Drinking water supplies are often impacted by taste and odour (T/O) episodes caused by algal volatile organic compounds (AVOCs) from algal blooms. Treatment and control of these events is important to utility operators, as customer confidence in the safety of public drinking water supplies is based primarily on their palatability and odour. To manage T/O outbreaks successfully, knowledge about treatment responses of AVOCs and anticipation of their outbreaks are thus of major importance to the water industry. The Glenmore Reservoir and water treatment plant (GWTP) supplies drinking water to over 50% of the ca. 1 million consumers in Calgary (Alberta). Despite low nutrients and high raw water quality, the reservoir experiences periodic outbreaks of fishy/floral T/O, caused by chrysophytes and diatoms (Uroglena americana, Dinobryon spp., Synura petersenii, Asterionella formosa). These odours are produced by the unsaturated C7-C10 alkenes 2,4-heptadienal, 2,4,7-octatriene, 2,4-decadienal and 2,4,7-decatrienal, generated during from the enzymatic breakdown of algal polyunsaturated fatty acids (PUFAs). The formation, persistence and stability of these compounds in both the raw water and treatment plant is not well understood.

Algal metabolites have major impacts on water quality. Some of the more potent of these compounds can produce severe source-water odours furthermore, there is now growing evidence that some algal volatile organic compound,, (AVOCs) function in aquatic chemical ecology. However, the processes that influence the levels and dynamics of AVOC production are not well understood. There has been some success in linking to individual taxa, but it is often difficult to predict taste and odour (T/O) Outbreaks because algal communities show significant among- and within-species variation in AVOC production. Thus, while it is generally established that resources Such as nutrients and light govern the growth of many odour-causing freshwater algae, it is not clear how these resources influence algal odour production and whether this is simply via changes in standing stock or in per capita AVOC yield or chemistry. We examined this question for three bloom-forming chrysophytes, the two bacterivores Dinobryon cylindricum Imhof and Uroglena cf. americana Calkins, and the autotroph Mallomonas papillosa Harris & Bradley, Using data from several studies. These chrysophytes can produce severe rancid or fishy odours through the production of unsaturated fatty acid derivatives (2,4-heptadienal, 2.4-decadienal and 2,4,7-decatrienal), which differ considerably in potency [with odour threshold concentrations (OTCs) of 25, 0.3 and 1 mug l (1) respectively]. We measured in vitro growth, biomass, AVOC production and total odour yield (calculated from OTC-normalized AVOCs), under low and high P, N, Fe and light. The results showed significant resource-related changes in biomass and AVOC production and chemistry, which differed among species. For Mallomonas, biomass, AVOC and odour yields were all correlated closely with resource levels. For the two bacterivores, biomass was related strongly to light and P, but odour production varied between species and treatments. Dinobryon AVOC yield was related significantly to P and N, but not directly to light Or Fe; Uroglena AVOC yield responded to light and P. For the two bacterivores, AVOC production was not related clearly to grazing activity. The final total odour produced by each species did not necessarily reflect changes in their total biomass or AVOC mass yield because of variation in their relative and absolute per capita production of individual aldehydes, and because of the large differences in the relative potencies of these compounds. This implies that biomass reduction (e.g. via nutrient abatement) will not always reduce algal-derived T/O, because this is a function of changes at both the cell and Population levels. This is also important from an ecological viewpoint, because the potential roles of these AVOCs as chemical signals or allelogens is likely to be Coupled closely with the dynamics of their production.

Actinomycete isolations were made from water, suspended sediment, periphyton, macrophytes, and mussels in the Lake Ontario system. Isolated actinomycetes were tested for production of the taste and odor compounds geosmin and 2‐methyl‐isoborneol. Actinomycetes found in this system were associated with suspended sediment, indicating a terrestrial origin, or were associated with mussels, indicating that some species may reside in association with mussel beds. Actinomycetes in the genus Streptomyces and actinomycetes in other genera both produced taste and odors, but not all isolates produced taste and odors, and those that did, did not do so under all conditions. This research highlights the reasons for the lack of correlation between actinomycete isolations and taste and odor levels, and the need for further study on actinomycetes in relation to mussels.

