GIS-based analysis of drinking-water supply structures: a module for microbial risk assessment

International Journal of Hygiene and Environmental Health

Int. J. Hyg. Environ. Health 203, 301-310 (2001) © Urban & Fischer Verlag http://www.urbanfischer.de/journals/intjhyg

GIS-based analysis of drinking-water supply structures: a module for microbial risk assessment

Thomas Kistemann, Susanne Herbst, Friederike Dangendorf, Martin Exner Institute for Hygiene and Public Health, University of Bonn, Germany Received August 11, 2000 · Accepted March 26, 2001

Abstract Water-related infections constitute an important health impact world-wide. A set of tools serving for Microbial Risk Assessment (MRA) of waterborne diseases should comprise the entire drinkingwater management system and take into account the Hazard Analysis and Critical Control Point (HACCP) concept which provides specific Critical Control Points (CCPs) reflecting each step of drinking-water provision. A Geographical Information System (GIS) study concerning water-supply structure (WSS) was conducted in the Rhein-Berg District (North Rhine-Westphalia, Germany). As a result, suitability of the existing water databases HYGRIS (hydrological basis geo-information system) and TEIS (drinking-water recording and information system) for the development of a WSSGIS module could be demonstrated. Spatial patterns within the integrated raw and drinking-water data can easily be uncovered by GIS-specific options. The application of WSS-GIS allows a rapid visualization and analysis of drinking-water supply structure and offers huge advantages concerning microbial monitoring of raw and drinking water as well as recognition and investigation of incidents and outbreaks. Increasing requests regarding health protection and health reporting, demands for a better outbreak management and water-related health impacts of global climate change are major challenges of future water management to be tackled with methods including spatial analysis. GIS is assumed to be a very useful tool to meet these requirements. Key words: drinking-water supply, water-related diseases, GIS, HACCP, MRA

Introduction In Germany, in contrast to the USA, Canada and the majority of European countries with comparable drinkingwater supply structures, waterborne disease outbreaks have not been reported for 20 years (Kistemann, 1997; Lack, 1999; Kramer et al., 2001, see Table 1). Considering the possibility of water-related epidemics in areas with highly developed drinking-water supply structures, it appears reasonable to develop a comprehensive set of tools serving for Microbial Risk Assessment

(MRA) during incidents and outbreaks as well as for interpreting of surveillance data (Haas et al., 1999; Gale, 1996). Such a set of tools should include certain modules which comprise the entire drinking-water management system. The Hazard Analysis and Critical Control Point (HACCP) concept has initially been created to improve food safety by identifying and monitoring relevant points within the production process (Untermann, 1996). This concept has been transferred to drinkingwater production. Specific Critical Control Points

Corresponding author: Dr. med. Thomas Kistemann MA, University of Bonn, Institute for Hygiene and Public Health, SigmundFreud-Straße 25, D-53105 Bonn, phone: + 49/(0) 228/287-5534, fax: 49/(0) 228/287-4885, E-mail: [email protected] 1438-4639/01/203/4-301 $ 15.00/0

302

T. Kistemann et al.

(CCPs) have been identified reflecting each step of drinking-water provision (Havelaar, 1994, with additions): – protection of water resources (catchment area, protection zone) – protection of well-head – water abstraction system – drinking-water treatment facility (coagulation, flocculation, sedimentation, filtration) – disinfection point – drinking-water storage system – distribution system – customers’ health outcome The utility of a Geographical Information System (GIS)based module for MRA in catchment areas of drinkingwater reservoirs has recently been demonstrated (Kistemann et al., 2001). Water courses, water protection zones, land use patterns, sewage discharge and farm sites have been integrated into a GIS. The spatial information has been analysed using different GIS facilities (i. e. overlaying, buffering, distance calculation). This paper deals with the application of GIS on water treatment and distribution structures as a precondition for a HACCP-based microbiological monitoring of drinking water and for identification of populations at risk.

