1.
INTRODUCTION
Urban Flooding occurs when the capacity of a drainage system can’t be able to
carry stormwater inflow or to infiltrate water into the soil. It is different from rural
flooding since urbanization leads to create catchments which cause the rise of flood picks
and flood volumes. Urban flooding is mainly caused by the combined result of intense
rainfalls, the Occupation of flood plains by residential and commercial buildings, high
coverage of impervious areas, and inadequate drainage system bringing an excessive
runoff in developed areas where the water doesn’t have anywhere to go. Furthermore,
climate change plays a big role in urban flooding by changing precipitation patterns.
Mathematical modeling of watersheds is a popular approach followed by
engineers to identify solutions for flood problems. This kind of watershed modeling with
a proper model for the representation of the events that need to be analyzed requires data
in order to calibrate and verify the model outputs. Adequate resources are required for
both physical and financial acquisition of the data and the model. Some of the widely
used mathematical modeling packages in urban flood problems are The USEPA Storm
Water Management Model (SWMM), PCSWMM, MIKE URBAN, QQS, DR3M, and
HEC-HMS. These models take the advantage of different mathematical formulas to
calculate flood picks and to design drainage systems.
Stormwater modeling is the mathematical modeling that enables designers to
analyze and explore innovative stormwater design options and efficiently manage project
deliverables. Effectively managing stormwater is an important part of land development
and roadway design projects and is an essential aspect of public service in any
community. Designing an effective system requires balancing the design between
oversizing, which increases capital budgets, and under-sizing, which can lead to potential
flood problems. Using hydrologic and hydraulic helps to understand how rainfall
translates into stormwater runoff and interacts with streets, channels, pipes, ponds, and
other related stormwater infrastructures. Surface channels need to be sized and aligned
correctly so that they relieve flooding problems.
1.1. Background
Floods are the second destructive type of natural disaster in turkey, after
earthquakes. The biggest cause of floods and overflows in Turkey is climatic conditions.
Heavy rainfall lasting for days in April-May causes floods in some places. Both human
and geographical factors are important factors in the occurrence of floods and overflows.
Ankara, the capital of Turkey, is one of the major cities affected by stormwater floods.
Heavy rains in Ankara always occur in these periods, in convectional (short-term and
limited area) form. According to the Chamber of Civil Engineers (IMO) Ankara Branch,
when statistical data are examined, it is a known situation that a total of 40 mm of
precipitation may occur every 5 years in 24 hours, 45 mm in every 10 years, 55 mm in
every 25 years, and 95 mm in every 500 years. When the temporal extreme distributions
of precipitation are examined, it is seen that approximately 40% of 24-hour precipitation
can occur in 15 minutes.
1.2.Project Aim and Purpose
The aim of the project is to build, adjust, and validate an urban event-based rainfall-
runoff model for the Boğaziçi neighborhood in the Mamak district of Turkey based on the
flood that happened in early May of 2018. In addition, different sets of data, such as rainfall
data of different storm events will be used in calibration and validation processes in order to
get more accurate results. Finally, different modifications will be applied to improve the
efficiency of the drainage network in the research area.
The purpose of the project is to combine the urban flood simulation models with
statistical methods so they can be used to analyze current drainage system performance
regarding stormwater excess management.
1.3. Limitation
Some limitations come across to make the project more intensive and
comprehensive are described as follows:
a. Since there are no clear rules for the division of sub-catchment areas, it is difficult to
decide which method is better to get a better result.
b. As with most of the models, this model also takes the average slope as the slope parameter
of the sub-catchment area, which brings errors to the model in a mechanism.
1.4. Previous Researche Projects
Some previous research related to urban flood modeling using the SWMM package
and its comparison with other software has been approved and published. However, those
research projects vary in the focus of the discussions, objectives, and case study.
S. S. Wanniarachchi and N. T. S. Wijesekera, 2016 have used SWMM as a tool for
flood plain management in an ungauged urban watershed. In this report, the authors
analyzed that SWMM was applicable for urban flood modeling of the Matara Municipal
Council area, which is ungauged. This study used literature reported values and data collected
during field visits to runoff modeling to obtain order of magnitude values with easy
comparison of engineering alternatives for stormwater management (Wanniarachchi &
Wijesekera, 2016).
H. Haris et al., 2016 have compared some of urban stormwater management
software according to functionality, accessibility, characteristics and the components in
quality analysis of the representative models. EPA SWMM used for planning and design
functions, as well as it is accessible for public domain. Although other software like HEC-
HMS can be used for operational purpose only (Haris et al., 2016).
M.L. Waikarand and Undegaonkar Namita, 2015 have applied an urban flood
modeling by using EPA SWMM 5 for an educational campus located in Vishnupuri, Nanded.
This research is concluded that the capabilities of EPA SWMM 5 in simulating the
response of a catchment to rainfall events in which runoff, water depth profile, and outflow
hydrograph are obtained. Additionally, this project used rational formula for comparison
purposes to obtain Runoff (Waikarand & Namita, 2015).
