Dynamics Model of Eco-Security Surveillance System for
River Tamsui in Taipei City
Hsiao-Lin Wang’ Yu-Feng Ho? Ching-I Wu"
Graduate School of Architecture and Urban Design,
Chaoyang University of Technology, Taichung, Taiwan
1 Email: whl1435@hotmail.com
2 Email: hyfarch@ms32.hinet.net
3 Email:gardenarS71@gmail.com
Abstract
Current research on the river ecosystem in Taiwan is mostly focused on the water
conservancy, ecology, and afforested viewpoints. There is a lack of integrated strategy
on urban river ecosystem management. In this study which aims to examine the river
quality based on ecological safety. By means of the systems engineering technology,
the related ecological safety operation mechanisms have been analyzed. Also, through
the new Fuzzy Delphi expert survey, the index system of urban river ecological safety
(IES) has been compiled in order to explore the variables. The key indicators affecting
the urban river ecology safety can be fully defined by sensitivity analysis in order to
engage in thr evaluation of effectiveness and to create a dynamic simulation model.
The study results indicate that the strategic implementation of improved embankment
vegetation structure, reduced degree of river channelization, and maintenance of a
high degree of longitudinal connectivity of rivers can effectively enhance the urban
river ecology risk prevention, strengthen the efficient use of resources, and promote
the sound development of the urban river ecology.
Keywords: ecological safety monitoring, Tamsui River, Fuzzy Delphi, system
dynamics
1. Introduction
With the overall goal of “building a land of esthetic ecological urban and rural
environment” and the vision of a “vibrant city and sustainable Taiwan”, the ecological
optimization of urban development in Taiwan has become an important indicator for
Taiwan’s future land development. Taiwan is situated in the sub-tropics with abundant
rainfall. In addition, due to the uneven distribution of rainfall in terms of the time and
place, coupled with the steep mountains and short river currents, a total of 350
typhoons have struck Taiwan and up to one thousand heavy rains have taken place in
the past 101 years. With the rainfall concentrations, short rivers and rapid currents,
flooding has been the most serious natural disaster in Taiwan (Water Policy and
Management Research Center, 2010). Therefore, flood management is Taiwan’s
long-term goal in river management. However, the urban flooding caused by Typhoon
Morakot and Typhoon Toraji seems to imply that the urban river ecological system
can no longer bear the destruction brought about by the urban development. In recent
years, the Taiwan government has actively promoted flood prevention and river
remediation programs; however, river management has been implemented under the
premises of flood management and maximum economic benefits of lands, of which
numerous hydraulic structures have been built in order to protect people’s lives and
property. By doing so, the original ecological habits have been changed at the same
time. Because the ecological functions had often been neglected in the past, the river
ecological systems were subject to destruction. This negligence may likely hinder the
sustainable urban development while the proper water consumptions of rivers and the
self-purifying water quality and biological diversity functions have gradually
diminished, thereby threatening the long-term survival and well-being of mankind.
Thus, the necessity and value of an effective early warning system have been
confirmed. In addition, the effective implementation of the early warning system can
also greatly reduce the occurrence of disasters (Tung Ching-Ping, et.al, 2008). For
city residents, the early warning system can minimize the risk of living in insecure
environments.
At present, environmental safety issues have been developed toward
macroscopic and microscopic trends. “Security”is the status of safety and eco-security
means the security to maintain human beings’ ecosystem upon healthy and sustainable
development. The ecology means multiple artificial eco-security system constructed
by nature, economy and society(Xie,2007;Wu,2011) . A large number of researches
have applied the ecological theories to the economic development considerations.
Instead of having fragmented national policies, they have been turned into alliance
cooperation, thereby putting a stop to environmental degradation and slowing down
the global climate changes through the economic mechanisms. Take the “Kyoto
Protocol” signed in 1997 for instance; through the carbon tax system, the goal of
restoring the global ecological balance has been achieved. Discussions on the urban
ecological development issues are unlike other general ecological system related
studies. First of all, the urban ecological system covers the natural, social, and
economic system aspects, not a static one-way system; second, the formulation of the
urban ecological management strategies must be forward-looking and implementable.
Rivers, like human blood, must perform the functions of ecological environment
conservation, transportation, and cleaning. The idea of urban river eco-security is to
keep urban residents life in safety and to keep nature water resource comes from
urban river which people used no longer in risk both(Wu,2011).The rivers and green
spaces that demarcate the water from the land are ecological habitats that stabilize the
urban biological species and provide the city residents with open spaces for recreation
purposes. The review of the river ecological system from the security-based viewpoint,
the establishment of a cordon for monitoring the river ecological functions, and the
review of the complex relationship between the river ecology and human
development are essential for the ongoing sustainable urban development.
2. Study Methods
From the ecological-based viewpoint, rivers are a dynamic and continuous water
ecological system (Wang Ching-Ming, 1992). The environmental makeup of rivers is
constituted by water environments and catchment areas. The ecological system
structures accommodate a cluster of biological species (such as producers, consumers,
decomposers, and other biological components), which are made up of physical and
chemical environments that are habitats for biological species (such as air, sunlight,
water, soil, chemicals, etc.) (Kuo, Monica, 2001). The combination of geography and
landscape ecology also offers another direction of thinking: the biological species
migrate along the river, forming a narrow band of “river corridor” (Kuo, Monica,
2001). This band stabilizes the water cycle and acts as a buffer for maintaining
ecological balance; it is also a biological habitat that provides biological species with
shelter and food and allows them to nurture their young after birth. The creatures in
clusters in the rivers are subject to effects of both the external (land) and internal
(water) environmental factors; that is, the interaction between nature and humanity
related environmental factors possess more complex structures and functional
processes. Although the urban rivers are subject to limitations of the riverbank and
flood-prevention facilities, the urban river ecological system is not only about water
or river areas as are often discussed, but also includes the water resources and
network systems of the natural land environment and resources that rely on and
maintain the river functions so that the overall river ecological system can be better
expressed.
The urban river ecology is a part of the urban ecology. Although the complex
ecological system is made up of social, economic, and natural aspects, it is still
consistent with the mechanism of urban ecology. The urban river ecology features the
following:
1. Integrity: The urban river ecological system is a complex and independent
functional unit in the environment.
2. Causality: The urban river ecological system’s functions are affected by the internal
and external variables of the system, of which the factors affect each other as a
unified system.
3. Restorability: The urban river ecological system has stable and feedback-regulating
internal mechanisms.
In this study, for the sake of obtaining complete and continuous research quality,
the research discussions and purposes in every stage of the paper are independent; yet,
they are logically related. It is therefore necessary to adopt cross-disciplinary
approaches to engage in integrated research.
2-1. Systems Engineering
Systems engineering takes into consideration system related discussions, as well
as the uncertain factors. In addition, the order in rank within the system elements has
been found through logical and organized tools and steps, and the correlation between
the system elements and the entire system have been measured in order to explore the
entire system organizational structure. In view of the massive and complex urban
river ecological system, this study has adopted the systems engineering theories and
methods in order to formulate the preliminary index system.
2-2. Fuzzy Delphi
Ishikawa et.al (1993) used the cumulative frequency distribution and fuzzy
integral concepts to integrate the expert opinions and transform them into fuzzy
numbers. In addition, the geometric mean serves as the basis for the group
decision-making screening and evaluation criteria in order to avoid extreme values
and ensure better results in the evaluation factor selection, all of which are
collectively known as the Fuzzy Delphi Method (FDM). In this study, the amended
method of fuzzy number construction proposed by Cheng (2005) has been adopted.
That is, the traditional Delphi Method has been adopted to evaluate the indicators
using single values. On the other hand, after obtaining the mean, the researcher
evaluated the consensus reached among the experts in order to determine the degree
of importance of the expert opinions and derive at the triangular fuzzy number
graphics. In this study, the FDM has been adopted in order to enhance the objective
and sciential urban river ecological system index system. The FDM has advantages
such as low investigation frequencies, complete expression of the expert opinions, and
inclusion of fuzziness that inevitably appears in the investigation process (Wang, 2007;
Wu, K.Y. et al, 2002; Wu Yi-Yen, 2008).
2-3. System Dynamics
System dynamics refers to the application of the causation and information
feedback concepts in analyzing the system structure, exploring decision-making that
affect system behaviors and the time lag phenomenon during the decision-making
process. It is ideal when used to deal with cyclical, long-term, multi-variable,
high-ranking, nonlinear, and time lag related issues, making it highly suitable for
studying complex problems related to the river ecological system. Moreover, it
features strong forecast and back-test functions, which help achieve the purpose of
integrating reality and history. In addition, the model simulation operation process
reflects the actual operation condition, which shall serve as a reference for policy
amendments and project evaluations. The urban river ecological system that involves
complex areas such as the overall urban development and ecological operation
mechanisms presents the “dynamic complexity” characteristics while the system
dynamics applied in the urban ecological system simulation analysis has a unique
research advantage.
