Date of Award

2015

Degree Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Civil and Environmental Engineering

First Advisor

Thomas Boving

Abstract

The use of onsite wastewater treatment system (OWTS) is a common practice in the U.S., especially in rural areas where the access to centralized wastewater treatment systems is limited. Onsite wastewater treatment systems include a soil treatment area or drainfield where contaminants are removed or attenuated. Ineffective OWTS are a source of microbial pathogens (bacteria and viruses), biological oxygen demand (BOD) and nutrients, which are among the major causes of contamination and water quality impairments in surface water in the U.S. The main objective of this research was to model the different chemical, physical processes, and removal mechanisms that influence the fate and transport of OWTS-derived contaminants using the HYDRUS 2D/3D software. In the first part of this study, segmented mesocosms (n=3) packed with sand, sandy loam or clay loam soil were used to determine the effect of soil texture and depth on transport of two septic tank effluent (STE)-borne microbial pathogen surrogates - green fluorescent protein-labeled E. coli (GFPE) and MS-2 coliphage - in soil treatment areas. In all soils, removal rates were >99.99% at 25 cm. The transport simulation compared (1) optimization, and (2) trial-and-error modeling approaches. Only slight differences between the transport parameters were observed between these approaches. Independent of the fitting procedure, attachment rates computed by the model were higher in sandy and sandy loam soils than clay loam, which was attributed to unsaturated flow conditions at lower water content in the coarser-textured soils. In the second part of this research, bacteria removal efficiencies in a conventional soil-based wastewater treatment system (OWTS) were modeled to elucidate the fate and transport of E. coli under environmental and operational conditions that might be expected under changing climatic conditions. The impact of changing precipitation patterns, initial bacteria concentrations, hydraulic loading rates (HLR), and higher subsurface temperatures at different depths and soil textures on bacteria removal was evaluated. Modeled effects of initial bacteria concentration shows that greater depth of treatment was required in coarser soils than in fine textured ones to remove E. coli. The initial removal percentage was higher when HLR was lower, but it was greater when HLR was higher. When a biomat layer was included in the transport model, the performance of the system improved by up to 12.0%. Lower bacteria removal (up to 5%) was observed at all depths under the influence of precipitation rates ranging from 5 cm to 35 cm, and 35 cm rainfall combined with a 70% increase in HLR. C Increased subsurface temperature due to climate change (23 °C) increased bacteria removal relative to a lower temperature range (5 °C to 20°C). It appears that the performance of OWTS may be impacted by changing climate. In the third part of this research, we also simulated the fate and transport of N in three different types of OWTS drainfield, or soil treatment areas (STA) using 2D/3D HYDRUS software to develop a N transport and fate model. Experimental data from a laboratory mesocosm study, including soil moisture content, and NH4 and NO3- concentration, was used to calibrate the model and a water content-dependent function was used to compute nitrification and denitrification rates. Three types of drainfields were simulated: (1) pipe-and-stone (P&S), (2) pressurized shallow narrow drainfield (SND) and (3) Geomat (GEO), a variation of SND. The results showed that the model was calibrated with acceptable goodness-of-fit between the observed and measured (average root mean square errors (RMSE) ranged from 0.18 to 9.65 for NH4+ and NO3-). The model predicted the N losses from nitrification and denitrification in all STAs. The modeled N losses occurred mostly as NO3- in water Outputs, accounting for more than 82% of N inputs in all drainfields. The highest N losses by denitrification were computed for the P&S drainfield and accounted for 17.60% of the influent total N. Our results showed that HYDRUS is a useful tool to predict the fate and transport of nutrients and microbial contaminants and help to provide practitioners with guidelines to estimate pathogens and nutrients removal efficiencies for OWTS under the effect of different operational and environmental factors. In addition, the modeling approach presented in this study, will be useful to predict the extent of contamination and spatial distribution for identifying non-point sources, and establish total minimum daily loads (TMDLs).

Share

COinS