Urban Water Journal, Vol. 4, No.

4, December 2007, 253 – 260

Microbial quality of rainwater supplies in developed countries:

a review

L. FEWTRELL*{ and D. KAY{

{Centre for Research into Environment and Health, Coppice House, Quakers Coppice, Crewe, Cheshire CW1 6FA, UK
{Hydrology and River Basin Dynamics Research Group, IGES, University of Wales, Aberystwyth SY23 3DB, UK

The current paper examines the reported microbial quality of rainwater supplies. The
majority of microbial contamination derives from debris and faecal material deposited on
the roof surface, principally from birds. The prevalence and level of contamination varies
widely, both in terms of indicator organisms and pathogens. Gastrointestinal pathogens
such as Campylobacter spp., Salmonella spp., Cryptosporidium spp. and Giardia spp. have
all been isolated from harvested supplies, although these may not always be of types
infective to humans. In the UK, there are currently no regulations in respect to the
microbial quality of harvested rainwater for non potable use, although there are generally
standards relating to the required plumbing to ensure that potable and non potable
supplies remain separate.

Keywords: Faecal indicator organisms; Guidelines; Microbial quality; Pathogens;

Rainwater

1. Introduction While harvested rainwater is not generally used (inten-

tionally) as drinking water in the UK, it is used within the
It is likely that rainwater harvesting has been employed in
house where people could come into contact with it and it is
some areas since 2000 BC, or earlier. Its use in the UK,
being used to replace a highly regulated drinking-water
however, to provide in-house water for toilet flushing and
source; thus, the quality of this supply, and in particular the
other non-potable uses is relatively recent and, in light of
microbial quality, is of interest. The present paper reviews
mounting interest (and need) for water demand manage-
the microorganisms that could be present in supplies (both
ment, is set to increase.
indicator organisms and potential human pathogens) based
Although it is a simple concept, where rain falling onto a
on literature relating to harvested supplies used in
roof surface is collected and subsequently used as a source
developed countries and, briefly, examines possible water
of water, systems can vary in complexity from a simple
quality guidelines that may be appropriate in relation to
water butt (where supplies are used for garden watering) to
rainwater use.
in-house systems. In the UK, in-house systems typically
consist of an underground storage tank, a filter to prevent
the entry of leaves and large solids, and a pump for
distributing the harvested water (to a header tank or 2. Microorganisms of interest
directly to appliances) via a system of pipes, separate (and
2.1 Pathogens
distinct) from those used for potable supplies. Systems also
generally have an automatic mains top-up device, which Pathogens are infectious microorganisms that can cause
may supply the storage tank, a header tank or the infection and disease. These may be classed as bacterial
appliances with mains water if harvested supplies run short (e.g. Salmonella spp.), fungal (Candida spp.), viral (e.g.
(figure 1). enteroviruses), protozoal (e.g. Cryptosporidium spp.) or

*Corresponding author. Email: [email protected]

Urban Water Journal

ISSN 1573-062X print/ISSN 1744-9006 online ª 2007 Taylor & Francis
http://www.tandf.co.uk/journals
DOI: 10.1080/15730620701526097
254 L. Fewtrell and D. Kay