2-4. Sensitivity analysis
Sensitivity analysis engages in the debugging of the decision-making analysis of
the actual system using models. By changing the parameters, structures, model
simulation, and comparative model outputs of the models, the extent of effects can be
4
determined. In the actual system, some parameter data are affected and controlled by
multiple factors. Thus, the ultimate purpose of the establishment of models is to pass
the debugging and seek more sensitive parameters in the models, which are the
leverage points sought in the actual system policy analysis, through which the best
policies can be found (Barlas, 1996; Sterman, 2002).
In this study, the systems engineering method has been adopted to analyze the
urban river ecological system operation. Through the Fuzzy Delphi expert opinion
survey, the index systems affecting Taipei City’s Tamsui River ecological system have
been compiled. Moreover, through the system dynamics method and using the Vensim
DSS 5.9 software, the main causes and causations of the urban river ecological system
have been constructed. Through the sensitivity analysis, the key factors affecting the
ecological system was grasped to engage in strategy efficiency evaluations, set up the
dynamic simulation model of the urban river ecological security monitoring system,
enhance the urban river ecological system risk prevention, strengthen the effective use
of resources, and promote the comprehensive development of the urban river
ecological system.
3. The Setup of the Urban River Ecology Security Evaluation Index
system
3-1 The Preliminary Formulation of the Index system for Urban River Ecology
Security
In this study, based on the domestic and foreign literatures and in consideration
to the urban river management characteristics and experiences, the engineering
system technology has been used to find objective, scientific, measurable, and
comparable system indicators from the exiting related important research literatures,
including 29 variables that are divided into five sub-systems: flood security, water
security, purification security, habitat security, and biosecurity. By application of the
Delphi Fuzzy Expert Survey Methods (Ho & Wang, 2008) including “group
decision-making” and “fuzzy evaluation,” the recommended variables were selected
from the perspectives of different disciplines and perspectives to output representative
viewpoints. In this study, after obtaining the respective indicator evaluations using the
Fuzzy Delphi Method, the Fuzzy Partition Method proposed by Zhao,R. and Govind
(1991) was applied to obtain the mean after fuzzification (BNP), which serves as the
integrated value consistent with the expert opinion. In addition, based on the fuzzy
numbers used in the study, the method of fuzzy number construction was applied to
analyze the expert values, integrate the urban river ecology security subsystems, and
set up the principle for selecting the reference values using single threshold values.
This method possesses the objectiveness of indicator selections using fuzzy numbers;
at the same time, as the survey content is clear and easy to implement, the writing and
numerical computations can be simplified, thereby enhancing the efficiency of the
Amalfi survey.
3-2 The Urban River Ecology Security System Index system
This research referenced local and international articles, while taking into
consideration the unique characteristics and experiences of Taiwan’s urban river
ecosystem. Engineering system techniques established at the initial stage an optimistic,
scientific, predictable and comparable variables set. In this study, the questionnaire
subjects invited include experts of landscape planning, urban planning, architectural
design, water resources engineering, and ecological conservation. A total of 20
questionnaire copies were distributed, and 20 copies were retrieved, thus achieving
100% recovery rate. Of the questionnaires, 20 were valid questionnaires, thus
achieving 100% efficiency rate. After obtaining the respective evaluations through the
Fuzzy Delphi Method, the respective indicator values in the questionnaires were
computed: the BNP value was 3.96, and the standard deviation was 0.57; therefore,
3.3 served as the threshold value selection indicator. Seven evaluation items,
including degree of river meandering, horizontal connectivity of the river, human
impact, survival rate of protected biological species, ratio of alien species, riverbed
sediment types, and environmental awareness have not reached the threshold values
and thus they have been eliminated. The indicators were decreased from 29 to 22,
which account for 75.86% of the overall ratio.
Among the index systems, the discharge security sub-systems including the
amount of discharge (four indicators), purification security sub-system (four
indicators), water security sub-system (five indicators), habitat security sub-system
(five indicators), and biosecurity sub-system (four items). The respective expert
evaluation is intended for experts’ degree of consensus regarding the importance of
the indicators on the systems and respective sub-systems. The survey results indicate
that the top nine in overall ranking are: per capita water consumption (4.72), riverbed
sediment stability(4.69), livelihood sewage and wastewater emissions(4.67),
riverbank _ stability(4.64), industrial sewage and wastewater emissions(4.64),
embankment vegetation structure(4.57), surface runoff(4.55), heavy metals in
water(4.55), and number of biological species(4.48). In general, the indicator items
related to water consumption, water quality, riverbank type, and vegetation condition
have been attached importance. This being said, the urban river ecological water
quality and discharge capacity remain the primary targets for maintaining the
ecological security system while the maintenance of the river structure types are
secondary in the construction of the Fuzzy Delphi index system for urban river
ecology security (Table 1).
4. The Dynamic Simulation Model of the Urban River Ecological
System
4-1. The Causal Feedback Diagram
The urban river ecological system is a complex organic body that belongs to the
decision-making model of a multi-factor decision-making model with considerable
complexity and uncertainty. The changes that affect the urban river ecological system
condition are quite extensive in scope. In this study, the purification security and
biosecurity serve as the core in identifying the five sub-systems that overlap in
feedbacks and affect one another. Through the causal feedback diagram, the various
feedbacks loops and causations have been expressed.
1. The Discharge Security Sub-system
Table 1 Index system for urban river eco-security evaluations
Central value distance of
Sub: the expert evaluation _d, Euzynumber Total
cous Evaluation factors ; Fouees 1G Expert Famk-
. Minimum Maximum Variance Lower Upper GPO ing
value value Mean gi A"3°° timit —timit_ —-Swigi “VSS
min ‘max a b BNP
Longitudinal
connectivity of rivers 0.7368 1.3684 0.8632 _0.5008 2.0186 _5.4678_3.5678 _3.6847_16
gl lee ili irl
security
River discharge __0.3158_1.7895_0.4947_0.5474_2.3417 5.1736 4.8889 _4.1347_13
Surface runoff 0.5789 1.3158 0.7737 0.3979 3.2010 _6.1805_ 4.2809 _4.5541_7
Bio-based flow 0.3158 1.7895 0.497 _0.5557_2.3577_5.1966_4.8967_4.1942_12
Degree of Water
Quality 0.7368 3.4737 0.9158 0.4480 2.7253 5.9960_4.4763 4.3992 _10
Heavy metal
-_ concentrations in
Purification ater 0.4211 2.3158 0.6421 0.3957 3.6794 6.1805 _4.2809 4.55417
subsystem Land use 0.4211 1.2632 0.6316 0.4074 1.7667 4.3606 3.8511 3.3237 20
Embankment
vegetation structure 0.6316 1.1579 0.8737 0.5490 3.1020 6.4797 4.1260 _4.5692_6
Total population of
the city 0.7368 3.4737 _0.9158 0.4480 2.7253 5.9960 4.4763 4.39929
Sewage treatment
rate 0.5789 1.5263 0.7105 0.3788 _2.7570_5.6861 4.5792 4.340811
Per capita water
‘Water consumption 0.4211 3.5789 0.7789 0.4092 3.7110 6.3162 4.1427 __ 4.72331
Industrial sewage and
wastewater emissions 0.4737 __ 1.3158 0.6684 0.3733 3.5045 6.2111 4.2186 4.64485
Livelihood sewage
and wastewater
emissions 0.6316 2.1053 0.9684 0.5804 3.3099 6.6126 4.0921 4.67153
Riverbank
(embankment)
stability 0.5263 2.3158 0.7684 0.3994 3.4493. 6.2699 4.2212 4.6468 4
__ Riverbed sediment
Habitat st abilit 0.4211 3.5789 0.7895 0.4043 3.6144 6.2924 4.1578 4.6882 2
Degree of river
channelization 0.6316 2.6316 0.8211 0.4006 _1.7788 4.8257 _3.5898 3.398119
Water depth 0.5789 2.3684 0.7947 __0.3800__2.2707_ 5.1713 3.3009 3.581017
Water connectivity 0.4737 1.3158 0.6684 0.3733 1.7889 4.4955 3.7814 3.3552 20
Number of biological
species 0.5789 3.5789 0.7763 0.2901 __3.1700_5.8083 4.4737 4.48408
y Ratio of water to
green area 0.6842 1.2105 0.9000 0.5394 1.6339 5.1382 3.7936 3.5219 _18
River course
cross-section area__0.7368 1.3884 0.8782 0.5013 _2.2357 5.4678 3.6678 3.7326 _15
Degree of vegetation
cover 0.7895__1.7368__1.0158 0.7282 _1.9879 6.0460 _3.5149 3.8496 _14
The purpose of this system is to evaluate the river flood-prevention capacity and
clarify the indicators that affect the river detention and discharge capacity in order to
explore the effects of the river environment and structure pattern and the landscape
changes on the system changes. The flood-prevention is not one of the capacities of
rivers; however, from the perspective of urban ecological security or for the purpose
of developing an urban river ecology security evaluation system, it is an important
target in ecological security to ensure human survival and reduce the potential risks of
disasters.
