Catchments are the first potential barrier to pathogen hazards in the water supply system. Reducing pathogen loads exported from catchments to drinking water reservoirs is thus an important priority in applying a risk-based approach to managing water supplies. Although predictive models are available to estimate sediment and nutrient loads, few models are available to predict either bacterial indicator or pathogen loads exported from catchments. This paper describes the application of a process-based mathematical model to predict pathogen (Cryptosporidium) and faecal indicator (E. coli) loads generated within and exported from the Sydney drinking water catchments. The model was derived from a conceptual model that identified key processes for microbial sources from animals, on-site systems and sewage treatment plants (STPs) and their subsequent transport within drinking water catchments (Ferguson et al. 2003). Inputs to the model include GIS land use and hydrologic data and catchment specific information. The model was initially applied to the Wingecarribee catchment in the Sydney drinking water catchment and a sensitivity analysis of the model was undertaken to determine components of the model that required further investigation (Ferguson et al. submitted). The model was then applied to all 27 individual catchments (and the 196 subcatchments) within the Sydney Catchment Authority (SCA) area of operations. The model predicts pathogen catchment budgets (PCB) and ranks the sub-catchments that generate the highest loads of pathogens and indicators (per km), as well as the sub-catchments that export the greatest load of pathogens to the downstream storages. Ranking the sub-catchments enables quick identification of those areas that are generating the highest pathogen and indicator loads facilitating the implementation of control measures. The outputs from the model show that in dry weather the highest daily loads of Cryptosporidium were predicted to be generated in Kellys Creek and Mittagong Creek subcatchments in the Wingecarribee catchment. These sub-catchments are heavily impacted by the effluent discharged from Bowral and Moss Vale STPs, respectively. However, in wet weather the wash off of faecal material into surface runoff predicts that large loads of Cryptosporidium are generated in sub-catchments dominated by improved pasture grazed by cattle. The slow decay of protozoan pathogens combined with their rapid transport in water during wet weather events results in a cumulative export of Cryptosporidium to downstream sub-catchments. For example, the PCB model predicts that Warragamba reservoir would receive 4 x 10 Cryptosporidium oocysts following a 100 mm in 24 h rainfall event in the Sydney catchment. The model predicts that in dry weather approximately 1 x 10 E. coli per day were generated in subcatchments that contain improved pasture with agricultural livestock with additional inputs from sub-catchments receiving STP effluent. The rapid die-off and limited transport of this microorganism in dry weather results in fairly localized impacts. However in wet weather significant loads of faecal indicator bacteria are mobilised to the stream network and transported to downstream sub-catchments with Warragamba reservoir and the Lower Wollondilly predicted to receive up to 5.4 x 10 E. coli following a 100 mm in 24 h rain event in the Sydney catchment. The pathogen and indicator wet weather export loads predicted by the PCB model can be used as input variables to the hydrodynamic reservoir model developed by Hipsey et al. (2005) thus enabling the estimation of the risk of their subsequent transport to the water storage offtake point in Warragamba Reservoir.

There is uncertainty whether Cryptosporidium oocysts attach to particles or to each other under ambient water conditions. Particle size distributions of Cryptosporidium oocyst suspensions were determined over a range of ionic strengths and pHs to determine under those environmental conditions that may promote oocyst aggregation. Cryptosporidium oocysts were shown to only aggregate in high ionic strength solutions (>0.45 M) and remain largely as single entities at ionic strengths and pHs that were likely to be encountered in surface runoff. Similarly, in loam soil suspensions, rather than attaching to the soil particles the majority of oocysts also remained as single entities. Overall, oocysts are expected to remain largely unattached to either themselves or soil particles in overland runoff. This has implications for pathogen transport and modelling since oocysts that are freely suspended are more likely to be transported in runoff to surface waters than if attached to more dense soil/faecal particles.

This paper describes the development and testing of a mathematical model as a tool to quantify pathogen loads in Sydney's drinking water catchments. It has been used to identify, quantify and prioritise sources of Cryptosporidium, Giardia and E. coli in the Wingecarribee catchment. The pathogen model promotes understanding of the relative significance of different sources of pathogen risks as well as their fate and transport as they move through the subcatchments. This pathogen model not only enables water utility managers to identify those catchment segments that may contribute the highest load of pathogens, but also where management options will be most effective.

AIMS:To generate field-relevant inactivation data for incorporation into models to predict the likelihood of viral contamination of surface waters by septic seepage.METHODS AND RESULTS:Inactivation rates were determined for PRD1 bacteriophage and Adenovirus 2 in two catchment soils under a range of temperature, moisture and biotic status regimes. Inactivation rates presented for both viruses were significantly different at different temperatures and in different soil types (alpha = 0.05). Soil moisture generally did not significantly affect virus inactivation rate. Biotic status significantly affected inactivation rates of PRD1 in the loam soil but not the clay-loam soil. Adenovirus 2 was inactivated more rapidly in the loam soil than PRD1 bacteriophage.CONCLUSIONS:Virus inactivation rates incorporated into models should be appropriate for the climate/catchment in question with particular regard to soil type and temperature. Given that PRD1 is similar in size to adenoviruses, yet more conservative with regard to inactivation in soil, it may be a useful surrogate in studies of Adenovirus fate and transport.SIGNIFICANCE AND IMPACT OF THE STUDY:A better understanding of the factors that govern virus fate and transport in catchments would facilitate the design of barrier measures to prevent viral contamination of surface waters by septic seepage.