helminth (Trichobilharzia spp.). Pathogens may be intro- The table excludes human viruses, which are excreted in
duced to water as a result of contamination (often from human faecal material, as they would not be expected to
faecal material) or may be naturally present in the found in rainwater supplies, although there is a theoretical
environment (autochthonous) (Fewtrell et al. 1994). A risk of contamination from aerosols derived from waste-
number of human pathogens that could, in theory, be water treatment works and a possible risk of contamination
found in UK rainwater supplies are summarized in table 1, during storage.
along with the illnesses they cause and case-fatality rates, The most common manifestation of water borne illness is
to give an indication of the severity of the illness. While this gastrointestinal upset (i.e. diarrhoea, nausea and vomiting)
table is not exhaustive, it does illustrate those pathogens generally of short duration (Health Canada 2006). How-
most frequently sought in harvested rainwater supplies. ever, a number of pathogens can cause serious (even fatal)
illness, especially in vulnerable groups such as the immuno-
compromised, young and elderly.
Most pathogens have specific and unique isolation
methods, for example the method for E. coli O157 is
different to that for Campylobacter spp., which is different
to that for Salmonella spp., and tests are usually time
consuming and may be very expensive. Thus, unless there is
interest in a specific pathogen it is not usual to detect
individual pathogens, rather ‘indicator organisms’ are
enumerated in order to indicate the faecal contamination
and the resultant possible presence of pathogens.
The principal source of pathogens in harvested supplies
in developed countries is likely to be avian. The intestinal
carriage of a number of pathogens by wild birds has been
recognised for a long time (e.g. Shirlaw and Iyer 1937). The
two most commonly studied and (where a variety of
pathogens have been examined) most frequently isolated
pathogens from birds are Salmonella spp. and Campylo-
bacter spp. (e.g. Girdwood et al. 1985, Craven et al. 2000,
Gautsch et al. 2000, Waldenström et al. 2002).

2.2 Faecal indicator organisms

Figure 1. Schematic of an in-house rainwater harvesting Microbial quality is usually assessed by measuring ‘faecal
system (adapted from EA 2003). indicator bacteria’ (also referred to as faecal indicator

Table 1. A selection of pathogens that could be present in UK rainwater supplies (adapted from Fewtrell et al. 1994).

Group Pathogen Infection Transmission Case fatality rate/100 000 cases

Bacteria Campylobacter spp. Gastroenteritis Oral 5*

Bacteria Escherichia coli O157:H7 Gastroenteritis Oral 8.3*
Bacteria Legionella pneumophila Legionnaire’s disease Inhalation 10 000{
Pontiac fever 0
Bacteria Mycobacterium avium complex (MAC) Respiratory and Inhalation, Fatal cases usually only
gastrointestinal Oral, Contact associated with those with
tract infection underlying infections{
Bacteria Salmonella spp. (non typhoid formsx) Gastroenteritis Oral 41*
Protozoa Cryptosporidium spp. Gastroenteritis Oral 22*
Protozoa Giardia spp. Gastroenteritis Oral 1*

*Mead et al. 1999 (table 3).

{
Benin et al. 2002 – community acquired cases.
{
Pond 2005.
x
S. typhi and S. paratyphi, which cause enteric fever (typhoid and paratyphoid) are host-adapted to humans and are not transmitted by other species
(Lightfoot 2004).
 Microbial quality of rainwater supplies in developed countries 255