Landscape changes affect the urban land’s water retention capacity and facilitate
large quantities of surface runoff water to flow into channelized rivers. They also
increase the instant peak flow. In order to reduce the frequency of urban flooding, two
directions are essential: one is to increase the river courses’ detention capacity, and the
other is to increase the urban land’s water retention capacity in order to reduce the
surface runoff. Therefore, the river ecological security and flood-prevention
sub-systems must exceed the scope of rivers or water, thereby linking the effects of
the system and urban development and presenting the detailed and realistic face of the
flood-prevention sub-systems.
This system uses river discharge as the main stock in constructing casual loop
relationships. Longitudinal connectivity means that the rate percentage of permeable
surface area of the river construction (Ladson et al » 1999 ; Yong-sheng Gao et al »
2007 ; E-nuo Wu et al » 2007). The connectivity between the longitudinal river and
surface affect the seepage of the river course flow (Landson,A.R. et al,1999 ; Kao
Yung-Shen et. al, 2007). The higher the seepage, the lower the water consumption that
needs to be transported by rivers. Similarly, the higher the permeability of the land
surface, the lower the load of river discharge; the lower the risks of floods, the higher
the river’s ecological security and number of biological species, thus a positive
relationst Rate of change of
a discharge capacity
Longitudit
connectivity
of rivers
ainfal
Planning of the
jamount of river
discharge
ne anauntot | tr Stee
Bio-based Bed sediment
fi stability
atey depth
‘Ton
Rate of change
water depth
Degree of river connectivity
5. Rate of change of channelization
vegetation cover
Fig. 1 The causal feedback diagram of the discharge security sub-system
2. The Purification Security Sub-system
This system is used to evaluate the urban rivers’ internal water purification
capacity, measure the event of environmental pollution, and understand the effects of
land use, vegetation green belt, and human intervention on the system. In the city,
livelihood and industrial wastewater flow into rivers through pipes while the
surrounding land pollutants flow into streams and rivers through the surface runoff or
groundwater. The city is burdened with the major task of sewage discharge. The urban
river water has become the carrier of pollutant transport while the rivers and
surrounding lands have become the concentration site of filth. The width of the river’s
surrounding green belt and vegetation types affect the quantity of surface pollutants,
the vegetation has the ability to metabolize pollutants, and the plants and biological
species in the river can break down a part of the pollutants and reduce the extent of
river pollution. The water quality was measured to reflect the extent of river pollution.
More seriously polluted rivers signify more insecure river ecological systems. When
the width of the green belt increases, the quantity of pollutants that flow into the river
decreases. The increased vegetation quantity and types enhance the river’s
self-purification capacity when in the face of water pollution, which in turn improves
the ecological system security.
This system has adopted the quantity of urban river pollution as the observation
value in order to construct the causal relationship. The river water quality pollutants
are constituted by the River Pollution Index and the Water Heavy Metal Index, which
jointly reflect the river water conditions and are important indicators for the
ecological security of rivers. A higher pollution index indicates lower water quality
for biological species and worse ecological security conditions, thus the negative
causation. It is characterized by an increase of the “index of the characteristic plants
along riverbanks completely cover the embankment vegetation structures” and thus
the ratio of water to green area will also increase. More biological habitats are then
provided while the increase of vegetation will enhance the heavy metal breakdown
capacity of rivers. The land use affects the land’s run off in which the common action
will improve the water quality, thus the positive cyclic relationship (An, K.-G. et al,
2002 ; Collins, K. et al,2007 ; Wang Lin, 2007) (Fig. 2)
3. The water security sub-system
This system evaluates the urban system factors that affect river water quality,
including the preservation of river water, the introduced and discharged water quality,
the pollutants affecting the water quality, and the effect of the urban environmental
policy on the sub-system. Before the issue of sustainability of resources of
environment are discussed, it had always been thought that the water supply for urban
residents was endless. However, people now understand that in order to retain
sufficient clean water for use, the convergence of the river water must be treasured in
order to lessen the burden of water purification. The re-use of recycled
wate
consumptio
Heavy metal
concentrations in
water Degreen of
Water Quality
Ratio of water
‘and green area
Land yse
Biological
species variation!
‘Amount of river
pollution
Dagree af river
hannelization
a
Fig. 2 The causal feedback diagram of the purification security sub-system
water and reduced water inputs can reduce the burden of the river ecological services
and provide the river ecological system sufficient buffer and nurture. In terms of the
river ecological security system, not only the water consumption and water quality
issues, but also urban development policies, particularly the environmental protection
policies and extension of the environmental protection related facilities, ought to be
included in the sub-system in order to explore the effects and interaction of the urban
river ecology security sub-system on the respective environmental protection policy
standards.
In this system, with the total population of the city as the main stock, which is
the characterization of the water needs, the causal loop relationship has been
constructed. The human use of river water resources involves the discharge of
wastewater after use into rivers; therefore, population increases will lead to increased
livelihood wastewater as well. The industrial wastewater is characterized by the
economic pressure on the river ecology; therefore, the higher the industrial sewage
and wastewater emissions, the worse the ecological security will be. Through sewage
treatment rate control, the river water security can be improved. Therefore, higher
sewage treatment rates and increased recycled water consumption will lead reduce the
river water supply load. Thus, reduced wastewater discharged into rivers will improve
the river water security (Fig. 3).
Rate of change of
discharge capacity. ~
Per capita water
Population of the city
variation
Lo
of
Planning of >
mount of river
Rate of population
Changes in the
amount of
Srainage
Livelihood
wastewater
‘lscharge
Sewage
treatment fate
Degreenof
Water Quality Rate of change of y
ater depth
water depth variates
Amount of river
pollution
Fig. 3 The causal feedback diagram of the water security sub-system
4. The Habitat Security Sub-system
This system examines the effects of the substantial changes of rivers in urban
development on the natural ecologies. In addition, by gaining an insight into the
biological habitat changes, the trend of the living environment changes was obtained.
The concept of “to conduct oneself virtuously” in terms of resource utilization has
jeopardized the foundation of human survival. We must not compromise the ecologies
of the species in rivers for our own survival, the significance of which is far beyond
the ideal of sustainable development. In fact, the ecological networks of river species
reflect the desirability of the environment for human survival. That is if all living
creatures die out, we cannot possibly survive. We have achieved the expected
discharge at rapid rates at the expense of the river vegetation’s natural detention
capacity, which have in turn altered the riverbank habitats and reduced the
biodiversity. Enhanced urban river ecology security can improve the habitats of
species, restore natural forms of the riverbank environment, and maintain the
environmental quality and stability in the long-run.
This system has adopted the urban river habitat water depths as the main stock.
In consideration to the factors affecting the environmental quality of the biological
habitats provided by rivers, the trend of river ecological habitat security development
has been analyzed. Rivers and pools function as habitats and breeding sites for
organisms in the water and stabilize the rivers’ physical forms to improve the habitat
quality, thus forming a positive causation. Channelization is the characterization of
the degree of human-level river course changes. The higher the degree of
channelization, the faster the flow of rivers and the more drastic the fluctuation ranges
will be. Single and straight rivers decrease the likelihood of biological survival and
decrease the river ecological habitat security, thus the negative causation of the two.
Moreover, rivers with water connectivity not only provide stable habitat water depths
for organisms, but also buffer the water consumption during flood peaks and enhance
the habitat security and discharge capacities, thus a positive correlation (Fig. 4).
Rate of change of
discharge capacity
Population of the city
varati 5
Planning
Rate of population
change
Sewage
reatment rat
‘stability
Waterdepth 5
Rate of change of wore depth
water depth waste,
Biological
Amount af river”
pol species variation
Water
Degree ofriver connectivity
channelization
Rate of change of
vegetation cover
River
‘cross-section area
Fig. 4 The causal feedback diagram of the habitat security sub-system
5. The biosecurity sub-system
The urban river biological composition and quantity have been evaluated, and
the effects of the alien species and environmental changes have been explored in order
to explore the trend of the river ecological system security changes. The quantity and
types of river organisms reflect the environmental quality of rivers, which include the
nutrient levels and the quantity of the dependent loads. If greater vegetation cover
areas can be provided, the biological habitat areas and river ecological quality can be
enhanced, thus bringing positive benefits to river biodiversity and more secure river
ecologies.
The number of biological species is the characterization of the internal and
external changes on the physical environment and biological habitats. This system has
adopted the number of biological species as the stock of the security sub-system in
order to explore the interaction of the important factors in the system that affect the
biological survival. The higher the biodiversity, the more closely related and stable the
ecological system and nutrient levels, thus the positive correlation. By increasing the
12
degree of vegetation cover, more secure habitat spaces can be provided for organisms
while more green plants can improve their water purification capacities. As the
coverage ratio increases, the biosecurity condition will improve as well, thus the
positive causal relationship between the two. At the same time, the degree of
vegetation cover affects the ratio of water to green area, thus the positive correlation
for the environmental security sub-system (Fig. 5).