1. Algal taste and odour is usually associated with open water blooms and eutrophic systems. However, some algal species can produce high biomass under ice‐cover, even at low nutrient concentrations, that can impact water quality. This paper describes a winter odour outbreak in oligotrophic Glenmore Reservoir (Calgary, Alberta, Canada), the major algal species, volatile organic compounds (VOCs) and some treatment implications.2. Using sensory, chemical and microscope analyses, we monitored odour, algal biomass and taxa, bacteria and major nutrients. In a preliminary assessment of the effectiveness of standard water treatment with this type of algal biomass and odour, we used bench‐scale tests and sampled raw water from the Glenmore treatment plant at successive treatment stages.3. In the winter of 1999–2000 Glenmore ice‐cover was delayed, nutrients were characteristically low (TP < ∼5 μg L–1), but organic carbon and bacteria were higher than in previous years.4. During this period there was an increase in algal biomass dominated by the mixotrophic chrysoflagellate Dinobryon divergens. Temporal dynamics of this species were inversely correlated with bacteria, and biomass declined following the establishment of ice‐cover, while depth profiles showed the highest abundance at subsurface layers. This suggested that the population outbreak was triggered by high bacteria abundance but depended on a minimum amount of light, consistent with in vitro studies of other mixotrophic chrysophytes.5. Other non‐bactiverous taxa were also numerous, notably Asterionella formosa, cryptomonads, dinoflagellates and the synurophyte Synura petersenii.6. Raw water odour was characteristically fishy, mainly caused by the VOCs 2,4,7‐decatrienal, 2,4‐heptadienal and 2,4‐decadienal. Based on algal population and VOC dynamics, these compounds were attributed to Dinobryon. Trace amounts of 2,6‐nonadienal (S. petersenii) and 1,3,5 and 1,3,6‐octatriene (A. formosa) were also detected. It was concluded that 2,4,7‐decatrienal was the major source of the raw water odour.7. Sensory and microscopic analyses of pre‐ and post‐treatment samples in the treatment plant indicated a complete removal of odour, but only a 30–60% removal of algal biomass and evident rupture of residual algal cells. Laboratory experiments showed that using standard treatment, chlorination rapidly oxidized 2,4,7‐decatrienal and 2,6‐nonadienal but had little effect on 2,4‐hepta‐ and decadienal.

Two volatile organic compounds (VOCs) responsible for the majority of taste and odour events in drinking water are geosmin (trans-1,10-dimethyl-trans-9-decalol) and MIB (2-methylisoborneol). These VOCs have odour threshold concentrations at ng l(-1) levels. Quantitative analysis of such trace concentration levels has required the use of technically complex and time-consuming analytical methods. In this study, a simple headspace solid phase microextraction procedure (HSPME) was developed and applied successfully to the qualitative and quantitative analysis of source water and drinking water samples. Concentrations of geosmin and MIB measured in these samples ranged from 1 to 1000 ng l(-1). The HSPME method achieved typical precision of 5-12% in a wide variety of sample matrices. Materials required were low cost and the protocol allowed sample turnover of 60-75 min per sample. HSPME proved to be a very practical technology for the analysis of geosmin, MIB and other potent volatile compounds in surface waters. (C) 2000 Elsevier Science Ltd. All rights reserved.

An efficient extraction method is needed to measure trace levels of taste and odour compounds in surface waters. This is usually accomplished by costly and involved analytical procedures. We have developed a simpler alternative, using a commercially available microextraction apparatus (SPME). With this technique we successfully monitored trace levels of some target organoleptics (unsaturated aldehydes e.g. heptadienal, nonadienal, and related compounds) which commonly cause aquatic taste and odour. We identified these compounds in culture material, and analyzed for them during the development of odourous chrysophyte blooms in two ponds. Preliminary work has also found a good recovery of some important off-flavour terpenoids (e.g. geosmin and MIB). SPME is labour and cost efficient, and therefore appealing to water treatment facilities for detection and monitoring. In addition, SPME requires only small sample volumes, and is therefore suitable for culture work.