organisms or FIOs). Thermotolerant coliforms (also differences in maximum indicator concentrations and the
termed faecal coliforms), Escherichia coli and intestinal frequency with which indicators are isolated. The pre-
enterococci (also termed faecal streptococci) are generally valence of E. coli in tank water, for example, ranges from 0
harmless bacteria that are present in high numbers in faecal to almost 80%.
material and are the most commonly examined FIOs. A number of investigators (Coombes et al. 2000, Thomas
(Note: E. coli O157 is pathogenic.) Thus, their presence in and Greene 1993, Uba and Aghogho 2000, Yaziz et al.
water samples is used to indicate the presence of faecal 1989) have also examined the microbial quality of rain-
pollution and the possibility that faecally associated water prior to contact with roof surfaces. E. coli,
pathogens may also be present. Other, less frequently thermotolerant coliforms and enterococci were not found
utilised FIOs include Clostridium perfringens and coliphage in any samples. Pseudomonas spp., including the opportu-
organisms. The ecology and environmental survival char- nistic pathogen Ps. aeruginosa, have been found in
acteristics of bacterial, viral and protozoal pathogens vary rainwater before contact with roof surfaces (e.g. Coombes
and so there is no single ideal indicator organism et al. 2002).
(Savichtcheva and Okabe 2006). Total coliforms have, in The presence of a number of pathogens in rainwater
the past, also been used as faecal indicators but they are supplies has also been investigated in a small number of
now considered to be too non-specific and they have fallen studies. The levels of pathogens are often not enumerated
out of favour. in water samples, with investigators simply determining
their presence or absence. Prevalence data and, where
available, pathogen concentrations are summarized in
3. Rainwater quality
table 3. Aeromonas spp., Mycobacterium spp., and
The main sources of external contamination of rainwater Pseudomonas aeruginosa were isolated in each study which
supplies are pollution from the air (which may include looked for them; Cryptosporidium spp., Campylobacter
microbial pollution – Evans et al. 2006), bird and animal spp. and Salmonella spp. were isolated in half of the
droppings, insects and materials washed out of roofing studies which looked for them. Giardia spp. were isolated
material. A number of studies conducted in various in one of the four studies where analyses were conducted,
countries have demonstrated that roof-water quality while E. coli O157, L. pneumophila and Shigella spp.
varies, with the greatest pollution load being related to were not isolated in any of the studies. Although
the first flush (e.g. Yaziz et al. 1989, Förster 1999, Che Cryptosporidium and Giardia have forms which are
2001, Ghanayem 2001, Gardner et al. 2004). Debris can pathogenic to humans, it is most likely that those isolated
build up on roofs during dry spells, meaning that the from rainwater supplies are avian forms (Birks et al. 2004,
initial runoff is likely to be full of debris, sediment and be Sinclair et al. 2005), which are generally not human
highly turbid. Rainfall intensity and the number of dry pathogens.
days preceding a rainfall event can markedly affect the
quality of runoff water. Gardener et al. (2004) reported
4. Water quality guidelines
that faecal coliform levels measured in first flush events
from a single property in Australia varied widely over 11 There are no regulations in the UK covering the microbial
rainfall events. Discarding the first 1 mm of runoff reduced quality of harvested rainwater, for non-potable purposes,
the bacterial load entering the tank by between 9 and 62% however a number of guidelines exist, that outline levels of
for individual rainfall events. Stored water, however, still faecal coliforms, E. coli or intestinal enterococci which may
contained significant levels of microbial contamination be appropriate for non-potable use (table 4). The World
with faecal coliform concentrations up to 480 cfu/100 ml Health Organization (WHO) has recently updated their
being reported (Gardner pers. com.). In addition, Evans Guidelines for the safe reuse of water (WHO 2006),
et al. (2006) investigated the effect of weather (such as however, this does not include specific microbial targets
wind direction) on the microbial composition of roof- but is, rather, a methodology using health-based targets for
water and found that weather patterns, in conjunction establishing appropriate guidelines at a local, regional or
with other factors (such as relative source location), can country level.
significantly influence the bacterial load of roof run- It can be seen that there is little consistency between these
off. They also noted that airborne microorganisms guidelines, with the USEPA guidelines requiring virtually
are significant contributors to the bacterial load of roof- drinking-water quality standards. Many of the systems
water. detailed in table 2, would fail to achieve most of guideline
Studies measuring rainwater quality, using faecal in- values on at least some occasions.
dicators, in Europe, Australasia and the USA are It is likely that appropriate microbial standards could aid
summarized in table 2 (adapted from Sinclair et al. 2005). the more widespread introduction of rainwater harves-
It can be seen from this table that samples exhibit wide ting systems (Nash, personal communication). A suggested
 256

Table 2. Presence of faecal indicator bacteria in roof-water and stored rainwater from Europe and other developed countries (adapted from Sinclair et al. 2005).