Table 1. Reported waterborne disease outbreaks associated with drinking water and bodies of water used for recreational purposes in 19 European countries, 1986–1996 (Lack, 1999; Kramer et al. 2001). Country

Total no. of outbreaks*

Albania Croatia Czech Rep. England and Wales Estonia Germany Greece Hungary Island Latvia Lithuania Malta Norway Romania Slovak Rep. Slovenia Spain Sweden Yugoslavia, Fed. Rep

14 29 18 20 12 0 2 27 1 1 0 162 0 57 61 45 208 53 68

* n = 778

Materials and methods Study area According to the hydrological situation, the proportion of drinking water Abstracted from surface water to that Abstracted from ground water varies between the regions of Germany. Different water resources are used in varying extents for drinking-water production. The total average amount of drinking-water comes from ground water (64 %), purified surface-water (28 %) and spring water (8 %). The water companies in the Federal State of North RhineWestphalia produce 57.5 % of their drinking- water from surface water resources. This exceeds the German average by 26 % (BGW, 1996). The Rhein-Berg District (North Rhine-Westphalia) is situated on a geological border, which causes the use of both, surface-water and ground water within a relatively small area. During the last century complex patterns of water supply structures have been established. Therefore, this district appears to be an appropriate site to prove how GIS may support recording of water supply structures in detail for HACCP purposes. Data and GIS Two existing databases on state and district level concerning the monitoring of water quality in North Rhine-Westphalia have been used. The state hydrological basis geo-information system (HYGRIS) served as source for information on raw water Abstracted by the waterworks. HYGRIS contains spatial data and detailed attribute data of waterworks and their wells, i. e. co-ordinates of abstraction points, origin and amount of abstracted water, treatment procedures and sampling results. It allows basic analytical data processing, but without respect to spatial relation. Unfortunately, HYGRIS does not provide any interface to GIS software which is able to process spatial information. Copies containing spatial information, attribute data and two sampling results per well and year including several chemical and microbiological parameters were available from HYGRIS. 285 raw water sampling data sets, with the majority comprising a set of 3 parameters (Table 2), from 20 sample points at the waterworks which cover a period of seven years (1991–1997) have been extracted from the HYGRIS database. A single waterworks did not provide any microbiological test results over the entire investigation period. Drinking-water data sets have been exported from the state drinking-water data recording and information system (TEIS; MAGS 1994) at the Public Health Department of the Rhein-Berg District. TEIS is required to provide spatial information, attribute data, chemical and microbiological test results on sampling points representing the following categories: waterworks, feeding points, sampling points and private wells. At the Public Health Department of the Rhein-Berg District TEIS was not introduced before 1995. Linkage to geographic information systems is not provided so far, but is intended by the Public Health Institute of the state (Lacombe and Fehr, 1999). Spatial information was partly lacking for the study period. Data was available concerning water companies, water origin, water treatment procedures, amount of

GIS-based analysis of drinking-water supply structures delivered water at feeding points and test results. The database contained test results of 484 sampling points. According to the German Drinking-Water Ordinance (Trinkwasserverordnung, 1990), the number of test results and the time period between samplings varied strongly with respect to the category of sample point. The total amount of drinkingwater test results, which have been transferred from the TEIS database into the GIS, is given in Table 2. The data covers a period of 23 years (1976–1998), thus reflecting a retrospective data input into TEIS by the Public Health Department. About 80 % of the available microbiological data, however, belongs to the last 4 years (1995–1998) of the investigation period (after installation of TEIS). The set of parameters which has been transferred to the GIS (Table 2), has been selected with respect to both continuous availability and relevance for assessing hygienicmicrobial risks within the drinking-water supply system. Total coliforms and E. coli may be interpreted as direct indicators for faecal contamination. Colony forming units (CFU) serve as indicators for effectiveness of water purification and of microbial re-growth beyond water treatment, whereas nitrate can be seen as an indirect indicator for human activities within the catchment areas (WHO, 1996). To assess microbial health risks related to consumption of drinking water a wide range of information is needed. Much of this information has a spatial dimension. A Water Supply Structure GIS (WSS-GIS) module has been built up to support the handling and utilisation of available quantitative data concerning the water supply system, i. e. drinking-water origin, water treatment procedures, water disinfection and water distribution. Parameters reflecting the hygienic quality of raw and finished water were attached as attribute data. Data about the number of supplied individuals per water company was included to assess the size of populations at risk. It saves manpower to integrate routinely collected data sets which are suitable to the GIS. For that reason CCPs which were included in the existing databases (HYGRIS, TEIS) were adopted and transferred to the WSS-GIS module. The available data consists of direct or indirect spatial information.

