A System Dynamics Model for the Investigation of Additional Source of Raw Water
from the Reclaimed Effluent W ater from a Constructed Wetland Domestic
Sewage Treatment Plant in the University of Lagos, Nigeria
Adelere Ezekiel Adeniran
Civil & Environmental Engineering Department and Director of Works
University of Lagos, Nigeria
eadeniran@unilag.edu.ng
Abstract
There exist a gap in the demand and supply of water supply to the University of Lagos,
Nigeria (Unilag). The University mainly depends on internal boreholes and municipal
supply (the Lagos State Water cooperation) as sources of water supply to the University.
While a number of boreholes serve as the source of raw water to the University's water
treatment plants, the municipal water is pumped directly for distribution. In addition to
water shortages that do arise occasionally from these sources, the combined quantities of
the internal and municipal water supply are far below the current water demand of the
University. It has been established, from another study, that the quality of water from the
University's constructed wetland based domestic sewage treatment plant (CWDSTP) is
acceptable as source of raw water for further treatment. In this study, a System Dynamic
Model is developed to examine the quantity and impact of the water reclaimed from the
CWDSTP in reducing water stress in the Unilag by closing the existing gap between water
demand and supply. The results obtained from the study shows that 76.2% of the
University's water demand can be met as against the current 42.6% supply level. An
additional 3,141m3 of raw water can be added to the available raw water sources.
Keyword: Water supply, water stress, reclaimed water reuse, supply and demand gap.
INTRODUCTION
The importance of water supply to a community cannot be over emphasized. Both
humans and every living thing need water to survive. Water covers over 70% of the
Earth. The conservation of water is a global issue as global water crisis deepens Fishman
(2011). The University of Lagos (Unilag), Nigeria, one of the foremost Universities in
Africa was established in 1962. The current population of the University is about 85,000
with no significant improvement made to the water supply system over the years. The
water demand has risen from a figure of 2.48 million litres per day (mlpd) 1991 to 10.75
mlpd 2013, whereas the water supply situation has declined to 5.051 mlpd in 2013. This
has led to a serious gap of about 5.70 mlpd between water supply and demand in the
University (Adeniran et al, 2013). The two sources of water supply to the University
(Municipal supply and internal treatment facilities are can no longer cope with the rising
demand for water. There is therefore the need to find additional sources of water supply
to increase the current supply level. An option yet to be taken advantage of is the
reclaimed effluent from the University's Constructed Wetland Sewage Treatment Plant.
Water reuse has gained prominence in recent time and reclaimed water is an important
component of water management. Reclaimed water is derived from domestic wastewater,
industrial wastewater and storm runoff that have been processed or treated. The process of
reclaiming water, sometimes called water recycling or water reuse, involves a highly
engineered, multi-step treatment process that speeds up nature's restoration of water
1
quality. The process provides a high-level of disinfection and reliability to ensure that only
water meeting stringent requirements leaves the treatment facility A my et al., (2005). The
treatment of wastewater for reuse in the drinking water system of Windhoek ,Namibia was
found, after a three-month trial period, to be of exceptional high quality measured by
national and intemational water quality criteria with respect to organics, particle(turbidity)
and bacterial(faecal coliform) removal (Menge, 2005). Also Adeniran (2011) reported the
reuse of reclaimed domestic sewage effluent under tropical conditions for toilet flushing,
flower wetting and catfish farming. Reused water is usually a constant and reliable supply,
particularly with sources such as treated sewage effluent or industrial discharges. Many
waters suitable for reuse are produced in large volumes, which if not used, would be
merely discharged into the environment or the receiving water bodies. In addition, the
reuse of wastewaters for purposes such as agricultural irrigation reduces the amount of
water that needs to be extracted from environmental water sources (Gregory 2000,
USEPA 1992). The use of recycled water for drinking is less common because many
people are repelled by the thought of water that has been in our toilets going to our taps.