Aims: To generate field-relevant inactivation rates for Cryptosporidium oocysts in soil that may serve as parameter values in models to predict the terrestrial fate and transport of oocysts in catchments.Methods and Results: The inactivation of Cryptosporidium oocysts in closed soil microcosms over time was monitored using fluorescence in situ hybridization (FISH) as an estimate of oocyst 'viability'. Inactivation rates for Cryptosporidium in two soils were determined under a range of temperature, moisture and biotic status regimes. Temperature and soil type emerged as significantly influential factors (P < 0.05) for Cryptosporidium inactivation. In particular, temperatures as high as 35degreesC may result in enhanced inactivation.Conclusions: When modelling the fate of Cryptosporidium oocysts in catchment soils, the use of inactivation rates that are appropriate for the specific catchment climate and soil types is essential. FISH was considered cost-effective and appropriate for determining oocyst inactivation rates in soil.Significance and Impact of the Study: Previous models for predicting the fate of pathogens in catchments have either made nonvalidated assumptions regarding inactivation of Cryptosporidium in the terrestrial environment or have not considered it at all. Field-relevant inactivation data are presented, with significant implications for the management of catchments in warm temperate and tropical environments.

ABSTRACT The dispersion and initial transport of Cryptosporidium oocysts from fecal pats were investigated during artificial rainfall events on intact soil blocks (1,500 by 900 by 300 mm). Rainfall events of 55 mm h −1 for 30 min and 25 mm h −1 for 180 min were applied to soil plots with artificial fecal pats seeded with approximately 10 7 oocysts. The soil plots were divided in two, with one side devoid of vegetation and the other left with natural vegetation cover. Each combination of event intensity and duration, vegetation status, and degree of slope (5° and 10°) was evaluated twice. Generally, a fivefold increase ( P < 0.05) in runoff volume was generated on bare soil compared to vegetated soil, and significantly more infiltration, although highly variable, occurred through the vegetated soil blocks ( P < 0.05). Runoff volume, event conditions (intensity and duration), vegetation status, degree of slope, and their interactions significantly affected the load of oocysts in the runoff. Surface runoff transported from 10 0.2 oocysts from vegetated loam soil (25-mm h −1 , 180-min event on 10° slope) to up to 10 4.5 oocysts from unvegetated soil (55-mm h −1 , 30-min event on 10° slope) over a 1-m distance. Surface soil samples downhill of the fecal pat contained significantly higher concentrations of oocysts on devegetated blocks than on vegetated blocks. Based on these results, there is a need to account for surface soil vegetation coverage as well as slope and rainfall runoff in future assessments of Cryptosporidium transport and when managing pathogen loads from stock grazing near streams within drinking water watersheds.

A methodology has been developed to apply the materials budget concept, used in sediment and nutrient studies, to construct a pathogen budget for drinking water catchments, taking into consideration pathogen origin, deposition, inactivation and movement within a catchment. These processes can be described in terms of stocks (pathogens) and flows (movement of stocks). In south-eastern Australia, the majority of pathogen loading to major tributaries was predicted to occur during and after high intensity rainfall events where in-stream resuspension was not of great relative importance. In contrast, during dry weather the transit time within the studied catchment was sufficiently long that in-stream processes became relatively important. Total pathogen unit (TPU) budgets were constructed for the parasitic protozoa Cryptosporidium and Giardia. This approach enables water utility managers to identify those catchment segments and processes that are contributing or removing the greatest load of pathogens, and thus where management options will be most effective. With improved knowledge of pathogen ecology this approach can be further refined to provide budgets of infectious pathogen units (IPU), more directed to public health risk endpoints.

This study examined water and sediment samples for a range of indicator and pathogenic microorganisms from six sites in an urban estuary, Sydney, Australia. Water quality was affected by rainfall and sewage overflows which were associated with significant increases in the concentration of faecal coliforms, faecal streptococci, Clostridium perfringens spores, F-RNA bacteriophage, Aeromonas spp., Giardia and Cryptosporidium spp. However, in sediments, only faecal coliform concentrations were significantly increased by rainfall, although sewage overflow again resulted in increased concentrations of faecal coliforms, faecal streptococci, C. perfringens spores and Aeromonas. Isolation of Salmonella appeared to coincide with wet weather events and occasionally identical serotypes were detected in sediments at several locations within the estuary. However, isolations of enteric virus were sporadic and did not appear to be exclusively related to wet weather events. C. perfringens was identified as the most useful indicator of faecal pollution and was the only indicator significantly correlated to the presence of pathogenic Giardia (r = 0.41, p < 0.05) and the opportunistic bacterial genus Aeromonas (r = 0.39, p < 0.05). F-RNA bacteriophage was not significantly correlated with any of the pathogens examined.