Modelling the response of influent volumes of wastewater treatment plants under current and future conditions for effective wastewater management in combined sewerage systems
Date
2023-09-28Author
Saikia, Sukanya D.
Metadata
Show full item recordUsage
This item's downloads: 127 (view details)
Abstract
Wastewater treatment plants (WWTPs) are critical infrastructure globally and are
essential to protect public health and the environment. With factors such as population
growth, urbanisation, increase in water consumption etc., the amount of wastewater
generated has increased significantly, which impact the operations of WWTPs. Of
particular concern are WWTPs with combined sewerage systems (CSSs) which treat foul
and storm wastewater collectively. Such WWTPs are also influenced by changes in the
intensity and frequency of precipitation events. During instances of increased
precipitation intensity and frequency of storm events, WWTPs with CSSs might
encounter hydraulic overloading and release of untreated wastewater called combined
sewer overflows (CSOs). On the other hand, lack of precipitation can lead to reduced
flow and increased contaminant loading. In both the cases, with climate change and the
associated changes in precipitation patterns, WWTPs with CSSs may become more
susceptible to system failures that pose a threat to the receiving waters and the
surrounding natural environment.
As stricter environmental regulations are enforced to limit the occurrences of CSOs, it
has become increasingly important to identify the variables that impact the functioning
of WWTPs with CSSs. However, unavailability of real data is a key challenge in the
research area of monitoring the performance of WWTPs and sewerage systems. Studies
have conventionally used modelled data simulated from hydraulic models that do not
cater to local characteristics of individual WWTPs and hence are often associated with
large uncertainties. While considerable attention has been given to the impacts of climate
variables (current and future) and urbanisation on effluent quantity and quality and the
performance of the sewerage systems focusing on CSOs, the same cannot be said for
wastewater influent volumes. Influent volume characteristics function differently as
compared to CSOs and effluent volumes. Once CSOs leave the sewerage system, there
are still variations in the flow that proceed towards the WWTPs during or after the spill.
Hence influent volumes have a significant impact on subsequent WWTP processes and
predicting how they might change can help prevent occurrences of overflows and aid in
achieving resilience of WWTPs. Studies investigating the degree to which precipitation
change (current and future), tidal level, river level, and urbanisation impacts influent
volumes of WWTPs with CSSs remain unexplored. This research addresses these gaps by studying influent volume response characteristics
of 14 WWTPs of varying sizes, connected with CSSs, that are spatially distributed over
Ireland. The thesis uses real spatio-temporal datasets of precipitation, influent volumes
and location-specific data of tidal and river level. The objective of this study was to
develop methodologies under practical data constraints to build models that could define
meaningful relationships among the different variables. Daily precipitation and daily
mean river level were found to be statistically significant predictor variables of influent
volumes at a daily scale. On a monthly basis, monthly average daily precipitation, number
of wet days in a month (and thus zero rainfall days) were observed to be statistically
significant. The daily and monthly variations in influent volumes for each of the WWTPs
were assessed with the help of simple and multiple linear regression modelling analysis.
These individual WWTP models helped to capture local characteristics specific to each
WWTP. In addition, a novel pooled model was developed through spatio-temporal
analysis across all the 14 WWTPs to derive generic trends in influent volumes across any
WWTP. Probability of exceedance curves linking daily precipitation and influent
volumes were also developed that could aid in identifying storm overflow events under
various precipitation categories. These graphs could be potentially used for future climate
scenarios using precipitation projections to estimate the projected frequency of storm
overflow events.
This research also analysed, for the first time, the evolution of influent volumes during
mid-century period (2041 – 2060) as compared to current period. It predicted future
influent volumes by leveraging high resolution multi-model regional climate model
projections of precipitation intensity and extreme events for each WWTP and linking
them to the developed data-driven models and probability of exceedance curves. This
analysis offers valuable insights into how WWTPs might get impacted in future (e.g.,
exceedance of peak design capacity under extreme weather conditions) due to climate
change.
Finally, this research aims to investigate the degree to which urbanisation might
potentially impact influent volumes of WWTPs with CSSs. Landsat 5 and Landsat 8
satellite images were used to perform landuse landcover classification of all the 14
agglomerations corresponding to each WWTP to estimate the change in built-up area.
Percentage change in built-up area relative to agglomeration area was found to be
statistically significant with moderate degree of correlation with influent volumes across
all agglomerations. The findings of this research will help wastewater utilities
(particularly the ones connected with CSSs) as end-users, take informed decision in their
planning and adaptation strategies in order to establish resilient wastewater infrastructure
at regional and local WWTP scales.