Setting Parameter Roof run-off (/100 ml) Tank water (/100 ml) Comments Reference

EUROPE
Denmark, urban area E. coli Prevalence 78.6% (n ¼ 14) Samples taken from Albrechtsen (2002)
Range 4 – 990 7 rainwater harvesting
Median 245 systems
London, urban area E. coli Prevalence 94% (n ¼ 88) Rainwater collected Birks et al. (2004);
Range 0 – 16 000 (prior to treatment) Birks (pers. com.)
Median 52 from the roof of the
Millennium Dome
Enterococci Prevalence 97% (n ¼ 34)
Range 0 – 680
Median 4200
England, semi rural E. coli Prevalence 34.6% (n ¼ 26) Samples taken from Day (2002)
Range 0 – 53 outside tap, possibility
of mains water top-up
Faecal streptococci Prevalence 57.7% (n ¼ 26)
Range 0 – 79
Germany, urban E. coli n ¼ 972 102 rainwater tanks Holländer et al. (1996)
Median 26
Maximum 410 000
Faecal streptococci n ¼ 969
Maximum 4 10 000
Scotland, rural area Thermotolerant coliforms Geometric mean range Sampling from farm roofs Kay and Edwards (2003)
383 – 3627 (n ¼ 68) (3 of the 4 farms were
housing livestock at the
time of sampling).
L. Fewtrell and D. Kay

The lowest counts were

recorded from the farm
without livestock.
Faecal streptococci Geometric mean range
482 – 46580 (n ¼ 68)
AUSTRALIA
Australia, rural towns E. coli Prevalence 25.5% (n ¼ 47) Bannister et al. (1997)
Range 0 – 370 cited by Sinclair et al., 2005
Mean 15
Australia, urban area Thermotolerant coliforms Highest average figure (743) n ¼ 30 Poor design and construction Coombes et al. (2000, 2002)
recorded at rain depths Maximum 800 of gutters and tanks (including
of 0.5 – 1 mm Mean 119 the installation of stormwater
detention tanks rather than
rainwater tanks) believed to
contribute to poor water quality
Australia, rural, Thermotolerant coliforms Prevalence 100% (n ¼ 24) Mean counts 55 – 125 Data presented in graphical form Thomas and Greene (1993)
urban þ industrial Range 75 – 125
Australia E. coli Not detected Thurman (1995) cited by
Sinclair et al. 2005
(continued)
 Microbial quality of rainwater supplies in developed countries 257

approach to developing non-potable water quality

standards in the UK, is to use the EC bathing water

Simmons et al. (2001)

directive as a baseline (e.g. MTP 2007). Given the
markedly different levels of exposure between full body

Savill et al. (2001)

water recreation and toilet flushing with harvested rain-
water (say), a more logical approach would seem to

Lye (1987)
Reference

be employing a health impact assessment approach

(employing quantitative microbial risk assessment),
where specific hazards are identified and their health
impact quantified (Fewtrell and Kay, in press). Dose-
response relationships could be used to establish an
appropriate maximum level of microbial contamination
systems examined, with

based on accepted tolerable risk levels (e.g. WHO

125 rainwater harvesting

2004).
sampling from the

30 tanks sampled
household tap

5. Discussion
Comments

It can be seen that harvested rainwater supplies vary widely

MPN

in terms of microbial quality and are not a ‘pure’ supply of

water that they are perhaps perceived to be. Faecal
indicator organisms are often, if not usually, isolated from
Prevalence 56% (systems)
Prevalence 38% (n ¼ 24)

roof run-off and harvested supplies, sometimes at very high

Prevalence 3% (tanks)
Typical values 10 – 20
Tank water (/100 ml)

levels. In addition, pathogens can also occasionally be

(Continued ).

isolated from supplies, although they are looked for less

Range 0 – 4900
Range 0 – 111

Range 0 – 840

frequently than FIO.

The principal source of pathogens in harvested rainwater
supplies in developed countries is avian faeces and it is
likely that where pathogens are isolated many of them (e.g.
Table 2.