In 2014, Lei Jiang et al. have conducted an urban flood simulation model using SWMM
to analyze the simulation of urban flooding in the rapidly urbanized southern China’s city of
Dongguan. The research is concluded that the SWMM model is likely favorable for urban
flood forecasting. However, a few limitations of SWMM were also observed during this
research. These limitations are generally associated with the absence of surface runoff
routing in SWMM that makes it difficult in forecasting urban floods precisely (Lei Jiang
et al., 2014).
Kourtis et al., 2017 stated that in both calibration and validation events in the combined
drainage network, SWMM model could simulate the shape of the hydrograph, the peak
discharge and the peak time quite precisely. Overall SWMM model was found to be a very
helpful modelling tool, which can be used for the simulation of urban drainage networks
(Kourtis et al., 2017).
In 2018, Junaidi et al. have evaluated the drainage system around Sungai Sapih
district hospital of Padang by calculating the quantity and quality of surface runoff from
each catchment area, flow discharge, flow depth, and water quality in each pipeline and
drainage channel during the simulation period. This study also analyzed that the drainage
planning design (DED) for study area is not enough to carry the drainage load of the area,
either now or in the future (Junaidi et al., 2018).
Sunny Agarwal and Sanjeet Kumar, 2020 have analyzed urban flood modeling using
SWMM for historical and future extreme Storm events under climate change scenario. The
research is concluded that flood occurred in few nodes of the region and there were some
overflow sections. This overflow in the section and nodes comes from heavy rainfall and
imperviousness of the surface. Additionally, this kind of project is helpful in assessing the
future floods, mitigating the risk of urban flooding and beneficial for policy makers for the
designing of suitable urban drainage system in developing urban cities (Sunny Agarwal and
Sanjeet Kumar, 2020).
2. LITERATURE REVIEW AND THEORETICAL BASE
2.1. General Description of a Mathematical Model
The term “model” has been widely used in different sector areas that sometimes
ones fell its meaning has become obscured. So it is very important to describe what a
“model” is and how it is used in the hydrologic design. Basically, a model can be defined
as a representation of phenomena in the world using something else to represent it,
making it easily understandable. It is a simplified form of a real-world system.
The reason behind using model is the complexity of the real-world system where
reflecting reality is impossible or where access to reality is inadequate in that certain
moment. A model has different forms, either physically or mathematically. A physical
analogy for operating a system is provided by the physical model and in most cases, it is
not possible to represent it physically. In this respect, emphasis will be directed to the
other form of model called “mathematical model”. In a hydrologic system, one normally
can’t model a watershed or rainfall physically. That means, all hydrologic models are
mathematical models which uses mathematical equation that characterize the prototype
system, giving relationships between variables enabling description, analysis and
prediction under the conditions to be modeled (Suleyman Oner, 1997).
Even though there are different types of mathematical models, they have a common
basic principle in allowing the analysis of the system in order to evaluate system
performance. Depend on the system performance, the one who use the model, determine
if or not the system will function at some desired level. On other hand, some acceptable
level of accuracy must be maintained if the model is to be acceptable (Gordon, 1985).
A model mainly used for: -
➢ Simulation
➢ Prediction/Forecasting
➢ Designing/performance Evaluation
➢ Prognostics/Diagnostics
➢ Control System Design
A mathematical model can represent simple or complicated systems. A simple
example of a mathematical model is an equation representing a certain process. A
manning’s equation (2-1) is a hydraulic model used in an open channel to calculate flow
depth. Rational equation (2-2) is also other example that used in small watershed to
calculate the runoff exposed to storm rainfall.
2.1
Where Q represents river flow, A, the cross-sectional area, R, hydraulics radius,
S, the hydraulics gradient and n, the roughness coefficient.
2.2
Where Q represents the discharge from catchment, C, the runoff coefficient, I, the
rainfall intensity and A, the drainage area.
Complicated mathematic models are those models which require large computer
storage capabilities and advanced mathematics and programming. These models
represent physical phenomena more accurate and precise than the simpler models and
currently used in variety of task ranging from real-time prediction of impending flood
events to the design of policies and structures for mitigating extreme hydrologic events
(e.g., floods and droughts). For the solution of such practical problems, these hydrologic
models must be calibrated regularly (Luis. A. Bastidas at el,2002).
The introduction of the unit hydrograph concept by Sherman in 1932 and the
development of the infiltration theory by Horton in 1933 marks the beginnings of
mathematical modelling of watershed hydrology. The 1960s decade witnessed digital
revolution that made possible the integration of different component of the hydrologic
cycle and simulation of virtually the entire watershed hydrology. Since then, the power
of computer has increased exponentially and improvements in watershed hydrology have
occurred at an extraordinary rate.
Figure 2.1 is a schematic representation of mathematical modeling process. The
modeling process set up with the specification of the problem through the construction of a
simplified and idealized version of the problem by detecting the crucial components of the
model and by making hypotheses. Then, the problem is interpreted into a mathematical model
consisting of mathematical expressions variables and describing the relationships among the
variables. The problem is then analyzed and resolved in order to find mathematical results.
The results are further interpreted in terms of the simplified real-world situation. Finally, the
solution generated for the simplified version of the problem is verified (Aysel sen Zeytun at
el.,).