Population
variation
Planning of tho
mount of river
ischarge
Amount of
drainage divide
Degreen of
Water Quality
Land yse “Amount of
and green area
Biological
species variation
Amount of river”
Number of
Riverbank biological species
vegetation structure
‘Amount of change
of vegetation cover
Degrée of river
channelization
Rate of change of
vegetation caver ‘cross-section area
Fig. 5 The causal feedback diagram of the biosecurity sub-system
4-2 The systems dynamics flow diagram
The elaboration of the causation among the urban river ecology security system
indicators provide an approach for exploring the relevance of the feedback system,
which shall serve as the basis for the dynamic model construction. In this study, the
statistical data of Taipei City’s Tamsui River from 1998 to 2009 serves as the input
source of the various exogenous variables. Based on the causal feedback loop from
the above analysis and based on the historical evolution of Taipei City’s Tamsui
River’s ecological security and calibration parameters, the model mathematical
equations were screened and setup. Indicator changes and non-linearity have been
processed using Vensim lookup, thus the establishment of the system dynamic flow
diagram (Fig. 6).
Planing ofthe
amount of river
dlscharge
Longitudinal
a ~ connectors
% Total population i
ofthecity | Seale aaa
Rate of change of
discharge capacity, Changes in ie amount of drainage
[Amount of ver] grmount of
ischarge _Mginage divide
Rateot popuation —Popslaio
pe variation
Rael
Water
Peete weiss consumption Total dschargeot
ve sewage and waste isda
Livelihood Shitace#—Nqulie flow
sewage veer = Riverbed flue
treatment rate ischrge ran stably multiplier ate
Industral sewage vere lpi
ardvasgemiions undue Retnk—sacirreabiy Werden [Seth] sensi
Degreanof Sevoge and suracenmot “eto Lae! =
Water Quality wastyemibions table function Riverbank ere Degree iver
stebity he er channlat
Riverbank oni
Amountot river TTatankvegetaion vegetation stare 7G dale clone
prt icone incon mer oon ae
iver pllation
gall — scale funtion
a Median Ratio of water and
concentrations in
water
waste emissions
muller
yater green area
ratio multiplier
Rate ol change of|
vegetation cover
[Amount
vegetation
‘Table function of
vegetation coverage ‘Amountof
vegetation cover
change
Fig. 6 The dynamic flow diagram of the urban river ecology security system
The urban river discharge security sub-system includes an level (River
discharge), one rate (Variation of discharge), eight auxiliary variables (longitudinal
connectivity of rivers, rainfall, variation of discharge capacity, the discharge
difference, surface runoff, bio-based flow, riverbed sediment stability multiplier, total
sewage and wastewater emission multiplier), and one constant (the river discharge
planning). In the dynamic flow diagram of the discharge security sub-system (Fig.
5-8), the total of river discharge is one level, which is the accumulated change in the
amount of drainage inflow. The product of river discharge and discharge rates of
change are the change values of the annual river discharge, indicating the total of the
annual river discharge. Through the longitudinal connectivity of rivers and the degree
of river channelization, the variation of discharge was measured and calculated. Then,
based no the variation of discharge, the Variation of discharge was calculated, and the
river discharge security sub-system was calculated. The main dynamic equations are
as shown in Table 2.
Table 2 The dynamic equation of the discharge security sub-system
Index type Index name Equation
Level River discharge INTEG Variation of discharge, 5.88813e+008)
Rate Variation of discharge River discharge* Variation of discharge capacity
Auxiliary Longitudinal connectivity Exogenous variables, Source Environmental Protection
variable _ of rivers) Administration, Executive Yuan, R.O.C. (Taiwan)
Auxiliary Rainfall Exogenous variables, Taipei City Government, The
variable Central Weather Bureau
Auxiliary — Variation of discharge Degree of river channelization*longitudinal
variable capacity connectivity*rainfall lookup (rainfall)* Sewage and
wastewater emissions* Total sewage and wastewater
14
emission multiplier
Auxiliary Discharge difference River discharge planning-River discharge
variable
Auxiliary Surface runoff Land use* rainfall
variable
Auxiliary — Bio-based flow River discharge*riverbed sediment stability*riverbed
variable sediment stability multiplier
Auxiliary Riverbed sediment stability Determined based on the non-linear relationship between
variable _ multiplier the riverbed sediment stability and bio-based flow
Auxiliary — Total sewage and Determined based on the non-linear relationship between
variable wastewater emission the total wastewater discharge and variation of discharge
multiplier
Constant River discharge planning — Exogenous variables, 1.3608e+009
The purification security sub-system is constituted by six auxiliary variables (the
amount of river pollution, Degree of Water Quality, heavy metal concentrations in
water, embankment vegetation structure, land use, and sewage and wastewater
emission multiplier. The amount of river pollution is the total of the Degree of Water
Quality and heavy metal concentrations in water; the land use is the product of the
different land use types and areas and permeability, resulting in disparities in the
surface runoff and changes in the heavy metal concentrations in water; the heavy
metal concentrations in water is the product of the embankment vegetation structure
and embankment vegetation structure lookup; the Degree of Water Quality are
subjects to the effects of surface runoff, ratio of water to green area, and sewage and
if first-hand data of the embankment vegetation structure
values cannot be acquired from the database, historical aerial photographs shall be
used to depict the number of trees and shrubs and the areas they occupy. The total
number of the trees and shrubs is the number of the vegetation structures. The main
dynamic equation is as shown in Table 3.
wastewater emissions.
Table 3 The dynamic equation of the purification security sub-system
Index type Index name Equation
Auxiliary
variable Amount of river pollution Heavy metal concentrations in water
Surface runoff lookup (surface runoff flow)*sewage and
Degree of Water Quality wastewater emissions*sewage and wastewater emissions
multiplier
Auxiliary Heavy metal concentrations Embankment vegetation structure*embankment vegetation
Auxiliary
variable
variable in water structure lookup (embankment vegetation structure)
Auxiliary Embankment vegetation Exogenous variables, determined based on historical aerial
variable structure photographs.
Auxiliary
variable Land use Exogenous variables, source: Taipei City Government
Determined based on the non-linear relationship between
Auxiliary | Wastewater discharge the sewage and wastewater emissions and Degree of Water
variable __ multiplier Quality.
The water security dynamic model includes one level (the total population of the
city), one rate (population variation), seven auxiliary variables (rate off population
change, per capita water consumption, sewage and wastewater emissions, sewage
treatment rate, water consumption, livelihood sewage and wastewater emissions, and
sewage and wastewater emissions). The sewage and wastewater emissions indicate
the comprehensive performance of the urban water consumption and sewage
treatment rate. At present, Taipei City’s sewage treatment rate (sewage and
wastewater pipes) has shown a yearly increase trend, and the rate reached 100% in
2010. Therefore, the sewage treatment rate after 2010 has been set as 100%. The main
dynamic equation is as shown in Table 4.
Table 4 The dynamic equation of the water security sub-system
Indicator type Indicator name Equation
Level Total population of the city INTEG (Population variation, 2.63994e+006)
Total population of the city*rate of population
Rate Population variation change/1000
Exogenous variables, source: Directorate-General of
Auxiliary Budget, Accounting and Statistics, Executive Yuan,
variable Rate of population change _R.O.C. (Taiwan)
Auxiliary Per capita water Exogenous variables, source: Water Resources Agency,
variable consumption Ministry of Economic Affairs
Auxiliary | Sewage and wastewater — Exogenous variables, source: Water Resources Agency,
variable _ emissions Ministry of Economic Affairs
Auxiliary
variable Sewage treatment rate Exogenous variables, source: Taipei City Government
Auxiliary
variable Water consumption Total population of the city*per capita water consumption
Auxiliary Livelihood sewage and Water consumption*0.8*100- Sewage treatment rate)/100
variable __ wastewater emissions
Auxiliary | Sewage and wastewater Industrial sewage and wastewater emissions + livelihood
variable __ emissions sewage and wastewater emissions
The habitat security sub-system includes one level (water depth), one rate (water
depth variation), six auxiliary variables (Rate of water depth variation, riverbed
sediment stability, the degree of river channelization, riverbank stability multiplier,
and river discharge multiplier), and one constant (water connectivity). As the riverbed
sediment stability is the product of the riverbank stability multiplier, embankment
vegetation structure, and embankment vegetation lookup, the river channelization and
water connectivity were to measure the rate of water depth variation. Based on the
rate of water depth variation, the water depth values were determined. The river
course cross-section area and river stability jointly affect the number of biological
species, which are considered important indicators of the habitat security sub-system.
The main dynamic equation is as shown in Table 5.
Table 5 The dynamic equation of the habitat security sub-system
Indicator type Indicator name Equation
Level Water depth INTEG (Water depth variation, 9.54474)
Rate
Water depth variation Water depth* Rate of water depth variation
Auxiliary Rate of water depth Water connectivity*the degree of river
variable __variation channelization*riverbed sediment stability
Auxiliary Exogenous variables: source: Environmental Protection
variable Riverbed sediment stability Administration, Executive Yuan, R.O.C. (Taiwan)
16
Riverbank stability multiplier*embankment vegetation
eri Riverbank stability structure*embankment vegetation lookup(embankment
vegetation structure)/le+006
Auxiliary Degree of river Exogenous variables; estimated and determined by the
variable channelization history aerial photographs
Determined based on the non-linear relationship between
Auxiliary Riverbank stability the embankment vegetation structure and riverbank
variable multiplier stability
Auxiliary Determined based on the non-linear relationship between
variable __ River discharge multiplier River discharge and water depth
Exogenous variables: estimated and determined based on
Constant Water connectivity historical aerial photographs
The biosecurity dynamic model contains two levels (number of biological
species and the amount of vegetation cover), two rates (variation of biological species
and the amount of change of vegetation cover), six auxiliary variable(River
cross-section, ratio of water to green area, variation of vegetation cover, water green
area ratio multiplier, river channelization multiplier, and green hydraulic multiplier).