Cryptosporidium spp. and Giardia spp.) may not be

generally infective to humans, but are avian forms (Birks
et al. 2004 and Sinclair et al. 2005). A number of pathogens
Roof run-off (/100 ml)

such as those responsible for avian cholera (Pasteurella

multocida) are not pathogenic for humans (due to
differences in species susceptibility) via the waterborne
route (Friend and Franson 1999), while other serious
human pathogens (e.g. those responsible for typhoid –
S. typhi) are host adapted to humans and are not carried by
avian sources (Lightfoot 2004).
Thermotolerant coliforms

Thermotolerant coliforms

There is generally no routine monitoring of pathogens or

FIO in rainwater supplies used for non-potable supplies,
but pathogen types are likely to vary according to the
MPN, most probable number enumeration.

location (and local wildlife) and pathogen concentrations

Enterococci
Parameter

may well show temporal variations, making any prediction

of occurrence difficult.
E. coli

First flush diverters, which remove a proportion of the

initial roof runoff, may offer a way of reducing microbial
New Zealand, rural area

New Zealand, rural area

contamination in supplies. In Australia, for example, a

study found that quality improved as rain depth increased,
with the first portion of the rainfall, which effectively acted
USA, rural area
NEW ZEALAND

to clean the roof, being discarded. Although, where

harvested rainwater is used for non-potable purposes, to
which people have limited contact, it may be more useful
Setting

to maximize the volume of the supply, rather than the

USA

quality.
258 L. Fewtrell and D. Kay

Table 3. Prevalence of pathogens in roof-water and stored rainwater (adapted from Sinclair et al. 2005).

Setting Pathogen type Parameter Roof run-off Tank water Reference

EUROPE
Denmark, urban area Faecal Campylobacter spp. 11.8% (n ¼ 17) Albrechtsen (2002)
Cryptosporidium spp. 35.2% (n ¼ 17) max
50 oocysts/l
Giardia spp. 0% (n ¼ 17)
Opportunistic Aeromonas spp. 14% (n ¼ 14) max 30/ml
Legionella spp. 71% (n ¼ 7)
L. pneumophila 0% (n ¼ 14)
M. avium 7% (n ¼ 14)
P. aeruginosa 7% (n ¼ 14) max 20/100 ml
London, urban Faecal Campylobacter spp. 0% (n ¼ 2) Birks et al. (2004)
Cryptosporidium spp. 0% (n ¼ 2)
E. coli O157 0% (n ¼ 2)
Giardia spp. 50% (n ¼ 2) 0.2 cysts/l
Salmonella spp. 0% (n ¼ 2)
Shigella spp. 0% (n ¼ 2)
Opportunistic L. pneumophila 0% (n ¼ 2)
Germany, urban Faecal Campylobacter spp. 0% (n ¼ 142) Holländer et al. 1996
Salmonella spp. 0.1% (n ¼ 798)
Shigella spp. 0% (n ¼ 342)
Opportunistic P. aeruginosa 11.1% (n ¼ 710)
Legionella spp. 0% (n ¼ 418)
AUSTRALIA
Australia, rural towns Faecal Campylobacter spp. 13% (n ¼ 42) Bannister et al. (1997)
Cryptosporidium spp. 0% (n ¼ 20)
Giardia spp. 0% (n ¼ 20)
Australia, urban Opportunistic Pseudomonas spp. Max 146 100% (n ¼ 18) max Coombes et al. (2002)
723/100 ml 15 200/100 ml
Australia Faecal Salmonella spp. 0% (n 4 60) Thurman (1995)
Shigella spp. 0% (n 4 60)
Australia Opportunistic Mycobacterium spp. 35% (n ¼ 205) Tuffley and
Urban tanks had the Holbeche (1980)
lowest prevalence
NEW ZEALAND
New Zealand, rural Faecal Campylobacter spp. 38% (n ¼ 24) max Savill et al. (2001)
0.56/100 ml
New Zealand, rural Faecal Campylobacter spp. 0% (n ¼ 115) Simmons et al. (2001)
Cryptosporidium spp. 4% (n ¼ 50)
Giardia spp. 0% (n ¼ 50)
Salmonella spp. 0.9% (n ¼ 115)*
Opportunistic Aeromonas spp. 16% (n ¼ 125)
Legionella spp. 0% (n ¼ 23)

*Salmonella typhimurium.