303

Table 2. Raw water and drinking-water parameters which have been integrated into the WSS-GIS module. Indicator

number of raw water samples (1991–1997)

number of drinking water samples (1974–1998)

total coliform bacteria Escherichia coli colony forming units [CFU, 20°C] colony forming units [CFU, 36°C] nitrate

240 – 234 – 267

3075 4457 4571 4525 683

Geographical co-ordinates represent direct spatial information, whereas street addresses are hidden or indirect spatial information. GIS utilises these spatial information (Bill, 1999). With regard to the various data formats, a preliminary data conversion plan was worked out and was strictly followed during the integration process (Table 3; see Cannistra, 1999). Since feature type restricts data analysis it was important to choose the appropriate feature. The conversion process was based on a geo-referenced topographical map in 1 : 25,000 scale (geographical co-ordinates). Administrative boundaries of the investigation site were provided in digital format by the district authority. Location and size of drinking-water companies and private wells were ascertained by inspection of records at the Public Health Department. More detailed information about the drinkingwater supply structures was derived from a questionnaire which was sent out to and answered by the waterworks operators. Data concerning the public distribution grid was digitised manually from maps of the Public Health Department and of the water supply companies. Geographical co-ordinates of raw water sample points and water treatment facilities were available from HYGRIS. Most TEIS sample points had to be digitised manually as their geographical co-ordinates were not available from the database.

Table 3. Data sources and data conversion plan. information

data source

conversion

feature

topography

topographic map (1:25,000 scale)

scanned map

image

administrative boundaries

street registry of Rhein-Berg District

ALK-ATKIS interface EDBS

polygon

supply areas

maps of local health department and own investigations

digitising

polygon

distribution grid

maps of water supply companies

digitising

line

raw water sampling points

HYGRIS query

use of coordinates

point

water purification facility sampling points

TEIS query

use of coordinates

point

feeding points

TEIS query

digitising

point

grid sampling points

TEIS query

digitising

point

water storage tank sampling points

TEIS query

digitising

point

private wells (supply for 1 family)

TEIS query

digitising

point

private wells (supply 2 to 15 families)

TEIS query

digitising

point

304

T. Kistemann et al.

Information regarding specific water supply structures together with the HYGRIS and TEIS databases constitutes two functional units of the WSS-GIS module. The different object groups are represented as layers in the WSS-GIS. The waterquality data were stored in a relational database. Structure and function of the WSS-GIS module is shown in Fig. 1.

Results The resulting WSS-GIS module comprises the following CCP classes: – – – – – – –

well-heads, water purification facilities, disinfection points, drinking-water storage tanks, grid feeding points, distribution grid, private wells.

Each class of CCPs as well as supply areas and administrative boundaries constitute separate layers within the WSS-GIS. In addition to store, analyse, map, change and update drinking-water data the WSS-GIS allows

Fig. 1. Concept of the water-supply structure (WSS) GIS module.

interactive screen queries. By clicking on a sample point of any CCP layer a window appears displaying any available information on that point. Objects may be selected by logical queries combining several conditions. A query can, for instance, simultaneously be limited to a specific category of sample points, to the exceeding of a legislative limit of a hygienic parameter within a defined time period, and to a defined area (Fig. 2). By utilising the assigned spatial information the query results can be displayed in different output formats. Other information related to these objects can be included into the output. By such GIS-specific options, spatial patterns within the integrated raw and drinking-water data may easily be uncovered. Spatial distributions of CCPs with specific qualities can be shown in dot maps as well as in choropleth maps, related to municipalities or supply areas, for instance. The application of the WSS-GIS allows a rapid visualisation and analysis of the drinking-water supply structure and drinking-water flows between the water companies. It enables the user to produce a huge amount of result output. Only few examples can be demonstrated here.

