonstrated unequivocally that such systems can be designed, constructed and implemented with due regard to public and environmental health.

A continuing problem worldwide is the lack of local guidelines for specifying what water quality is suitable for non-potable applications.  Plumbing codes may allow for dual plumbing for both potable and non-potable water but decisions concerning water quality are often left to municipal permit departments and plumbing inspectors. Because of a general lack of experience, there is reluctance by local agencies to approve plans such as rainwater collection systems for buildings also served by community water systems.

Risks of contamination appear to be limited to those rainwater systems that do not have proper design, proper treatment procedures or adequate disinfection procedures. Policies concerning the best designs and most effective maintenance routines of rooftop collected rainwater systems need to be established for minimizing contamination of these water sources.  More epidemiological studies are needed such as those reported by Heyworth et al. (2006) where risk factors for illness were directly related to consumption of rainwater. Their report suggested that consumption of rainwater from rooftop collection systems did not increase the risk of gastroenteritis relative to public water sources. They also suggested that more studies are needed to determine if new consumers of collected rainwater are at a greater risk of infection than those individuals accustomed to this type of water.

There are many aspects of rainwater collection systems that need to be addressed by developing technologies.  More effective design and engineering of systems needs to be documented.  Identification of appropriate construction materials is an immediate area in need of research (Kim et al. 2005a). Studies such as Evans et al. (2006) suggest that the extent of contamination occurring in rainfall runoff may be predictable by the local environmental conditions and weather patterns at any given site. This would assist local agencies in identifying those geographical areas where rainwater collection would yield low quality water.

Magyar et al. (2007) noted that large scale integration of rainwater catchment systems will require governmental and legal considerations. Such systems would represent a transfer of responsibility from local water authorities to property owners. Occupants of buildings that contain rainwater collection systems would assume legal liabilities for ownership, operation and maintenance of such systems. Even when identical local codes have been followed, individual systems are still unique in some aspect. Well designed catchment systems will have to address fit-for-purpose water qualities. If design regulations do eventually become implemented, they will be useful for the physical parameters of a system but may not address the actual performance of each system.  Maintenance requirements must be achievable and ensure the long-term success of rainwater systems integration into urban water systems.

Guidelines for rooftop runoff collection systems are being developed throughout the world. Research studies such as those reviewed in this report have identified the following parameters as particularly important for optimum performance of rainwater collection systems:

▪     Proper design/sizing of all parts of the system

▪     Use of most appropriate materials in construction

▪     Proper treatment/disinfection materials and procedures

▪     Regular schedules of maintenance for existing systems

▪     Intended use of the water

▪     Periodic testing of water quality

▪     Education and certification of individuals associated with the governing, approval, and usage of these types of systems.

The views expressed in this article are those of the individual author and do not necessarily reflect the views and policies of the United States Environmental Protection Agency (EPA).


5.1  References

Albrechtsen, H-J.,   2002.  Microbiological investigations of rainwater and graywater collected for toilet flushing.  Water Science and Technology  46(6-7):311-316.

Berndtsson, J. C., Emilsson, T., Bengtsson, L.,  2006.  The influence of extensive vegetated roofs on runoff water quality.  Science of the Total Environment 355:48-63.

Birks, R., Colbourne, J., Hills, S., Hobson, R.,  2004.  Microbiological water quality in a large in-building, water recycling facility.  Water Science and Technology 50(2):165-172.

Canadian Guidelines for Household Reclaimed Water for Use in Toilet and Urinal Flushing.   2007.

Chang, M., McBroom, M.W., Beasley, R.S.  2004.  Roofing as a source of nonpoint water pollution.  Journal of Environmental Management 73:307-315.

Dillaha, T.A., Zolan, W.J.  1985.  Rainwater catchment water quality in Micronesia.  Water Research 19:741-746.

Evans, C.A., Coombes, P.J., Dunstan, R.H.  2006.  Wind, rain and bacteria: The effect of weather on the microbial compsotion of roof-harvested rainwater.  Water Research  40:37-44.

Evans, C.A., Coombes, P.J., Dunstan, R.H., Harrison, T.  2007.  Identifying the major influences on the microbial compostion of roof harvested rainwater and the implications for water quality.  Water Science and Technology 55(4):245-253.

Han, M.Y., Mun, J.S.  2008.  Particle bahaviour consideration to maximize the settling capacity of rainwater storage tanks.  Water Science and Technology 56(11):73-79.

Heyworth, J.S., Glonek, G., Maynard, E.J., Baghurst, P.A., Finlay-Jones, J.  2006.  Consumption of untreated tank rainwater and gastroenteritis among young children in South Australia.  International Journal of Epidemiology 35:1051-1058.

Hollander, R., Bullermann, M., Gross, C., Hartung, H., Konig, K., Lucke, F-K., Nolde, E.   1996.  Mikrobiologisch-hygienische Aspekte bei der Nutzung von Regenwasser als Betriegswasser fur Toilettenspuling, Gartenbewasserung und waschewaschen.   Gesundheitswesen  58:288-293.

Kim, R-H., Lee, S., Lee, J-H., Kim, Y-M., Suh, J-Y  2005a.  Developing technologies for rainwater utilization in urbanized areas.  Environmental Technology 26:401-410.

Kim, R-H., Lee, S., Kim, Y-M., Lee, J-H., Kim, S-K., Kim, S-G.  2005b.  Pollutants in rainwater runoff in Korea: their impacts on rainwater utilization.  Environmental Technology  26:411-420.

Lye, D.J.,  2002.  Health risks associated with consumption of untreated water from household roof catchment sy stems.  J. Am. Water Resour. Assoc.  38:1301-1306.

Magyar, M.I., Mitchell, V.G., Ladson, A.R., Diaper, C.  2007.  An investigation of rainwater tanks quality and sediment dynamics.  Water Science and Technology  56(9):21-28.

Melidis, P., Akratos, C.S., Tsihrintzis, V.A., Trikilidou, E.  2007.  Characterization of rain and roof drainage water quality in Xanthi, Greece.  Environmental Monitoring Assessment 127:15-27.

Polkowska, Z., Gorecki, T., Namiesnik, J.   2002.   Quality of roof runoff waters from an urban region (Gdansk, Poland).  Chemosphere 49:1275-1283.

Signor, R.S., Ashbolt, N.J., Roser, D.J.  2007.  Microbial risk implications of rainfall-induced runoff events entering a reservoir used as a drinking-water source.  Journal of Water Supply: Research and Technology – AQUA  56(8):515-531.

Simmons, G., Hope, V., Lewis, G., Whitmore, J., Gao, W.  2001.,  Copntamination of potable roof-collected rainwater in Auckland, New Zealand.  Water Research 35(6):1518-1524.

Spinks, J., Phillips, S., Robinson, P., Buynder, P.V.  2006.  Journal of Water and Health 4(1):21-28.

USEPA.  1999.  National Recommended Water Quality Criteria-Correction.  EPA 822-Z-99-001.

USEPA.  2004.  Guidelines for Water Reuse   EPA/625/R-04/108.



By: Dennis J. Lye



MS  314

26 W. Martin Luther King Drive

Cincinnati, Ohio     45268,   USA            


Phone: 513-569-7870

FAX    513-569-7170