Table 4. Summary of microbial quality guidelines that could be applicable to the non-potable use of rainwater (adapted from
CIRIA 2003).

WHO (2003)
Bathing water WHO (1989) USEPA (1992) Recreational water Bathing water
directive (EC 1975) Reuse guidelines* Water reuse WRAS (1999) guidelines{ directive{ (EC 2006)

Faecal coliforms 95% Faecal coliforms E. coli 51/100 ml Faecal coliforms Intestinal enterococci E. coli 95th percentile
52000/100 ml GM 1000/100 ml 510 000/100 ml 95th percentile 51000/100 ml
Intestinal enterococci 5200/100 ml Enterococci 95th percentile
5100/100 ml 5400/100 ml

*Category A – irrigation of crops likely to be eaten uncooked.

{
Marine sites to be classified as good.
{
Freshwaters.
GM, geometric mean.
 Microbial quality of rainwater supplies in developed countries 259

Day, M.M., A report on ‘Freerain’, the rainwater recycling system. Report

6. Conclusions for Gusto Construction Ltd., Severn Trent Water and the Environment
Agency, 2002.
Data relating to the microbial quality of rainwater supplies Environmental Agency. Harvesting Rainwater for Domestic Uses: An
in the UK and developed countries have not been exten- Information Guide, 2003 (EA: Bristol).
sively published. Some small-scale studies monitoring both EC. Bathing Water Directive. 76/160/EEC. European Community, 1975.
FIO and a number of selected pathogens at several sites Cited by CIRIA, 2003.
EC. Directive 2006/7/EC of The European Parliament and of The Council
over a year-long period would provide additional data
of 15th February 2006 concerning the management of bathing water
which would be useful for health impact assessment and quality and repealing Directive 76/160/EEC. Official J. European Union
standards development. 2006, L64, 37 – 51.
Where supplies are used solely for non-potable purposes Evans, C.A., Coombes, P.J. and Dunstan, R.H., Wind, rain and bacteria:
there is often no requirement for quality monitoring. The effect of weather on the microbial composition of roof-harvested
rainwater. Water Res., 2006, 40, 37 – 44.
There are a number of possible guidelines that could be
Fewtrell, L., Godfree, A., Jones, F., Kay, D. and Merrett, H., Pathogenic
adapted or adopted for determining the microbial quality Microorganisms in Temperate Environmental Waters, 1994 (Samara
of non-potable rainwater supplies, but these vary widely Publishing: Tresaith, UK).
and, based on published data, supplies are unlikely to con- Fewtrell, L., Kay, D., Matthews, I. and Smith, K. Rainwater harvesting –
sistently achieve the standards. Ideally, any proposed an HIA of rainwater harvesting in the UK. In: A guide to the health
impact assessment of flooding and sustainable water management.
standards should be based on a health impact assessment
(in press) (IWA Publishing: London).
approach, to ensure that microbial concentrations are at an Friend, M. and Franson, J.C., Field Manual of Wildlife Disease 2, 1999 (US
appropriate level to protect human health. Department of the Interior, US Geological Survey).
Förster, J., Variability of roof runoff quality. Water Sci. Technol., 1999, 39,
137 – 144.
Acknowledgement Gardner, T., Baisden, J. and Millar, G., Rainwater first flush devices – are
they effective? In: Sustainable Water in the Urban Environment, 2004
This work was funded by an Engineering and Physical Conference, Brisbane, Australia, 30 – 31 August 2004.
Sciences Research Council grant (Ref: GRS 18373/01). Gautsch, S., Odermatt, P., Burnens, A.P., Bille, J. and Ewald, R., The role
of common starlings (Sturnus vulgaris) in the epidemiology of bacterial,
potentially pathogenic, disease agents. Schweizer Archiv fur Tierheilk-
References
unde, 2000, 142, 165 – 172. In German.
Albrechtsen, H.J., Microbiological investigation of rainwater and greywater Ghanayem, M., Environmental considerations with respect to rainwater
