Distinguishing Sources of Faecal Pollution in Australian Inland and Coastal Waters using Sterol Biomarkers and Microbial Faecal Indicators
This report was produced for the Urban Water Research Association of Australia, a now discontinued research program.
Report no. UWRAA 204
The project aims were to: (i) estimate the numbers of faecal indicator microorganisms (bacteria and bacteriophages of Bacteroides fragilis) and concentrations of sterols present in the faeces of native and domestic animals and in sewage to validate whether profile differences measured in pilot studies could be used to quantify sources of faecal contamination, (ii) determine the relationship between, and relative persistence of, faecal sterols and indicator microorganisms during microcosm field experiments and (iii) develop a quantitative method which could distinguish how much faecal matter is present from various sources in a polluted receiving water and test the technique in the field.
The results for objective (i) indicate that the differences in faecal sterols and bacterial indicators between the animal groups were consistent over a range of environments. The concentrations of faecal sterols and bacterial indicators measured can be variable, but the ratios of the parameters, on the basis of populations, are consistent within animal groups and compared to reference samples. Reference samples were collected during field studies from sewer mains or overflows, livestock sale yards and bird resting sites adjacent to field sites and results from these samples provided corroborating in situ ratios to estimate faecal contributions to specific environments. The independent use of at least two and preferably four subgroups of bacterial indicators (thermotolerant coliforms, Escherichia coli, faecal streptococci and enterococci) as a common denominator of all faecal pollution, strengthens the degree of confidence in the estimates. Estimates from calculations performed for field studies, where the wider range of bacterial indicators was used, generally varied by less than 10%. The use of a range of ratio values based on the variance measured in the faecal and field samples enables a range of contributions to be expressed. These are summarised in the legend of Figure4 in the report.
The bacteriophages to various strains of B.fragilis used to characterise human, pig or poultry, cattle and sheep faeces were often absent from the faeces of individual animals. Nonetheless, when bacteriophages were present in the faeces of cattle or pigs, they consistently contained only phages specific to their respective host strain, and human sewage always contained the bacteriophages to HSP40, the human-specific strain. As observed for the HSP40 strain in North America, the results indicated that only about10% of individuals excreted their respective bacteriophages, and if faecal contamination from cattle, pigs or humans was present, then up to 1000 plaque forming units of their specific phages per millilitre of effluent could be expected. Hence, given the likely low numbers of phages in environmental waters, enrichments for the specific phages probably will be needed to confirm the presence of cattle, pig or human faecal contamination.
With the use of any single or multiple indicator system, the efficacy of the result or relevance of the interpretation is a function of time and / or distance from the source. Results for objective (ii) indicated that faecal sterols degrade at the same rate as each other and most bacterial indicators die-off more rapidly than faecal sterols degrade, with C.perfringens spores being the exception. However, significant divergence of the ratios between faecal sterols and bacterial indicators are unlikely to occur within2-3 days. Under the environmental situations of field studies undertaken so far, this time constraint was generally not considered to have been exceeded. This was because the survey / catchment areas were not large and even low flow rates occurring in dry weather were estimated to transport faecal matter along the length of the catchments inside 2-3 days. During wet weather when faecal contamination is most prevalent, residence times are even shorter. A settling experiment was conducted to examine the particle size distribution for faecal sterols and bacterial indicators. The results indicated that only in the very fine particle range (equivalent settling velocity to 9 ф quartz particles) was there any significant increase in bacterial indicators. This means that only in standing waters would the ratio of bacterial indicators to faecal sterols increase.
Results from field studies showed the technique conclusively distinguished episodes of human faecal contamination. Faecal contamination from humans and herbivores such as cattle, sheep, kangaroos, etc. could be distinguished from each other with a high degree of accuracy. Based on the findings of this study, faecal contamination from dogs might also be evident, if present in significant amount, because of a distinctive faecal profile of high C. perfringens spores(≈ equal to thermotolerant coliforms) and high cholesterol concentrations. However, in field studies performed during the period of this study, dogs were not considered to be a significant source of faecal contamination, hence the efficacy of the technique to distinguish faecal contamination from dogs remains untested. In the absence currently of a specific marker for bird faeces, the faecal profile for birds is high abundances of thermotolerant coliforms and faecal streptococci, low abundances of C. perfringens spores and high concentrations of Δ5-sterolswith 5ß-stanols absent or in environmentally insignificant trace amounts. Other possible sources of faecal contamination which share a similar faecal profile are invertebrates, reptiles and fish. Because there are presently no specific biomarkers for faeces from these species, faecal contamination from these sources are designated “birds, dogs and other diffuse sources”. However, in several field studies, birds were proposed to be the dominant contributor to this grouping because they were ubiquitous and prolific.
In summary, the technique can, with a high degree of certainty, determine and distinguish human versus herbivore faecal contamination and these two sources versus other sources. The technique can then, with a diminished degree of precision, estimate the contribution from birds, dogs and other diffuse sources from total faecal contamination. The results from field studies show the technique is a great benefit to water managers for making effective management decisions.
The water industry has already responded positively to the concept of using additional indicators of faecal contamination to characterise and understand catchments. No one indicator or one approach could represent all the facets and issues associated with contamination of our waterways with faecal matter. The technique described herein has, as of publication, been used in over 20different environments for local councils, environment protection agencies and water authorities with consistent results. This technique is only a first step toward devising a set of tools that water managers will have at their disposal to investigate the entry of faecal contamination and other organic matter inputs into aquatic systems.
In light of the results of this study, it is recommended that further research be conducted in several complementary areas. Firstly, other classes of lipid biomarkers (e.g. cholanic bile acids and bile pigments) show distinctive profiles between animals. Additional research to discover specific biomarkers for birds and/or dogs would be highly desirable and appears achievable. Secondly, research on bacterial phages could provide qualitative confirmations of faecal sources to complement biomarker assays and to promote a diversity of approaches to solve current problems. Finally, combined PCR (polymerase chain reaction) and biomarker analyses of environmental samples could also provide an advantageous and diversified approach to distinguishing faecal pollution.