GIS-based analysis of drinking-water supply structures

Approximately 68 % of the area is supplied with drinking-water originating from surface-water, the remaining 32 % is provided with drinking-water stemming from purified ground water (Fig. 3). But reflecting the unequal population density of the administrative units, only half of the population is supplied with drinking water Abstracted from surface water. The other half of the population receives ground water. In the Rhein-Berg District, 22 drinking-water companies distribute at least 5,000 m3 of drinking-water per year to the consumers (Fig. 3). In the South, the drinking-water supply structures coincide for the most part with the municipality boundaries. In the North, a lot of small water providers reflect the persistence of historical drinking-water supply structures. The city of Leichlingen (approx. 26,000 inhabitants) is provided by nine water companies of which only two serve more than 50,000 m3 per year. As an exception for the North of the district, one small water company (WVG Heddinghofen) distributes mainly selfabstracted ground water. It is situated in the city of Burscheid. Within the district, the number of water distributing companies exceeds the number of water producing companies substantially. Due to this, there exists a complex system of drinking-water flows between abstracting and purifying waterworks (with or without own distribution grid) and non-producing water supply companies (Fig. 4). These interrelationships are the predominant pattern within the management of drinking water abstracted from surface-water (dams), whereas ground water is Abstracted by the distributing companies themselves to a large extent. Three out of four surface-water purification facilities (two being situated outside the district) produce drinking-water which is distributed to the consumers by 18 water supply companies via their distribution systems. All the supply companies which distribute treated surface-water have to buy their total drinkingwater amount. One surface-water purification facility (Fernwasserversorgung Große Dhünn-Talsperre) produces only drinking-water for remote water supply outside the Rhein-Berg District. The drinking-water production of four waterworks is based on ground water. One waterwork is situated outside the Rhein-Berg District (RGW, Cologne) and sells a small amount of its total production to BELKAW (Bergische Licht-, Kraft- und Wasserwerke GmbH). Mainly, BELKAW distributes self-abstracted ground water. The waterworks of Rösrath and Bechen (municipality of Kürten) distribute only ground water abstracted by themselves. The surface-water treatment process comprises sand filtration and continuous disinfection for all sur-

305

· · ·

Fig. 2. Query result of the WSS-GIS module.

face waterworks, whereas sand filtration is not applied for ground water purification. For other than microbial purposes, a gravel filter is installed at the Rösrath waterworks. Continuous ground water disinfection is only driven by the Bechen waterworks (Figure 4). There exists a very low number of private wells within all cities and municipalities. From 0.01 % up to 0.9 % of the municipalities’ inhabitants are supplied by private wells. The most rural municipality with the lowest population density (Kürten) has the highest amount of inhabitants being supplied by private wells.

Discussion Microbial contamination of drinking water in regions with a high population density and highly developed drinking-water supply may occur for different reasons. Highly microbial polluted raw water can obstruct the purification process. Recontamination of the purified drinking-water in storage and distribution facilities can occur due to bio-films, cracks, animals and sewer pipe leakage. During the last years, waterborne outbreaks of faecal origin have occurred in many countries with high-

306

T. Kistemann et al.

BELKAW

Bergische Licht-, Kraft- und Wasserwerke GmbH (Bergisch Gladbach)

FWV

Fernwasserversorgung

GW

Gemeindewerk

GWW

Gemeindewasserwerk

RGW

Rheinische Gas- und Wasserversorgung Aktiengesellschaft (Kšln)

WBV

Wasserbeschaffungsverband

WTV

Wahnbachtalsperrenverband (Siegburg)

WVV

Wasserversorgungsverband

Source:

