Statistical analysis of parameters affecting Legionella and total cell growth in premise plumbing systems within buildings: A field study based on an empirical data set.

During the storage and distribution of water in buildings, the excessive growth of pathogens can deteriorate the quality of drinking water. This study aims to investigate the factors influencing this growth and propose technical measures for prevention. The analysis is based on an empirical data set comprising 1361 samples from 204 domestic premise plumbing systems. In 14 systems, ultrafiltration plants were installed as microbiological barriers. Legionella cultivation and flow cytometry were used to determine microbiological properties. The study identified elevated total cell counts in tapping valves and pipe end lines in numerous premise plumbing systems, indicating prolonged water stagnation prior to sampling, which facilitates microbiological growth. Higher contamination rates were observed in these systems, with peripheral taps often being contaminated in lieu of the entire system. These systems were classified as microbiologically unstable due to the relevantly higher total cell numbers at hot water taps compared to the hot water tank (>25%). Furthermore, these systems exhibited a Legionella contamination rate that was 22.3% higher than in microbiologically stable systems. In some cases, peripheral contaminations may not accurately represent the entire premise plumbing system. Increasing the discard volume during sampling from 1 L to 3-5 L could provide more precise results during standard testing. Legionella species were primarily detected in the first 1 L of water after tap activation. Additionally, statistically significant relationships were observed between direct temperature and total cell number, as well as between the presence of ultrafiltration and total cell numbers at cold water taps.

Monitoring of microbially mediated corrosion and scaling processes using redox potential measurements.

The use of redox potential measurements for corrosion and scaling monitoring, including microbially mediated processes, is demonstrated. As a case study, monitoring data from 10years of operation of an aquifer thermal energy storage (ATES) site located in Berlin, Germany, were examined. (Fe(2+))-activities as well as [Fe(3+)]-build up rates were calculated from redox potential, pH, conductivity, temperature and dissolved oxygen measurements. Calculations are based on assuming (Fe(3+))-activity being controlled by Fe(OH)3-solubility, the primary iron(III)-precipitate. This approach was tested using a simple log-linear model including dissolved oxygen besides major Fe(2+)-ligands. Measured redox potential values in groundwater used for thermal storage are met within ±8mV. In other systems comprising natural groundwater and in heating and cooling systems in buildings, quantitatively interpretable values are obtained also. It was possible to calculate particulate [Fe(3+)]-loads in the storage fluids in the order of 2µM and correlate a decrease in filter lifetimes to [Fe(3+)]-build up rates, although observations show clear signs of microbially mediated scaling processes involving iron and sulphur cycling.

Inside-sediment partitioning of PAH, PCB and organochlorine compounds and inferences on sampling and normalization methods.

Comparability of sediment analyses for semivolatile organic substances is still low. Neither screening of the sediments nor organic-carbon based normalization is sufficient to obtain comparable results. We are showing the interdependency of grain-size effects with inside-sediment organic-matter distribution for PAH, PCB and organochlorine compounds. Surface sediment samples collected by Van-Veen grab were sieved and analyzed for 16 PAH, 6 PCB and 18 organochlorine pesticides (OCP) as well as organic-matter content. Since bulk concentrations are influenced by grain-size effects themselves, we used a novel normalization method based on the sum of concentrations in the separate grain-size fractions of the sediments. By calculating relative normalized concentrations, it was possible to clearly show underlying mechanisms throughout a heterogeneous set of samples. Furthermore, we were able to show that, for comparability, screening at < 125 µm is best suited and can be further improved by additional organic-carbon normalization.