Now showing 1 - 2 of 2
No Thumbnail Available
Publication

Modelling Quantified Microbial Source Specific Pollution from Domestic Wastewater Treatment Systems during High Flows

2016-11-15, Corkery, Aisling, O'Sullivan, J. J., Deering, Louise, Demeter, Katalin, Ballesté, Elisenda, Masterson, Bat, Meijer, Wim, O'Hare, G. M. P. (Greg M. P.)

The ability to model the transport of source specific faecal bacteria contamination in river networks can equip water resource managers with information of the different pathogens that are present. Such information can be particularly useful in catchment management plans for rivers from which potable water is extracted or where these rivers discharge to coastal zones where bathing or aquaculture is prominent and where the evaluation of human health risks is of primary importance. This paper presents an application and performance assessment of the commercially available MIKE11 Hydrodynamic model for evaluating the fate of faecal bacteria of human origin, Human Gene Marker HF183f, from Domestic Wastewater Treatment Systems within the Dargle catchment. The Dargle is a spate river and the upper catchment is characterised by steep slopes that incorporates peat bogs and land used for forestry and agricultural purposes. Residential dwellings within the area are predominantly single detached units that rely on septic tanks for wastewater treatment. In the context of faecal bacteria of human origin, malfunctioning systems are of concern, particularly in terms of surface ponding, leakage to groundwater and direct discharge to surface waters. The MIKE11 model was calibrated in a two stages process. Firstly, the model was calibrated for prediction of discharge and microbial water quality parameters, namely E. coli and Intestinal Enterococci (IE), using data from a real-time sensor network within the catchment that comprised rain gauges, weather stations and water level recorders, data from which was used to determine flow records from stage-discharge ratings. E. coli and IE concentrations were determined from high resolution sampling during storm events. Following this, water quality samples taken during storm event sampling were used to identify and quantify the human gene marker HF183f using quantitative polymer chain reaction (qPCR) techniques. Results from the qPCR analysis were used to further calibrate the model at sub-catchment level for the transport of microbial bacteria derived from human origins. Using non-compliance statistics from the EPA National Inspection Plan, domestic sources have been calculated based on the percentage number of malfunctioning septic tank units and average daily faecal gene marker concentrations per household. The study highlights issues with how the fate of the human gene marker is modelled using MIKE11, particularly in terms of advection-dispersion inputs and the requirement to associate microbial concentrations with total runoff when modelling surface and groundwater pathways.

No Thumbnail Available
Publication

Integrated modelling of urban, rural and coastal domains for bathing water quality prediction – Smart Coasts and Acclimatize Projects

2018-06-27, O'Sullivan, J. J., Bedri, Zeinab, Corkery, Aisling, Deering, Louise, Demeter, Katalin, Meijer, Wim, Masterson, Bartholemew, O'Hare, G. M. P. (Greg M. P.)

This paper presents the results of the Interreg funded SMARTCOASTS project in which an integrated catchment (MIKE11) and coastal (MIKE21 and MIKE3) modelling tool was developed for predicting the bathing water quality, at Bray, Co. Wicklow, on the east coast of Ireland. The Bray bathing waters had historically been prone to episodic short-term pollution, caused primarily by rainfall related catchment run-off. Accounting fully for the complexity of the pollution inputs for water quality prediction in the system required an integrated modelling approach. The approach for integrating the individual component models (NAM, MIKE 11, and MIKE 3 FM) was simple but efficient. The component models, interfaced to the core of the forecasting system, were run sequentially, i.e. in the form of a cascade with the forcing of each downstream model being the result of the model upstream of it. Rainfall (both forecasted and measured) drives the hydrological processes in the NAM model, which produces runoff that generates sub-catchment inflows into the river network. The output from NAM serves as the input to the MIKE 11 model which routes the flow and water quality variables in the river network and transports them to the coastal waters. Finally, the MIKE 3 FM coastal model uses flow and water quality outputs from MIKE 11, together with tidal and meteorological data, to simulate the current flow, transport and fate of water quality variables in the marine environment. Models were calibrated using measured data. Adjustment of the tidal constituents of the MIKE global model resulted in a markedly improved fit to measured water levels at five reference tidal gauges, used for calibration. Bottom friction was calibrated to produce good correlations of measured and simulated current speed and direction. When applied to water quality prediction, results of the transport model showed that the model adequately replicated measurements of E.coli and Intestinal Enterococci within the coastal domain. Computational simulations of bathing water quality are not without difficulty and a significant challenge in this work involved incorporating real-time meteorological data from a sensor network within the catchment into the model predictions. The work of Smart Coasts is currently being built on in the Interreg funded Acclimatize project. Acclimatize is focussing on the bathing waters of Dublin Bay and involves the development of a modelling platform that will facilitate a longer-term assessment of the likely pressures on bathing water quality in the context of a changed climate.