Figure 2.1 Mathematical modelling process (NCTM, 1989, p. 138)
2.2. Classification of Mathematical Models
Mathematical models can be divided in to two major catagories: Stochastic and
Deterministic models. The main difference between these catogeries is Stochastics models
are based on statistics and probability and more concerns about predictions While
deterministic models used to make forecast. Prediction is about set of simulated hydrographs
used for the purpuse of engineering design. On the othor hand, forecasting refers simulated
hydrographs of specific future events to be used to make in making opretional decisions
(Suleyman Oner, 1997).
In deterministic models such as manning’s and rational equations, outcomes are
precisely determined through known realationship between states and events without a place
for random variatios. There are two types of deterministics models based on spatial
representations, lumped and distributed mathematical models. The lumped model considers
individual sub-basins as a single unit in space without dimensions, whereas the distributed
model sub-divides each sub-basin in smaller cells as function of space dimensions. These
models may be steady-flow or unsteady- flow models based on the flow rate with respect
to time.
Event-based hydrologic modeling deals how a basin responds to an individual
rainfall history (e.g., surface runoff quantity, peak, peak time). In contrary, continuous
hydrologic modeling integrates hydrologic processes and phenomena over a longer time
period in both wet and dry conditions (Xuefeng Chu and Alan D. Steinman,2009). The
summary of general classification of mathematical models in hydrology is shown in
figure 2.2.
Figure 2.1 Classification of hydrologic models (adapted from chow et al,1988)
2.3. Urban Storm Managemnt Model
Urban Stormwater modeling is the mathematical modeling of stormwaters
mainly in urban areas to enable designers to analyze and explore innovative stormwater
design options and efficiently manage project deliverables. Rapid urbanization has
numerous undesirable impacts on the hydrological cycle by increasing impervious areas
and worsening water quality in stormwater runoff.
The first computerized Models capable of simulating stormwater quality and
quantity appeared in the earlier 1970s, and were developed primarily by US government
agencies, such as the USEPA (Zoppou, 2001; Mitchell, 2001). Since that time, several
urban watershed models have been developed and discussed in literature. These models
frequently used for planning, designing, flow predicting and screening purpose.
Rainfall-runoff modeling and transport modeling are the two basic components of an
urban storm water model and the illustration between these models is given in figure 2.3.
Figure 2.3 Overview of process included in a storm water model (Zoppou, 2001)
There are many software packages developed by academic institutions, regulatory
authorities, government departments and engineering consultants that are capable of
modeling and simulating water quality and quantity primarily in an urban areas such as
Storm Water Drainage System design and analysis program (DRAINS), Urban Drainage
and Sewer Model (MOUSE), Info Works River Simulation (InfoWork RS), Hydrological
Simulation Program-Fortran (HSPF), Distributed Routing Rainfall-Runoff Model
(DR3M), Storm Water Management Model (SWMM), XP Storm Water Management
Model (XPSWMM), MIKE-SWMM, Quality-Quantity Simulators (QQS), Storage,
Treatment, Overflow, Runoff Model (STORM), and Hydrologic Engineering Centre-
Hydrologic Modelling System (HEC-HMS).
Most of these models are 1-D models and are based on the principle of mass,
energy, and momentum conservation. Some models are event-based models and others
are of continuous simulation type (Vinay Ashok Rangari at el,.2015). The summary of
comparison between these urban flood models is given in table below.
Table 2.1. Comparison of Urban Flood attributes (Donigan et al.,1995)
Attributes DR3M HSPF STORM SWMM
Sponsoring agency USGS EPA HEC EPA
Simulation type (Event- Both Both Continuous Both
based/Continuous)
No of pollutants 4 10 6 10
Rainfall/Runoff analysis ✓ ✓ ✓ ✓
Sewer system flow ✓ ✓ ❌ ✓
routing
Full, Dynamic routing ❌ ❌ ❌ ✓
equation
Surcharge ✓ ❌ ❌ ✓
Regulators, overflow, ❌ ❌ ✓ ✓
structures e.g., weirs
Storage analysis ✓ ✓ ✓ ✓
Treatment analysis ✓ ✓ ✓ ✓
Data requirement Medium High Low High
Complexity Medium High Low High
2.4. The EPA SWMM
EPA SWMM 5 is one of urban flood modeling software commenly used at present by
policymakers for planning and disign of suitable urban storm water drainage systems. It is a
stormwater management model which is first developed in 1971 by the U.S. Environmental
Protection Agency's National Risk Management Research Laboratory. It is a distributed
dynamic rainfall-runoff simulation model mainly used in urban araes for simulation of event
based or continious runoff quantity and quality.
The runoff component of SWMM operates on a collection of subcatchment areas that
receive precipitation and generate runoff and pollutant loads. The routing portion of SWMM
transports this runoff through a system of pipes, channels, storage/treatment devices, pumps,
and regulators. SWMM tracks the quantity and quality of runoff generated within each sub
catchment, and the flow rate, flow depth, and quality of water in each pipe and channel during
a simulation period comprised of multiple time steps (Huber and Dickinson,1988).