The amount of vegetation cover, riverbank stability, number of biological species, the
amount of river pollution, bio-based flow, and ratio of water to green area affect the
biological species’ survival rates, which further affect the number of biological
species; the river cross-section is the product of the degree of river channelization and
ratio of water to green area; Degree of Water Quality that jointly affect changes in the
number of biological species were used to calculate the amount of vegetation cover
that is relative to the Degree of Water Quality. The product of the said values and the
river cross-section was used to estimate the ratio of water to green area, and the action
of the three was used to calculate the number of biological species, thus the
constitution of the biosecurity sub-system core. Based on this correlation, the dynamic
flow diagram of the biosecurity sub-system was constructed. The main dynamic
equation is as shown in Table 6.
Table 6 The dynamic model of the biosecurity sub-system
Indicator type Indicator name Equation
Number of biological
Level Species INTEG (Variation of biological species, 21)
Level Amount of vegetation cover INTEG (The amount of change of vegetation cover, 145)
Riverbank stability* Number of biological species* The
amount of vegetation cover* Vegetation cover lookup
il F biologi ‘amount of vegetation cover)* River pollution look
Variation of biological (amou veg ver)* River pollu up
Rate species (amount of river pollution)* Bio-based flow lookup
. ° (Bio-based flow)/100*(Ratio of water to green
area*10000)* Water green area ratio multiplier
Amount of change of
Rate vegetation cover Amount of vegetation cover*variation of vegetation cover
Auxiliary River course cross-section Degree of river channelization* River channelization
variable _area multiplier
Auxiliary (Amount of vegetation cover/river course cross-section
variable _Ratio of water to green area area)*Degree of Water Quality*green hydraulic multiplier
Auxiliary Embankment vegetation structure*vegetation coverage
variable _ Variation of vegetation lookup(embankment vegetation structure)/10
17
cover
Determination of the non-linear relationship based on the
Auxiliary — Water green area ratio ratio of water to green area and variation of the number of
variable multiplier biological species
Auxiliary — River channelization Determination of the non-linear relationship between river
variable _ multiplier channelization and river course cross-section area
Auxiliary Determination of the non-liner relationship between the
variable __ Green hydraulic multiplier _ratio of water to green area and Degree of Water Quality
Regarding the system dynamic model validation, Coyle, Exelby (2000) and
Sterman (2002), etc. advocate that, the system dynamic model building and the
dynamic simulation process need to inter-contrast compare repeatedly the model
simulation results and the real system quantification information, in order to
constantly test and adjust the model validity, to confirm all the behavioral
characteristics of the model, and to demonstrate the behavioral characteristics of the
real-world.
The indicator parameter confirmation is to compare the simulated parameter
values with the real system values in order to make the simulated behavior of the
model reasonable and consistent with the behavioral characteristics of the real system.
By examining the data of 12 years from 1998 to 2009, this study analyzed the mean
deviation between historical values and the simulated values to confirm the validity of
the model. This urban river ecology security dynamic simulation system with the
level as the object, through the parameter test (Table 7), all mean deviations were
below 4.54%, and the mean deviation of all observed indicators were only 1.69%.
These results satisfy the threshold requirement of “the number of indicators with
deviation <5% accounted for 70%, and the deviation of a single indicator is not larger
than 10%”, fit quite accurately the real situation of the Tamsui Rive ecology security
in Taipei City. The forecast capability for the real historical trend is fairly accurate.
4.3 Variables Sensitivity Analysis
Through the variables sensitivity calculation and analysis, this study sorted out
the variables that had larger impact on the system to understand the main reasons for
the system status changes, divided the variables into the active set and the passive set,
and observed through the “Impact - Response Matrix” (Kano Noriaki 1999) the direct
impact of strategy changes on the observed indicators. The active set variables
contains all the exogenous variables in the system, accounting for a total of 12
indicators. The passive set variables are of five levels plus bio-based flow, the amount
of water pollution, Degree of Water Quality, and the ratio of water to green area,
accounting for a total of 9 indicators. The active set variables crossed the passive set
indicators, generating 1188 data and accounting for a total of 108 kinds of changing
relationships. This study conducted tests on 20% of the variation for each variable in
the active set to observe the change extent of the passive set variables. As the value of
the change extent may show positive impact of positive value or negative impact of
the negative value, so after the change of the exogenous variables, the active set
Table 7 Comparison table of the historical value and simulated value of the Tamsui River ecological security in Taipei City
Mean
Item | Year 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 besa 2g
n
Historical
River __Yalue | 588813408 | 289080576 | 444951360 | 317196864 99498240 | 139848768 | 324298080 412177248 | 375193728 | 444737952 |_ 287871840 | 226458720 |
dischar Simulated 0.00%, 0.0000
ge (M2) value 588812992 289080352 | 444950976 | 317196608 | 99498176 |_139848672 | 324297856 412176960 |_375193472 | 444737664 | 287871648 | 226458560 |
aviation! 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00%
The [Historical
total | value 2639939 | 2641322 | 2646474 | 2633802 | 2641856 | 2627138 | _-2622472|__-2616375 | __ 2632242 | __- 2629269 2622923 | _2607428
populati Simulated
onof | value 2639940 | 2641323 | 2646475 | 2633803 | 2641857 2627139 | -2622473 | 2616376 | 2632243 | 2629270 | _—-2622924 | 2607429 | 0.00% 0.0000
the
city(per Deviation
sons) 0.00%, 0.00%, 0.00%. 0.00%, 0.00%, 0.00% 0.00% 0.00% 0.00% 0.00% 0.00% 0.00%
Historical
Water |_ value 9.54 9.34 9.53 9.30 9.01 9.08 9.17 9.14 8.82 8.50 7.69 748
depth Simulated 821 8.34 | 3.09% 0.0169
(m) | value 9.54 9.42 9.48 9.37 9.28 9.29 9.30 9.29 9.16 9.05
Deviation’ 0.00% 0.92% 0.54% 0.829 °, 9 y D 9 5 6.78% 24%
.00% 92% 54% 82% 3.07% 2.31% 1.43% 1.67% 3.89% 6.44%
‘Number Historical
of | value 21.0 34.0 14.0 24.0 45.0 36.0 19.0 59.0 48.0 35.0 21.0 43.0
biologic Simulated 0.83% | 0.0013
al value 21.0 34.0 13.8 237 448 35.9 19.2 59.8 48.5, 35.2 20.6 42.9
species( ro vintion
species) 0.00% 0.00% 1.62% 1.26% 0.52% 0.18% 1.09% 139% 1.06% 0.54% 1.92% 0.33%
The Historical
amount|_ value 145.00 161.00 150.00 152.00 149.05 161.05 183.00 177.00 172.00 188.00 196.00 202.00
of Simulated
vegetati, value 145.00 149.33 154.53 159.08 163.79 168.68 175.94 181.13 186.45 194.49 202.88 208.86 | 4.54% | 0.0330
on
cover( Deviation
mt) 0.00%, 7.25% 3.02% 4.66% 9.89%, 4.74%, 3.86% 2.33% 8.40% 3.45% 3.51% 3.40%
Note: The simulation value of this model is by chi-square test to judge its degree of similarity with the historical value, a significant level is
0.05. If? < Xo os (n-1=11)= 19.675, it will be considered that the historical value of the variable is consistent with simulation value. ~=
19
| , where, a is the simulation value of system dynamics model; B is the historical value; n is the number of tests.
variables were taken to contrast with the original simulation data, to obtain the
arithmetic mean of the absolute value of the gap in 12 years and then conduct the
contrast with the mean difference of the original simulation data in 12 years. The
evaluation method is to give 0 points to the ones that completely have no change; 1
point to the ones with mean difference less than 1/8; 2 points for between 1/8 and 1/4 ;
and 3 points for larger than 1/4.
The results show (Table 8) that the variables in the active set that has the largest
impact on the overall system in sequence are per capita water consumption, sewage
treatment rate, rainfall, the degree of river channelization, longitudinal connectivity of
rivers, and the embankment vegetation structure; in the passive set, the ones that are
most affected by this active set are in sequence ratio of water to green area, bio-based
flow, the amount of river pollution, Degree of Water Quality, and the number of
biological species. In this analysis, those variables with high PS values as highly
sensitive indicators of the table the same way the system can be inferred the impact of
higher In this analysis, those variables with high PS values as highly sensitive
indicators, likewise the indicators impact highly on trend of system change. Overall,
water security-oriented strategy implementation has the highest correlation with the
development of urban river ecology security, followed by habitat security, and the
discharge security. This study uses the sensitive indicators of embankment vegetation
structure, the degree of river channelization, longitudinal connectivity of rivers, etc. as
the bases for scenario simulation to understand the extent of the impact of the active
set variables after the implementation of strategy, and by changes of the indicators of
the number of biological species, the amount of river pollution, the amount of river
discharge, etc., to conduct evaluation of the relevant strategy (Table 8).