collected for toilet flushing. Water Sci. Technol., 2002, 46, 311 – 316. harvesting. In: Proceedings of the 10th International Rainwater Catchment
Bannister, R., Westwood, J. and McNeill, A., Investigation of micro- Systems Conference (Rainwater International), Mannheim, Germany,
biological and chemical water quality in rainwater tanks in Victoria, 2001.
Melbourne. Water ECO science report no. 139/97 – Department of Girdwood, R.W.A., Fricker, C.R., Munro, D., Shedden, C.B. and
Natural Resources and Environment, 1997. Cited by Sinclair et al., 2005. Monaghan, P., The incidence and significance of salmonella carriage
Benin, A.L., Benson, R.F. and Besser, R.E., Trends in Legionnaires by gulls (Larus spp.) in Scotland. J. Hygiene, 1985, 95, 229 – 241.
disease, 1980 – 1998: declining mortality and new patterns of diagnosis. Health Canada. Bacterial waterborne pathogens – current and emerging
Clin. Infect. Dis., 2002, 35, 1039 – 1046. organisms of concern. 2006 (Health Canada: Ontario, Canada).
Birks, R., Colbourne, J., Hills, S. and Hobson, R., Microbiological water Holländer, R., Bullermann, M., Groß, C., Hartung, H., König, K.,
quality in a large in-building, water recycling facility. Water Sci. Lücke, F.-K. and Nolde, E., Microbiological and hygienic aspects of the
Technol., 2004, 50, 165 – 172. use of rainwater as process water for toilet flushing, garden irrigation and
Che, W., Wang, H., Li, J. and Meng, G., The quality and major influencing laundering. Gesundheitswesen, 1996, 58, 288 – 293. In German.
factors of runoff in Beijing’s urban area. In Proceedings of the 10th Kay, D. and Edwards, A.C., Evaluation of bacterial loads from farmyard
International Rainwater Catchment Systems Conference (Rainwater drainage systems, 2003. Report to the Scottish Executive. Centre for
International), Mannheim, Germany, 2001. Research into Environment and Health, Wales.
Construction Industry Research and Information Association. Model Lightfoot, D., Salmonella and other enteric organisms. In Waterborne
agreements for sustainable water management systems. review of existing zoonoses. Identification, Causes and Control, edited by J. A. Cotruvo,
legislation. RP 664. 2003 (CIRIA: London). A. Dufour, G. Rees, J. Bartram, R. Carr, D. O. Cliver, G. F. Craun,
Coombes, P.J., Kuczera, G. and Kalma, J.D., Rainwater quality from R. Fayer and V. P. J. Gannon, 2004 (IWA Publishing: London).
roofs, tanks and hot water systems at Figtree Place. Proceedings of the 3rd Lye, D.J., Bacterial levels in cistern water systems of Northern Kentucky.
International Hydrology and Water Resource Symposium, Perth, Water Res. Bull., 1987, 23, 1063 – 1068.
Australia, 2000. Mead, P.S., Slutsker, L., Dietz, V., McCraig, L.F., Bresee, J.S.,
Coombes, P.J., Kuczera, G. and Kalma, J.D., Rainwater quality from Shapiro, C., Griffin, P.M. and Tauxe, R.V., Food-related illness
roofs, tanks and hot water systems at Figtree Place. In Proceedings of the and death in the United States. Emerging Infect. Dis., 1999, 5, 607 –
3rd International Hydrological and Water Resources Symposium, Perth, 625.
Australia, 2002, pp. 152 – 157. Market Transformation Programme, Rainwater and greywater: review
Craven, S.E., Stern, N.J., Line, E., Bailey, J.S., Cox, N.A. and of water quality standards and recommendations for the UK. Market
Fedorka-Cray, P., Determination of the incidence of Salmonella spp., Transformation Programme, 2007.
Campylobacter spp., and Clostridium perfringens in wild birds near Pond, K., Water Recreation and Disease. Plausibility of Associated
broiler chicken houses by sampling intestinal droppings. Avian Dis., Infections: Acute Effects, Sequelae and Mortality, 2005 (IWA Publishing:
2000, 44, 715 – 720. London).
260 L. Fewtrell and D. Kay