Survey of the Environmental Agency of North Rhine-Westphalia (1996)

Fig. 3. Waterworks and drinking-water delivery in the Rhein-Berg District (North Rhine-Westphalia, Germany).

ly developed supply structures. The highest number of infected individuals (400,000) was reported for Milwaukee, Wisconsin, USA in 1993 (Kramer et al., 1996). In Germany, high standards of the purification processes minimise the threat of such a water-related disease outbreak, but it may nevertheless be possible under unfavourable circumstances. Contamination of drinking-water by wastewater can be a hazard in highly developed water supply situations. In Sweden a giardiasis outbreak was caused by tree roots growing into and fracturing a sewer system. The wastewater flowed into the water purification facility, was abstracted and distributed via the distribution grid (Neringer et al., 1987). Due to significant social, economic and demographic changes, the German health care system is assumed to require strengthening of preventive measures (health protection, health promotion). Information which is needed for political decisions concerning public health care have to be made available more readily to the public by health reporting on local, state and federal level (Statistisches Bundesamt, 1998).

Our study proved that GIS-supported recording, visualization and analysis of drinking-water supply structures is feasible on the basis of already existing databases. Thus, the WSS-GIS module is able to give useful support to emerging public health issues (i. e., interpretation of surveillance data, detection, management and prevention of incidents or outbreaks). Its outputs are ready for linkage with epidemiological data by combining the WSS-GIS module with an additional health data module and thus can be used as an input for basic health reports. Access to epidemiological data, however, may be a major obstacle in Germany. It would be much easier to run epidemiological studies, if smallarea based, anonymous health data would routinely be provided and made available for public health research issues (Beske and Hallauer, 1993). An incident is defined as an event which can have bad influence on human health. Incident management plans establish standard operating procedures for the event of water supply incidents. The target of such plans is to identify an incident quickly, to assess its consequences rapidly, to manage the crisis, to restore the water quality and to pass information to the public (Exner et al. 1999). Incorporating GIS at water supply companies and public health departments is expected to provide huge benefits in managing incidents. The demands which are given within the framework of incident management plans can be optimised using GIS (Weckenbrock, 1999; Exner and Kistemann, forthcoming). In the detection stage of an incident management process, a hygienic parameter may exceed a critical limit. Alternatively, clusters of diarrhoea notifications or consumers’ complaints may occur. It is easy to recognise spatial agglomerations of complaints or disease cases using GIS (NWW 1999). During the assessment stage, GIS provides decision makers with information on origin of distributed water, physical conditions of catchment areas, water treatment procedures, distribution structures and customers. On base of this information the number, location and health status of individuals at risk can be assessed to decide which short-term precautionary measures have to be taken. GIS supports the identification of causes (investigation stage) by providing detailed information on land use and physical conditions in the catchment areas (Kistemann et al., 2001) as well as information on distribution systems. Epidemiological outbreak investigations can effectively be supported by GIS tools (Eng et al., 1999). After normalization (stage 4), GIS supports concluding epidemiological investigations (Briggs and Elliott, 1995, Clarke et al., 1996, WHO, 1999 a) and developing consecutive long-term precautionary measures within the final stage of incident management (analysis, evaluation, prevention).

GIS-based analysis of drinking-water supply structures

Drinking-water companies above all implemented GIS to manage catchment areas and to administer distribution systems. Although it is both expensive and time consuming, GIS came up to link and store different data with spatial relation. These opportunities seemed to be the biggest advantages of GIS regarding water suppliers’ needs (Briechle and Bucher, 1998). British North West Water (NWW), who also uses GIS for these purposes, additionally applied their system to investigations on incidents and outbreaks. On base of

307

GIS they developed an “Event Based Reasoning System”. An incident is realised by the system automatically whenever a fixed and area-related limit of consumers’ complaints is exceeded (NWW, 1999). In Imo State, Nigeria, a GIS-based computer modelling approach called probabilistic layer analysis was used to assess the risk of diarrheal disease for children using different water sources (Njemanze et al., 1999). With respect to the global climate change, changes in the distribution patterns of waterborne diseases

Sources:

Public Health Department of Rhein-Berg District (1998) and Environmental Agency of North Rhine-Westphalia (1996)

Fig. 4. Water supply areas and sources of drinking-water distributed in the Rhein-Berg District (North Rhine-Westphalia, Germany).