Table 8 The sensitivity analysis (variables - response matrix) of urban river ecology
security system
Tdustria
Longitud Embank Rate of Pet lsewage The
inal ment ‘ate of capita | and | Sewage Riverbed Water degree of
: inf The river, Land populati
Variable name connecti e Fiver esetatio| water |wastewattreatmentsedimentconnectivi| river | PS
all(*) discharge use on
vity of amie change consump, et rate(*) stability ty channeliz
riversy(*) PINE structure! 8° “tion(*)_ emission ation(*)
yey s
River discharge 5 3.0 0.0 1 3 1 3 0 0 3 17
Total population of
POP 0 0 0 0 0 2,0 0° 0° 0 0 02
the city
Water depth 2 3 0 0.0 1 3 1 1 2 2 2 17
Numberof gi ) ft} 1} 3 fafa] 3)a]3]a]a1]2 de
biological species
Amount of
0 0 0 2 0 0 0 0 0 0 0 0 2
vegetation cover
Bio-based flow 5 3.0 0 0 1 3 1 3 3 0 3 20
Amount of river 1/2 1 3 /2/1 3 1 3 1 1 1 20
pollution
Degree of Water
Quality 1 3 1 1 3 1 3 1 3 1 1 1 20
Ratio of water to 1 3 1 2 3 1 3 2 3 1 1 3 24
green area
AS 12,18 4 ll 9 9 21 8 19 9 6 15
Note: 1. AS: impact value, AS = &-line value, indicating the summation of the active
impact value of the variables.
2. PS: response value, PS = X column values, indicating the summation of the
passive impact value of variables.
20
4-4 Analysis of Strategy Scenario Simulation
This study, by goal-orientation, set up clear strategy objectives and implemented
the strategy step by step (Legasto & Maciariello, 1980). With the increase of the
number of biological species as the check basis, the improvement of river ecological
conservation function, reduction of river pollution index, the enhancement of flood
dredging capacity of the river as strategic objectives, this study observed the trends of
behavioral change after the implementation of a single and complex strategies of the
system model on embankment vegetation structure, the degree of river channelization,
longitudinal connectivity of rivers, and the trends of the changes of 9-sensitive
endogenous variable indicators. This test method can be used to observe the behavior
characteristics performed by sensitive endogenous variables under the stimulation of
the input variables in the absence of other noise interference to assist decision makers
to find the leverage points of effective strategies.
1. Single Scenario Simulation Analysis
(1) Strategy 1 - Environmental purification strategy — implementation of multi-layer
planting of embankment vegetation, to create a hydrophilic green riverbank
Taipei City proposed the “Clean, Green, Hydrophilic Taipei City” concept as the
environmental and ecological development vision in 2020. Plans have been made to
build or rebuild many riverside parks to actively promote the development and
construction of park greens. In 2009, the Tamsui River reduced to moderate pollution
(PRI value is 4.18), but was limited by river management regulations. As trees cannot
be planted within 20 meters of the surface of the water, the enhancement of plant
diversity and amount of green is limited. This study targets to increase the numerical
value of the embankment vegetation structure by 2 times in 2020, and conducted
different approaches to enhance the vegetation structure during 2011-2020 by
simulation approaches that the Strategy 1-1 to increase by 7% slowly each year,
Strategy 1-2 to increase by 70% in the initial stage, while Strategy 1-3 to increase in
small scale in the initial stage and to increase in high-intensity in medium stage.
Strategy 1-1: after the interference of raise by 7.0% embankment vegetation
structure, the amount of vegetation cover continued to increase slightly, causing the
situation of a slight rise continuously in the number of biological species, ratio of
water to green area, while the heavy metal concentrations rapidly decline in
short-term, but showing increase in medium-term and long-term; the situation that the
amount of river pollution continues to increase slightly after initial dropping explains
that the major role of the interference rise of embankment vegetation structure was to
increase the amount of woods, shrubs, although there is long-term influence on the
growth of biological species for the river, it is the short term benefits in reducing the
amount of river pollution, but for long-term and medium-term administrative
implementation to curb the pollution is disadvantageous.
21
Strategy 1-2: to implement reinforced planting of multi-layer plants for the
strategy of green embankment. This case simulated river green renovation program to
increase amount of vegetation by 70% in the initial stage, and then with the increasing
rate of 3.5% per year starting from 2016 to raise the amount of river bank vegetation.
In this case, the number of biological species shows decline in short-term (2010-2012),
increased year by year in long and medium-term, the numb of biological species in
2020 compared to Strategy 1-1 were increased by 10%, the amount of river pollution
compared to Strategy 1-1 were increased only by 2%. In the short term reinforced
green strategy, the amount of river pollution rises only in the initial stage, but in long-
and medium-term showed fluctuations in a steady state, Therefore it is not conducive
to the purification security in the initial stage, but the long and medium-term trends of
steady state can help control of the implementation of changes in river pollution,
Reinforced green strategies should take into account of the status of both biosecurity
and purification security, in implementation must propose the immediately effective
measures to respond for river pollution prevention and control, by two-pronged
approaches of the increase of planting of multi-layer plants and reduction of the
amount of pollution to ensure the effective maintenance of the river ecological
security.
Strategy 1-3: continues to carry out 3.5% of embankment vegetation structure
raise till 2015 and then implement reinforced 70% increase of embankment vegetation
structure, observed the data performance, its mutation rate is similar to Strategy | -1,
the number of biological species only slightly increased by 2013, a substantial
increase after 2015. It shows Strategy 1-3 with slow-progress way to increase
multi-layers of embankment vegetation structure can stimulate the gradual increase in
the number of biological species; though the amount of river pollution is not obvious
in an initial stage, but sustains a steady state in the final stage and will not increase
any more but show a downward trend.
Strategy 1-1, the implementation of strategy of green embankment vegetation
structure by slow-progress in 7% showed a slight increasing trend in the biological
species and ratio of water to green area in the annual period of medium term, and
showed the status of an unstable increase in the amount of river pollution, and is not
conducive to long-term operation of river pollution control. Strategy 1-2 for the
control of the amount of river pollution shows rising slowly, the relative number of
the biological species shows decreasing trend initially and then increased year by year,
the number of species in 2020 has the highest value among the three simulation
strategies; the amount of river pollution shows increase in the initial stage then getting
steady at later stage. The observation of the medium and long-term process shows that
Strategy 1-2 has a positive efficacy in increasing river ecology security. Strategy 1-3
is a slow-progress way to increase the amount of embankment vegetation structure
year by year, increase the number of biological species, so mitigation of the amount of
river pollution is conducive to the control the amount of river pollution.
22
90 F —+—s1-3
80
met) ott
7 +
—*— $1-1
6 b
50 b ze --X-- Current
40 F 82-2
30 aren
20
1 s3-2
10 F
0 n 1 ‘$3-1
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020
Figure 7 Diagram of the changing trend of the number of biological species under a
single strategy
14
——s1-3
12 7
—a—s1-2
Ly ig, 81-1
0.8 --%-- Current
0.6 fF —*#— 82-2
04 + st
02 —+— 33-2
$3-1
0 1 1 \ 1 1 1 1 1 1 1
2010 «2011 «2012-2013, 2014'S 2015. 2016) 2017, 2018 =. 2019-2020
Figure 8 Diagram of the changing trend of the amount of river pollution under a
single strategy
350000000
8
300000000. F
4
250000000. F =an0 ia
200000000. = --¥-~ Current
150000000. + 522
100000000. + Sais
50000000. + see
so-1
rt SSS
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020
Figure 9 Diagram of the changing trend of the amount of river discharge under a
single strategy
23
(2). Strategy 2 - reduce the degree of river channelization, to improve breeding habitat
_, The cement layer at the bottom of the river is removed to enhance
opportunities
vegetation, and the cement slope is replaced by a masonry slope to revive the river’s
ecosystem. All concrete ditch rivers have the possibility of ecological restoration.
However, whether or not the effectiveness of its implementation will conflict with flood
control and discharge function is to be determined. In this study, by the simulated
Strategy 2-1 reinforced de-channelization strategy in short-term, Strategy 2-2 low
intensity de-channelization strategy in long-term, contrast the trend of change in
indicator value to observe the Tamsui River ecology security changes in Taipei City.
Strategy 2-1 implementation in the short-term (2011) is to remove the concrete
structure of the river bed and river banks in low intensity, then change the river course
structure in the way of slow-progress (-5.0%) year by year, initially producing no
stimulation change in the number of species, increasing at a medium stage year by year,
the number of species in 2020 will be higher than the simulated values by 10%,
indicating that the implementation of river de-channelization engineering methods to
improve the river ecology can relatively improve the ratio of water to green area, and
has a positive efficacy for the number of biological species. The amount of river
pollution did not generate a significant interference, indicating that the change of
channelization is not a beneficial practice for the control of the amount of river
pollution. The implementation of slow-progress of de-channelization caused a slight
increase of the amount of river discharge, but the effect is not significant, only after
medium and long periods were large differences in the amount of discharge observed.
