Faecal indicator bacteria have traditionally been used in the detection of faecal pollution in water, but due to concerns about the lack of reliability of these indicators, alternative methods have been developed. One of which is the detection of sterols present in human and animal excreta via GC-MS analysis of water in this study. The Szűcs method was used to detect six target sterols (coprostanol, cholesterol, dehydrocholesterol, stigmasterol, β-sitosterol, and stigmastanol) in environmental water samples. An initial study was done by analysing raw sewage and effluent (human faecal sterol biomarkers) and water samples were spiked with excreta from cattle, chickens, horses, pigs, and sheep to determine faecal sterol fingerprints. The method was evaluated for quantitation and differences between the water samples from each species. Following liquid-liquid extraction, silylation and derivatization, samples were analysed by GC-MS. Standard curve assays were linear up to 160ng and the limit for quantification was 20ng. The human faecal sterol biomarker was coprostanol, while herbivore profiles were dominated by terrestrial sterol biomarkers (stigmasterol and stigmastanol). Sterol fingerprints and differences in concentrations of sterols between various animals and between animals and humans occurred, providing the opportunity to determine whether faecal pollution was from humans or from animals. The method proved sensitive enough to evaluate faecal contamination in environmental water. Groundwater was collected from bore-holes and surface water samples were collected from the Baberspan Inland Lake. Physico-chemical parameters analysed indicated that pH for surface water samples was above 6.9. The total dissolved solids (TDS) in groundwater indicated that the water was not suitable for human consumption, but could be used for livestock watering. Surface water electrical conductivity (EC) and inorganic nitrates was too high to be used for irrigational purposes. Nitrates in groundwater were too high to be consumed by humans. In groundwater, the total coliform target water quality range (TWQR) was exceeded at 53% of sites analysed and faecal coliform TWQR were exceeded at 77% sites. Surface water samples complied with TWQR with regards to faecal coliforms for full contact recreational activities and livestock watering. The TWQR for E. coli, with regards to full contact recreational activities, was within a safe range for surface water. Faecal streptococci were found in 85% of groundwater sampling sites. And surface water faecal streptococci counts exceeded the TWQR for full contact recreational activities. There is no TWQR for faecal sterols in water, but concentrations of cholesterol and coprostanol was found at three of the groundwater sites analysed. This indicates faecal contamination from possible animal and human origin. Surface water samples analysed showed that the Harts River water is clean and free of faecal sterols, while the water analysed from the inflow, hotel and outflow, cholesterol eluted, which showed faecal contamination, possibly from animals. Faecal sterol markers could be detected in groundwater and surface water, adding an extra dimension to determining the quality of water systems. An optimization and sensitivity study of the method was done on waste water treatment plant (WWTP) raw sewage and effluent. The WWTP sample analysed form Potchefstroom and Carletonville WWTP yielded all six target sterols in the raw sewage water samples, but no sterols eluted in the effluent samples. The raw sewage water sample taken from the Fochville WWTP yielded all six target sterols as well, however, the effluent yielded an unknown compound as well as cholesterol. An alternative study was done where the effluent sample volume was increased. By increasing the volume of water, one can possibly increase the amount (“load”) of sterols extracted and analysed, resulting in a higher abundance of target sterols. By using the target qualifier ions of the six target sterols, and the GC-TOF/MS software, the target sterols could still be qualitatively determined. Optimal volume for raw sewage is 300 ml water sample as this is enough to yield all 6 target sterols. For optimum water quality monitoring via faecal sterol analysis of effluent and other environmental samples, at least 1L sample volume needs to be collected and analysed. The methods described here can be applied to the analysis of environmental water samples. The technical advantages also make it suitable for routine environmental monitoring of faecal pollution. / MSc (Environmental Sciences), North-West University, Potchefstroom Campus, 2015

Faecal indicator bacteria have traditionally been used in the detection of faecal pollution in water, but due to concerns about the lack of reliability of these indicators, alternative methods have been developed. One of which is the detection of sterols present in human and animal excreta via GC-MS analysis of water in this study. The Szűcs method was used to detect six target sterols (coprostanol, cholesterol, dehydrocholesterol, stigmasterol, β-sitosterol, and stigmastanol) in environmental water samples. An initial study was done by analysing raw sewage and effluent (human faecal sterol biomarkers) and water samples were spiked with excreta from cattle, chickens, horses, pigs, and sheep to determine faecal sterol fingerprints. The method was evaluated for quantitation and differences between the water samples from each species. Following liquid-liquid extraction, silylation and derivatization, samples were analysed by GC-MS. Standard curve assays were linear up to 160ng and the limit for quantification was 20ng. The human faecal sterol biomarker was coprostanol, while herbivore profiles were dominated by terrestrial sterol biomarkers (stigmasterol and stigmastanol). Sterol fingerprints and differences in concentrations of sterols between various animals and between animals and humans occurred, providing the opportunity to determine whether faecal pollution was from humans or from animals. The method proved sensitive enough to evaluate faecal contamination in environmental water. Groundwater was collected from bore-holes and surface water samples were collected from the Baberspan Inland Lake. Physico-chemical parameters analysed indicated that pH for surface water samples was above 6.9. The total dissolved solids (TDS) in groundwater indicated that the water was not suitable for human consumption, but could be used for livestock watering. Surface water electrical conductivity (EC) and inorganic nitrates was too high to be used for irrigational purposes. Nitrates in groundwater were too high to be consumed by humans. In groundwater, the total coliform target water quality range (TWQR) was exceeded at 53% of sites analysed and faecal coliform TWQR were exceeded at 77% sites. Surface water samples complied with TWQR with regards to faecal coliforms for full contact recreational activities and livestock watering. The TWQR for E. coli, with regards to full contact recreational activities, was within a safe range for surface water. Faecal streptococci were found in 85% of groundwater sampling sites. And surface water faecal streptococci counts exceeded the TWQR for full contact recreational activities. There is no TWQR for faecal sterols in water, but concentrations of cholesterol and coprostanol was found at three of the groundwater sites analysed. This indicates faecal contamination from possible animal and human origin. Surface water samples analysed showed that the Harts River water is clean and free of faecal sterols, while the water analysed from the inflow, hotel and outflow, cholesterol eluted, which showed faecal contamination, possibly from animals. Faecal sterol markers could be detected in groundwater and surface water, adding an extra dimension to determining the quality of water systems. An optimization and sensitivity study of the method was done on waste water treatment plant (WWTP) raw sewage and effluent. The WWTP sample analysed form Potchefstroom and Carletonville WWTP yielded all six target sterols in the raw sewage water samples, but no sterols eluted in the effluent samples. The raw sewage water sample taken from the Fochville WWTP yielded all six target sterols as well, however, the effluent yielded an unknown compound as well as cholesterol. An alternative study was done where the effluent sample volume was increased. By increasing the volume of water, one can possibly increase the amount (“load”) of sterols extracted and analysed, resulting in a higher abundance of target sterols. By using the target qualifier ions of the six target sterols, and the GC-TOF/MS software, the target sterols could still be qualitatively determined. Optimal volume for raw sewage is 300 ml water sample as this is enough to yield all 6 target sterols. For optimum water quality monitoring via faecal sterol analysis of effluent and other environmental samples, at least 1L sample volume needs to be collected and analysed. The methods described here can be applied to the analysis of environmental water samples. The technical advantages also make it suitable for routine environmental monitoring of faecal pollution. / MSc (Environmental Sciences), North-West University, Potchefstroom Campus, 2015