Report No FR0457


APRIL 1994



Performance of treatment processes for Cryptosporidium removal will assist in the selection, design and operation of plant for sites where Cryptosporidia are a potential problem, and will allow the performance of existing plant to be predicted.


To establish the performance of water treatment processes for Cryptosporidium oocyst removal, and the performance of ozone for oocyst inactivation, through pilot plant trials and laboratory experiments.


Outbreaks of cryptosopridiosis in the UK and USA, linked to water supplies, have led to concern over the ability of conventional water treatment to remove oocysts, regarding which there was very little available information. Ozone needed to be investigated as a disinfectant for Cryptosporidia, because of the ineffectiveness of chlorine for this application.


Pilot plant trials have indicated that well-operated conventional water treatment processes are capable of achieving better than 99% removal of oocysts, which would reduce concentrations in final waters to non-detectable or barely detectable levels for the maximum concentrations found in most raw water sources. For example, in 32 experiments with oocyst concentrations of between 300 and 800 oocysts per litre dosed to the feed water, concentrations in the filtered water were non-detectable for 50% of the time, and were detected at less than 0.1 per litre for 70% of the time. There were no indications of differences in performance between the different chemical coagulation based treatment streams used on the pilot plant.

At the doses normally used for water treatment, ozone did not appear to be very effective for oocyst inactivation, based on viability measurements using the excystment or staining techniques. In order to achieve a reduction in oocyst viability of over 90%, the ozone doses required would be well in excess of those currently used for water treatment, and would be associated with high operating costs. Results from other studies, in which viability was measured using animal infectivity, suggest ozone to be more effective. It is yet to be established which of the methods for viability assessment most closely models human infectivity.


The work has highlighted some of the areas of water treatment plant operation which require attention in order to achieve a consistently high degree of removal. Any measure which minimises filtered water turbidity will reduce the risk from Cryptosporidia, Care should therefore be taken to select the most suitable coagulants and chemical conditions, and ensure that these are maintained on the plant. The risk from higher than normal filtrate turbidity at the start of a rapid gravity filter run might be overcome by a slow start-up of the filters, for example with the filtration rate increased in four equal stages over the first hour of the run. However, more investigations are needed to confirm this. A reduction in efficiency for oocyst removal might occur without a corresponding increase in filtrate turbidity, and other indicators of possible reduced efficiency should be taken into account in plant monitoring. These would include increased true colour or dissolved metal ion coagulant concentration in the clarified or filtered water, or a sudden increase in clarified water turbidity.

Recycling of backwash water supernatant should not increase the risk to any significant extent, provided the backwash is well settled to give a good quality supernatant. Suitable targets would be a turbidity of less than 5 NTU or suspended solids of less than 10 mg/l. If necessary, polyelectrolyte could be used to achieve this quality supernatant. High rates of backwash supernatant return should also be avoided; a suitable target would be not more than 10% of the raw water flow.

Ozone should still be considered as part of a Cryptosporidium control strategy, when used in conjunction with conventional water treatment, offering a measure of protection against any oocysts which penetrate the treatment works.


The report describes laboratory and pilot plant experiments using water dosed with Cryptosporidium oocysts to investigate oocyst removal or inactivation by water treatment processes. The implications of the results of these experiments on water treatment plant design and operation are discussed.

Copies of the Report are available from FWR, price 25.00 less 20% to FWR Members