Strategy 2-2 implementation strategies remove the concrete structure in riverbeds
and riverbanks with high intensity in a short period (2011) as and Strategy 2-1 ,then
keep the river course structure in the same statue (leave 2), have it change to the
ecological engineering construction of river banks, and generated significant changes
to the number of biological species; although the number of biological species decline
in the short term, it showed the trend of continuous increase in medium and long-term,
in the final simulation value the number of biological species is much higher than the
simulation value. The ratio of water to green area increased substantially, the amount
of river discharge is also relatively increased, although in the long-term it showed a
downward trend, but its discharge value is still higher than the original simulation
value. The results show that de-channelization strategy can provide more habitats for
aquatic plants and animals; reduce the degree of channelization of riverbanks has a
positive effect on bio-security and habitat security.
(3) Strategy 3 - Strengthen the Longitudinal Connectivity of Rivers to Reduce Flood
Risk
Longitudinal connectivity of rivers characterized the change of budget of works
departments for the river dredging. Remove mud base of the river bottom and the
elimination of garbage and obstacles on river surface can increase the flow rate of
river discharge and enhance discharge capacity. River dredging for 1998-2009
changed in the amount of investment by the annual rate of change in a great scale
from -12% to 65%, and maintained an upward trend in the past two years. By 2011,
apart from maintaining a fixed amount for the original model, it is proposed to invest
1 — 0.5 times of budget through longitudinal connectivity of rivers at the initial stage,
by way of growth rate of 5% to observe urban river ecology security changes.
24
Strategy 3-1 After the implementation of longitudinal connectivity of rivers input
with | time dredging rate on annual growth rate of 2011, compared with the original
model, the amount of river discharge increased; the amount of river discharge was
stable in the initial stage, the rate of change increased gradually in the medium stage;
The rate of change reached up to 7.9% in 2020. The discharge value increased
exponentially, indicating that the change in the amount of river discharge was positive
subject to longitudinal connectivity of rivers. Bio-based flow showing an increasing
trend is a positive benefit for biological and habitat security. Analysis by the amount
of change shows no significant impact for the number of biological species and the
amount of river pollution. Strategy 3-2 after the implementation of longitudinal
connectivity of rivers with input of dredging rate 0.5 times of annual growth rate of
2011, compared with the original model, the amount of river discharge increased; the
bio-based flow and the amount of river discharge was stable in the initial stage, rate of
change increased gradually in the medium stage. The rate of change reached high up
to 5.1% in 2020; discharge amount increased with the improvement of the
longitudinal connectivity of rivers and bio-based flow showed an increasing trend, but
with a lower rate of increase than the value of simulation 7.7% as in Strategy 3-1.
Slightly declining in water depth, the simulation value of water depth is higher than
that of Strategy 3-1, but the difference is not significant, the adjustment of
longitudinal connectivity of rivers relatively reduced the height of the river water
depth.
Overall, to conduct control of longitudinal connectivity of rivers through the
river dredging rate of control, both Strategies 3-1 and 3-2 will result in a decline of
water depth indicators of the river generating a relatively upward affect to the amount
of river discharge and the amount of bio-based flow; but does not affect the indicators
of the number of biological species, the amount of river pollution, and ratio of water
to green area, etc. Compared to the river dredging rate of high-intensity, discharge
capacity of river dredging rate of low-intensity is lower, but in overall the input of
slightly and continuously increasing dredging rate has a significant effect for the
discharge security.
Found in a single scenario simulation was that to apply different strategic
measures to mobilize embankment vegetation structure, the degree of river
channelization, longitudinal connectivity of rivers will produce a number of biological
species, partly different from that the strategy expected. The strategic purpose of the
amount of water pollution, ratio of water to green area and the amount of river
discharge to a limit achieved the strategy objectives to enhance the number of
biological species, improve the environment purification, and reduce flood risk; if the
comprehensive strategy objectives is expected, a variety of strategy scenarios is
required to be provided simultaneously before multiple objectives can be achieved at
the same time.
2. Comprehensive Scenario Analysis
“Urban river ecology security” refers to the result of interaction of the
eco-system with the external interference of the environment, and more need for the
use of integrated multi-strategy mechanism to regulate state-level of the urban river
ecology security: As a single strategy input test cannot improve a variety of system
behavior simultaneously, therefore, it needs to explore the effectiveness of the
combination of a multifaceted strategy input to improve the system. In this study,
25
taking 2011 as the base period, compose the aforementioned strategies into two, to
conduct the complex strategy simulation and strategy evaluation.
(1). Complex strategy 1 (cs1) - increase the number of biological species, improve the
environment purification, and reduce flood risk
Complex strategy 1: Use this strategy to implement 7.0% slow-progress
promotion on the embankment vegetation structure, remove the degree of
channelization in low-intensity, raise longitudinal connectivity of rivers in high
intensity, resulting in the presentation of continuously and slightly increasing status on
the number of biological species, ratio of water to green area. The situation that the
amount of vegetation cover continued to slightly rise, the amount of river pollution
each year ramps-up shows the implementation of complex strategies 1 can improve
relatively ratio of water to green area, the number of biological species, but also affect
the changes in the amount of river pollution, resulting in the pressure on the control of
the river pollution at later stage. The amount of river discharge was slightly increased,
short and medium-term effects were not significant, but the long-term performance
showed a large difference in the amount of discharge.
(2) Complex strategy 2 (cs2): increase the number of biological species to improve
the environment purification, and reduce flood risk.
Complex strategy 2: Use this strategy for the green embankment to implement
planting of multi-layer plant species in high-intensity (70%), remove the concrete
structure of the riverbed and the riverbanks with high-intensity within the short term,
raise longitudinal connectivity of rivers with high-intensity. In this reinforced scenario,
the number of biological species showing status of short-term decline, medium and
long-term rise, the number of biological species in 2020 is more than that of the
Complex strategy 1 by 55%, the number of biological species in 2020 is 2.1 times of
the original simulation value (74 species increased to 155 species) showed that the
Complex strategy 2 for the number of biological species to enhance has a positive
effect, more able to increase the ratio of water to green area, is conducive to the
implementation of river conservation strategy. The amount of river pollution is more
than that of the Complex strategy 1 by only 2.7%, compared of the two the
implementation of Complex strategy 2 has a more positive effect on the indicators of
the number of biological species; the implementation of Complex strategy 2 can
improve the indicators values of the amount of river discharge, its discharge value is
lower than that of Complex strategy | by only about 5%, has a positive effect.
Overall, in comparison of simulation difference between Complex strategy 2
with powerful means and Complex strategy 1 by means of slow-progress, Complex
strategy 2 can provide the results of river management high number of biological
species, and high amount of discharge; for the amount of river pollution it shows
improvement in the initial stage and turned to be stable in the medium and long-term.
It shows the greening strategy that has been promoted for several years with the way
of slow-progress, in terms of greening construction for the river, space planning and
construction of the initial stage, controlling the planting in the performance of
ecological capacity in medium and long-term schedule, is a long-term effective
strategy. If planting seedlings set to 3-5 years as the waiting period for the root growth
of vegetation, to give the time for planting cluster to get matured, significant efficacy
26
can be seen, for construction costs and control of the ecological footprint is a better
choice. However, both Complex strategy 1 or Complex strategy 2 will cause the rise
of the amount of river pollution, Complex strategy | performed a rising trend year by
year, Complex strategy 2 after the rise in the initial stage the changes got smaller in
extent, the amount of pollution was greatly affected by land-use and foreign sewage
and waste discharge, it is proposed the implementation of these strategies should be
coped with the supportive measures for controlling the reduction of pollution sources;
the implementation of complex strategy 2 for the initial control of reduction of
pollution volume makes future control rectification of changes in smaller amplitude,
is more effective for the implementation of the strategy.
180
160 F
140
120 F
—*CS1-2
—=CSI-1
+ Curent
2010 «2011-2012, 2013, 2014-2015 2016S 2017_—««-H18-—- 2019-2020
Figure 10 Diagram for the changing trend of Number of biological species in complex
strategy simulation
al a
81-2
ee ee “Sst
06 7 + Current
2010-2011) -2012, 2013, 2014. 2015-2016) 2017s «2018» 2019-2020
Figure 11 Diagram for the changing trend of volume of river pollution in complex
strategy simulation
27
300000000
250000000" SS
—+——_4,-
200000000. aia.
—eCS1-2
150000000. F —=csh
—*— Current
100000000. F
59000000. F
2010 2011 2012 2013 2014-2015 2016-2017: «2018-2019» -2020
Figure 12 Diagram for the changing trend of the amount of river discharge in complex
strategy simulation
ret
cs
=
Lieneliang Comas =
= =
Ronan 4 ? sees
Tamsul
ante
Kinmen Cougs ARBRE ian Couey
Tamsui fo?