308

T. Kistemann et al.

have to be expected. Populations with restricted access to water in the home will be vulnerable to any climate related decreases in freshwater availability. The anticipated increase in extreme rainfall events, which may be associated with outbreaks of cryptosporidiosis, could pose a major challenge even for high quality water treatment services (Kovats et al., 1999). To address these problems of water-related ill-health, the need to establish co-operations and international networks has been identified. To meet that need, GIS-based health information systems are recommended to be developed (WBGU, 1998). At the Second European Conference on Environment and Health (Helsinki, Finland 1994) environment and health indicators as well as health and environment GIS were identified to be high-priority topics of research. Water quality and drinking water have been formulated to be specific health-related research areas (WHO, 1998). Concerning the use of transboundary watercourses and international lakes, a Draft Protocol on Water and Health was accepted by the Third Ministerial Conference on Environment and Health. “The objective of this protocol is the protection of human health and well being, within a framework of sustainable development, through improving water management, including the protection of water ecosystem, and through preventing, controlling and reducing water-related disease.” (WHO, 1999 b). The development of integrated information systems to handle information and provide authorities with successful solutions is explicitly encouraged. The demand of national and international health and environment policy agencies to establish research networks and public databases confronts health authorities and water-supply companies with new tasks. Effective data storage and evaluation with respect to their spatial relation have to be achieved. Developing integrated management systems to monitor and organise drinking-water supply is essential to support national and international databases corresponding with EC legal demands (EC, 1998) and WHO recommendations (WHO, 1993, 1997). GIS is a dynamic system which matches these demands by allowing to process and update huge amounts of geo-referenced data with spatial information. GIS produces powerful, easy-to-use and easy-to-understand planning and analysis information systems (Sweeny, 1999). The implementation of the database, however, is very expensive and time consuming. Substantial savings can be realised by integrating existing digital databases. This requires access to test results of raw and drinking-water, to data on supply structure, administrative boundaries, street registry and population data in digital form. Additionally, data quality has to be ensured by several cleaning procedures: checking, con-

verting, reformatting, correcting and editing of data to remove gaps and to eliminate or resolve errors either contained in the original data or introduced during data capture (Briggs and Elliott, 1995).

Conclusion A set of GIS modules which comprises data on drinkingwater supply structures (catchment area, abstraction, purification, disinfection, distribution grid), population consuming the tap-water and administrative organization provides a useful basis for MRA, incident and outbreak management and epidemiological investigations. GIS-supported studies provide more spatio-analytical options (i. e. aggregation, buffer and overlay) than conventional environmental and epidemiological studies do. Increasing requests concerning health protection and health reporting on the national level, the legislative demand for a better preparedness for incident and outbreak management on the EC level (EC, 1998), and water-related health impacts of global climate change are major challenges of future water management. Place is central to these problems. Therefore, the problems need to be tackled with methods including spatial analysis. GIS is assumed to be a very useful and easyto-use tool to meet these requirements. As demonstrated, the database for a Water-Supply Structure GIS module in principle is available in North Rhine-Westphalia. HYGRIS and TEIS turned out to be suitable databases for GIS application. Implementation of a WSS-GIS module offers huge advantages concerning microbial monitoring of raw and drinking water as well as recognition and investigation of incidents and outbreaks. It may be expected that experiences with GIS in the field of water hygiene and health would also be suitable for transfer to regions with less developed drinking-water supply structures. Acknowledgements. The authors wish to thank the Public Health Department of the Rhein-Berg District, especially Dr. Norbert Petruschke, who substantially supported this study throughout the entire project period. We sincerely thank all involved water-supply companies as well as the Environmental Agency of North Rhine-Westphalia for providing this study with data. This project was partly funded by a grant from the Association for Drinking-Water Reservoirs (Arbeitsgemeinschaft Trinkwassertalsperren e.V.).