Savichtcheva, O. and Okabe, S., Alternative indicators of fecal pollution: Uba, B.N. and Aghogho, O., Rainwater quality from different roof catch-
relations with pathogens and conventional indicators, current methodol- ments in the Port Harcourt district, Rivers State, Nigeria. Aqua 2000, 49,
ogies for direct pathogen monitoring and future application perspectives. 281 – 288.
Water Res., 2006, 40, 2463 – 2476. United States Environmental Protection Agency, Guidelines for Water
Savill, M.G., Hudson, J.A., Ball, A.A., Klena, J.D., Scholes, P., Reuse, EPA/625/R-92/004, 1992 (USEPA: Ohio).
Whyte, R.J., McCormick, R.E. and Jankovic, D., Enumeration of Waldenström, J., Broman, T., Carlsson, I., Hasselquist, D., Acherberg, R.P.,
Campylobacter in New Zealand recreational and drinking waters. Wagenaar, J.A. and Olsen, B., Prevalence of Campylobacter jejuni,
J. Appl. Microbio., 2001, 91, 38 – 46. Campylobacter lari, and Campylobacter coli in different ecological guilds
Shirlaw, J.F. and Iyer, S.G., A note on the variety of Salmonella enteritidis and taxa of migrating birds. Appl. Environ. Microbiol., 2002, 68,
isolated from pidgeons (sic). Indian J. Vet. Sci., 1937, 7, 231. Cited by 5911 – 5917.
Girdwood et al., 1985. World Health Organization, Health Guidelines for the use of Wastewater in
Simmons, G., Hope, V., Lewis, G., Whitmore, J. and Gao, W., Agriculture and Aquaculture, Technical Report Series 778. 1998 (WHO:
Contamination of potable roof-collected rainwater in Auckland, New Geneva, Switzerland).
Zealand. Water Res., 2001, 35, 1518 – 1524. World Health Organization, Guidelines for Safe Recreational Water
Sinclair, M.I., Leder, K. and Chapman, H., Public health aspects of Environments, Vol. 1: Coastal and fresh waters, 2003 (WHO: Geneva,
rainwater tanks in urban Australia. CRC for Water Quality and Switzerland).
Treatment; Occasional Paper 10, 2005 (Cooperative Research Centre World Health Organization, Guidelines for drinking-water quality. Third
for Water Quality and Treatment: Australia). edition, Vol. 1. Recommendations. 2004 (WHO: Geneva, Switzerland).
Thomas, P.R. and Greene, G.R., Rainwater quality from different roof World Health Organization, Guidelines for the Safe Use of Wastewater,
catchments. Water Sci. Technol., 1993, 28, 291 – 299. Excreta and Greywater, 2006 (WHO: Geneva, Switzerland).
Thurman, R., Evaluation of rainwater stored in collection tanks. Australian Water Regulations Advisory Scheme, Reclaimed Water Systems, WRAS,
Microbiol., 1995, March, 20 – 22. Cited by Sinclair et al., 2005. UK, 1999. Cited by CIRIA, 2003.
Tuffley, R.E. and Holbeche, J.D., Isolation of Mycobacterium avium – M. Yaziz, M., Gunting, H., Sapari, N. and Ghazali, A.W., Variations in rain-
intracellulare – M. scrofulaceum complex from tank water in Queensland, water quality from roof catchments. Water Res., 1989, 23, 761 – 765.
Australia. Appl. Environ. Microbiol., 1980, 39, 48 – 53. Cited by Sinclair
et al. (2005).