Beno
? shintin
SY Taishan
Sijhuang
Pin Coy snutn /
\) Tucheng \ \
OLig Stusyrea aan
Fig13 The study areas boundary
5. CONCLUSIONS
In this study, we use literature reviews, Fuzzy Delphi expert opinion survey, to
compile the indicator system for urban river ecology security by System Dynamics to
survey the causation of the Tamsui River ecology security system in Taipei City and
set up dynamic simulation model; through sensitivity analysis to grasp the key factors
of the river ecological security. After the implementation of repeated testing, rectify
the mode by means of constant testing of mode validity, to confirm all the behavioral
characteristics of the mode, to make it able to demonstrate real-world behavior, and its
validity tests including model purpose, structural behavior, confirmation of
parameters and other processes, use of simulation results to improve the effectiveness
of implementation of strategies, to achieve quality urban living environment.
This study made the strategic goal of securing efficient use of the ecological
resources of urban rivers, creating a hydrophilic green river banks, improving
breeding habitat opportunities for all organisms and reducing flood risk, the strategy
28
Wanlt
Shiging
with changing trends of 9 sensitive endogenous variables to observe system mode in
environmental purification - Conservation strategy to implement of planting
multi-layer vegetation at embankment, - Discharge strategy to reduce the degree of
river channelization, - The study of a single strategy of strengthening the longitudinal
connectivity rivers and of the trend of the behavioral changes after the implementation
of complex strategy shows that strategy simulation is viable, scientific, and flexible.
Some conclusions of this study are the following:
1. Multi-approach association can be systemized and applied to the study of urban
river ecology dynamics
Urban river ecology security is a high-order, nonlinear and time relay complex
social - economic - natural systems, with systems engineering as the framework,
through the systems engineering technology and science to construct the overall
system and sub-system; through the Fuzzy Delphi method, questionnaires and
consultation by experts effectively establish the objective, professional indicator
system. By the System Dynamics Mold method, the complex can be simplified,
clarify the causal and feedback relationship of each index variables in the system
and the impact size of variation, and be able to implement strategy evaluation as a
preventive measure, early warning facilities to become a feasible structure for the
study of the urban river ecology security monitoring system.
2. Complex Scenarios Often Have a Better Strategy Evaluation than a Single Scenario
In a single scenario simulation, by applying different strategic measures to
mobilize embankment vegetation structure, the degree of river channelization,
longitudinal connectivity of rivers will produce number of biological species, the
amount of water pollution, ratio of water to green area and the amount of river
discharge partly differed from that expected by the strategy objectives. It is more
likely to be unable to achieve the comprehensive objectives of the strategy.
Complex scenario simulation is under the implementation of a multi-strategy, each
still holds its own influence and can play a clustering effect, under the effects of
conduction and synergistic, more able to comprehensively ensure the increase of the
number of biological species, improve the environment purification and reduce the
flood risk for the achievement of strategic objectives.
3. Optimization strategy must take into account the strategic objectives and the
feasibility of limited resources
The optimization process of urban river ecology security strategy should first
establish clear objectives, through the establishment and operation of system
dynamics model, and run by the cause effect and logic of model, to identify strategy
leverage point to improve the behavior of the system. Through strategy simulation
to choose the preference to implement, in the general case, the program coordinated
facilities worth more than the implementation of better performance, but natural
resources are limited, should first look at the limitations of resources and the degree
of difficulty in implementation; and to implement accordingly will have the
optimum strategy evaluation, to achieve strategy objectives successfully.
29
REFERENCES
An, K.-G., Park, S. S., & Shin, J.-Y. 2002, (An evaluation of a river health using
the index of biological integrity along with relations to chemical and habitat
conditions", (Environment International) , Vol. 28, No. 5, pp 411-420.
Barlas, Y. & Carpenter, S. , 1990,"Philosophical Root of Model Validation: Two
Paradigm", «System Dynamics Review) ,Vol. 6 , pp 148-166.
Cheng, C.-B. ,2005,"Fuzzy process control: construction of control charts with fuzzy
numbers ", (Fuzzy Sets and Systems) , No. 154 , pp 287-303.
Ching-Ming ,Wang. ,1992, (Rivers conservation) , pp.189 , National Museum of
Natural Science, Taipei,
Chiung-Ying,Kuo., Yu-Ren,Guo.,Da-Yuan,Lin.,1995, (The Guidelines of River
Field Ecological Design) ,Environmental Protection Administration Executive
Yuan, ROC. (Taiwan), Taipei.
Chiung-Ying,Kuo.,2001, « Ecological considerations of River Design) , from
http://www.tbroc.gov.tw/tbroc99 _3w/EcoEM/(at : 11.March.2009)
Collins, K., Blackmore, C., Morris, D., and Watson, D. ,2007, "A systemic approach
to managing multiple perspectives and stakeholding in water catchments: some
findings from three UK case studies ", (Environmental Science & Policy) , Vol. 10,
No. 6, pp 564-574.
Coyle, R.G. , 1996, (System Dynamics Modeling: A Practical Approach) , Chapman
& Hall, New York.
Ho, Y. F., & Wang, H. L. ,2008, "Applying Fuzzy Delphi Method to Select the
Variables of a Sustainable Urban System Dynamics Model ", Paper presented at the
Proceeding of (the 26th International Conference of the System Dynamics
Society) , Athens, PATRAS University ,Greece.
Ho, Y.F., Wang, H.L., Wang, H.C. & Liu, C.C. , 2009, "A System Dynamics Model
for The Sustainable Management of Theme Park", « Proceeding of the 27th
International Conference of the System Dynamics Society) , Albuquerque, New
Mexico, USA.
Ishikawa, A., Amagasa T.S., Tomizawa G., Tatsuta R. and Mieno H. , 1993, "The
Max-Min Delphi method and fuzzy Deliph method viafuzzy in tegration ", (Fuzzy
Sets and System) ” , No. 55, pp 241-253.
Ladson, A. R., White, L. J., Doolan, J. A., Finlayson, B. L., Hart, B. T., Lake, P. S.,
et al. , 1999, Development and testing of an Index of Stream Condition for waterway
management in Australia, (Freshwater Biology) , Vol. 41, No 2 ,pp 453-468.
Legasto, A.A. & Maciariello, J. , 1980, "System Dynamics: A Critical Review",
(Studies in the Management Science) , Vol. 14, pp 23-43.
Tong ,Qing-Bin, Wu, Ming-Jin, Li,Ming-Xu, Hung, Nian-Min, (Establishment of
Indicators and Framework of an Early Warning System for the Climate Change
Impacts on Sustainability of River Assimilative Capacity) ,Research Project report
on the outcome,National Science Council Executive Yuan, ROC. (Taiwan), Taipei.
Sterman, J.D. , 2000, (Business Dynamics: Systems Thinking and Modeling for a
Complex World) , Irwin McGraw-Hill.
Wang, Y.-K., Stevenson, R. J. ,2002, "An ecological assessment framework for
watershed management ", (Journal of the Chinese Institute of Environmental
Engineering) , No.12, pp 209-215.
Wang, H., Ho, Y., Wu, C. and Chen, K. ,2007, "Fuzzy evaluation method of virtual
30
"
reality for urban landscape design ", Paper presented at the (the International
association of socienties of design research) ,Hongkong.
Wang, Lin,Gong Zhao-guo, Zhang Jiong, Li Yong-chang,2007,”Comprehensive
Index System for Evaluation of Urban River Ecosystem Health”, (China Water &
Wastewater) ,Vol.23,No10,PP.97-100.
Wang, H.-L., and Wu, Y.-Y. ,2006, "The Construction of Urban Ecotourism
Indicators-Application of Fuzzy Process Control ", (The 9th World Leisure
Congress) .
Wu,E-nuo, Che,Yue, Yang Kai,2008, "Conceptualization and Characterization of
River Health from a Distinct Criteria Approach by Using Content Analysis",
Resources and Environment in the Yangtze Basin, Vol.17,No.6, pp.932-938.
Wu, Ching-I, 2011, (The Early Warning System Dynamics Simulation Model of
Urban River Eco-Security) ,Thesis of Phd.,Graduate institute of Architecture and
Urban Design Chaoyang University of Technology.
"
Wu, Yi-Yen and Wang,Hsiao-Lin, ,2007,"Urban Ecotourism,a contradiction?
(International Ecotourism Monthly) , No. 90, pp 8-9.
Wu, Yi-Yen,2008, (The Study of Urban Ecotourism System ) ,Thesis of
Phd.,Graduate institute of Architecture and Urban Design Chaoyang University of
Technology.
Xie,F. Zhang,G.,Zhou, Q., 2007, "The Influence of Ecological Security on City
Ecosystem Service Value" J. Ecological Economy ,No.5S (2007),PP.362-364.
Yong-sheng Gao, Hao Wang,Fang + Wang,2007, "Construction of Evaluation Index
System for River's Healthy Life", Advances in Water Science, Vol.18 ,No.2,
pp.252-257.
Zhao, R. and R.Govind. ,1991 , "Algebraic Characteristics of Extended Fuzzy
Number", (Information Sciences) , No. 54, pp 103-130.
31