References Beske, F., Hallauer, J.: Das Gesundheitswesen in Deutschland. Deutscher Ärzteverlag, Köln (1993). BGW (Bundesverband der deutschen Gas- und Wasserwirtschaft e.V.) (Ed.): 107. Wasserstatistik – Entwicklung der öffentlichen Wasserversorgung 1990–1995. Bonn (1996).

GIS-based analysis of drinking-water supply structures Bill, R.: Grundlagen der Geo-Informationssysteme. Band 1. Hardware, Software und Daten. Wichmann, Heidelberg (19994). Briechle, D., Bucher, B.: Geografisches Informationssystem – ein Instrument für die Wasserversorgung. gwf – Das Gasund Wasserfach – Wasser Special, 139, 13: 10–14 (1998). Briggs, D. J., Elliott, P.: The use of geographical information systems in studies on environment and health. World Health Stat. Quaterly 48: 85–94 (1995). Cannistra, J. R.: Converting utility data for GIS. Journal of the American Water Works Association 91, 2: 55–64 (1999). Clarke, K. C., McLafferty, S. L., Tempalski, B. J.: On epidemiology and geographic information systems: a review and discussion of future directions. Emerging Infectious Diseases 2: 85–92 (1996). EC (European Community): Richtlinie 98/83/EG des Rates vom 3. November 1998 über die Qualität von Wasser für den menschlichen Gebrauch. Amtsblatt der Europäischen Gemeinschaften: L 330/32–54 (1998). Eng, S. B., Werker D. H., King, A. S., Marion, S. A., Bell, A., Isaac-Renton, J. L., Irwin, G. S., Bowies, W. R.: Computer-generated dot maps as an epidemiologic tool: investigating an outbreak of toxoplasmosis. Emerging Infectious Diseases 5: 815–819 (1999). Exner, M., Kistemann, T.: Strukturelle Voraussetzungen und Maßnahmen zur Kontrolle der Weiterverbreitung übertragbarer Krankheiten durch Wasser für den menschlichen Gebrauch. Maßnahmepläne und Störfallmanagement. Hässelbarth, U., Grohmann, A., Schwerdtfeger, W. (Ed.): Die Trinkwasserverordnung. Berlin (forthcoming). Exner, M., Lacombe, M., Lafontaine, J.: Störfall-Leitplanung Trinkwasserversorgung. Ministerium für Frauen, Jugend, Familie und Gesundheit des Landes Nordrhein-Westfalen. Düsseldorf (1999). Gale, P.: Developments in microbiological risk assessment models for drinking water – a short review. In: Journal of Applied Bacteriology, 81, 401–410 (1996). Haas, C. N., Rose, J. B., Gerba, C. P.: Quantitative microbial risk assessment. New York, Chichester, Weinheim (1999). Havelaar, A. H.: Application of HACCP to drinking water supply. Food control, 5, 3: 145–152 (1994). Kistemann, T.: Trinkwasserinfektionen – Risiken in hochentwickelten Versorgungsstrukturen. Geographische Rundschau, 49, 4: 210–215 (1997). Kistemann, T., Exner, M.: Water quality and health risks. In: Ehlers, E., Krafft, T. (Ed.): Understanding the Earth System: Compartments, Processes and Interactions. Springer, Heidelberg: 209–221 (2001). Kistemann, T., Dangendorf, F., Exner, M.: A geographical Information System (GIS) as a tool for microbial risk assessment in catchment areas of drinking water reservoirs. International Journal of Hygiene and Environmental Health 203, 3: 225–233 (2001). Kovats, R. S., Haines, A., Stanwell-Smith, R., Martens, P., Menne, B., Bertollini, R.: Climate change and human health in Europe. British Medical Journal, 318: 1682– 1685 (1999). Kramer, M. H., Quade, G., Hartemann, P., Exner, M.: Waterborne diseases in Europe – 1986–96. Journal of the American Water Works Association 93: 48–53 (2001).

309

Kramer, M. H., Herwaldt, B. L., Craun, G. F., Calderon, R. L., Juranek, D. D.: Waterborne disease: 1993 and 1994. Journal of the American Water Works Association 88, 3: 66– 80 (1996). Lack, T.: Water and health in Europe: an overview. British Medical Journal, 318: 1678–1682 (1999). Lacombe, M., Fehr, R.: Development of a comprehensive drinking water surveillance system. Fehr, R., Berger, J., Ranft, U. (Ed.): Environmental Health Surveillance. Results of an international workshop, 11–12 March 1997 at the University of Bielefeld. ecomed, Landsberg: 135–146 (1999). MAGS (Ministerium für Arbeit, Gesundheit und Soziales NRW): TEIS – Trinkwassererfassungs- und Informationssystem, Version 1.0. Ein System zur Erfassung, Darstellung, Auswertung und Weiterleitung von Trinkwassergütedaten für Gesundheitsämter. Rheinisch-Westfälisches Institut für Wasserchemie und Wassertechnologie, Mülheim (1994). Neringer, R., Andersson, Y., Eitrem, R.: A Water-borne Outbreak of Giardiasis in Sweden. In: Scandinavian Journal of Infectious Diseases 19: 85–90 (1987). Njemanze, P. C., Anozie, J., Ihenacho, J. O., Russell, M. J., Uwaeziozi, A. B.: Application of risk analysis and geographic information system technologies to the prevention of diarrheal diseases in Nigeria. American Journal of Tropical Medicine and Hygiene 61: 356–360 (1999). NWW (North West Water): The use of Geographical Information Systems (GIS) within NWW. Water & Health 27: 6–10 (1999). Statistisches Bundesamt (Ed.): Gesundheitsbericht für Deutschland. Stuttgart (1998). Sweeney, M. W.: Geographic Information Systems. In: Water Environment Research, 71, 5: 551–556 (1999). Trinkwasserverordnung (Verordnung über Trinkwasser und über Wasser für Lebensmittelbetriebe – TrinkwV) in der Fassung der Bekanntmachung vom 5. Dezember 1990 Bundesgesetzblatt Teil I, Nr. 66: 2613–2629 (1990). Untermann, F.: Risiko-Bewertung und Risikomanagement nach dem HACCP-Konzept: Ein Weg zu sicheren Lebensmitteln. Zentralblatt für Hygiene und Umweltmedizin, 199, 2–4: 119–130 (1996). WBGU (Wissenschaftlicher Beirat der Bundesregierung Globale Umweltveränderungen): Welt im Wandel – Wege zu einem nachhaltigen Umgang mit Süßwasser. Springer, Berlin, Heidelberg, New York (1998). Weckenbrock, P.: Bedeutung Geographischer Informationssysteme in unternehmensweiten Geschäftsprozessen. gwf – Das Gas- und Wasserfach – Wasser Special, 140, 13: 162–166 (1999). WHO (World Health Organization): Guidelines for drinking-water quality, Vol.1: Recommendations. WHO, Geneva (19932). WHO (World Health Organization): Guidelines for drinking-water quality. Vol. 2: Health criteria and other supporting information. WHO, Geneva (19962). WHO (World Health Organization): Guidelines for drinking-water quality. Vol. 3: Surveillance and control of community supplies. WHO, Geneva (1997).

310

T. Kistemann et al.

WHO (World Health Organization): Environment and health research for Europe. http://www.who.dk/london99/researchg.htm (1998). WHO (World Health Organization): Geographical information systems (GIS). Weekly Epidemiological Record 74: 281–285(1999a).

WHO (World Health Organization): Protocol on Water and Health to the 1992 Convention on the Protection and Use of Transboundary Watercourses and International Lakes. Adopted on 17 June 1999 at the Third Ministerial Conference on Environment and Health. http://www.who.dk/ london99/water02g.htm (1999b).