But a few countries like Singapore, Australia and Namibia, and states such as California,
Virginia and New Mexico are already drinking recycled water, demonstrating that
reclaimed wastewater can be safe and clean, and help ease water shortages. Eighty percent
(80%) of public water supply systems rely to some extent on ground water, which is a
form of recycled water in the natural water cycle. Raw water from a selected source
should be of sufficient quality and quantity such that it can be economically treated to
produce finished water which complies with the potable water quality requirements.
Factors that influence the choice of the raw water source should include reliability,
treatability, environmental impact, and economics. Raw water characteristics such as
microbiological quality, turbidity, pH, alkalinity, colour, Total Organic Carbon, Total
Suspended Solids, iron, manganese, algal counts and temperature determines the type and
extent of treatment required for a particular source of water (U.S. EPA, 1992). In the
University of Lagos, Nigeria underground water sources constitute 60% of the water
supply to the University, the remaining 40% comes from the municipal source which has
to be paid for. The combined internal and external sources can only meet about 46.89%
(5,041m3/day) of the estimated current water demand (10,750m3/day) of the University.
Adeniran et al (2013) have confirmed that reclaimed effluent water from the constructed
wetland is suitable for reuse as additional source of raw water for the University's water
supply system.
The objective of this work is to apply System Dynamics (SD) model in the examination of
water that can be reclaimed from the constructed wetland sewage treatment plant with aim
of reducing the existing water stress on the campus.
MATERIALS AND METHODS
The Study area
The study area, the University of Lagos, Lagos Nigeria, is located in the South Western
part of Nigeria on geographical coordinates of 6° 27' 11" North, 3° 23' 45" East and is
located in the heart of Lagos metropolis and has a direct link to the Lagos lagoon. Figure 1
shows the map of the University relative to the continent of A frica.
Figure 1: Map of the University of Lagos
University Of Lagos Water and W Tr Sy
Water Supply Treatment System
The processes involved in the production and supply of water in order to (i) make it
suitable for human consumption and (ii) make it available at the various end users; include
a complex of physical, chemical, biological and mechanical methods Twort et al. (1994).
The water treatment processing in the University involves not only purification and
removal of various unwanted and harmful impurities, but also transportation with the aid
of prime movers through conduits as well storage in specially designed pressure vessels
and tanks. The methods adopted in processing water include (a) those aimed at improving
organoleptic properties of water (clarification, decoloration, and deodorization), (b) those
which ensure epidemiological safety (chlorination) and (c) those by which the mineral
composition of water is conditioned (softening). The method of water processing is
chosen upon preliminary examination of the composition and properties of the raw water
source to be used and comparison of these data with the standard specification expected of
the final processed water. A section through the Water Treatment and Supply system is
shown in Figure 2.
‘RAW WATER PRODUCTION DISTRIBUTION
SUBSYSTEM SUBSYSTEM SUB-SYSTEM
Figure 2: Section through Unilag Water Supply System.
Wastewater Treatment and Reclaim System
The wastewater generated from the distribution system is then processed through a
constructed wetland based sewage treatment plant. The wastewater from the University
community are conveyed in sewers ranging from 100mm to 200mm diameter from homes,
hostels, offices, classrooms and laboratory to the central sewage pumping sump located at
Service Area. The wastewater is pumped to and held in the oxidation ponds, which are
planted with water hyacinth to prevent mosquito infestation and to increase the quality of
the sewage influent to the anaerobic reactor (Septic Tank). Large particles are screened off
in a primary treatment chamber containing stainless steel screen. The pretreated water is
further treated under anaerobic condition in a purposely designed Septic Tank. The
effluent from the Septic Tank is then channelled into the constructed wetland system
through a 150mm sewer pipe. The wetland was achieved in concrete with waterproof
underlay to prevent pollution to the underground water and eliminate water infiltration
into ground water. The total area of the nine (9No.) constructed wetland cells is 1540m2
with an average depth of 0.65m. Water hyacinth is planted on the influent sewage cell
while cyperus papyrus is planted on the remaining cells containing sand of average grade
size 0.1mm to 0.35mm. Figure 3 is the layout of the Constructed Wetland Sewage
Treatment system.
FROM SEWER LINES
‘Scwage Flow Line ———* —_istodge Flow Lin
Figure 3: Layout of Unilag Constructed Wetland Sewage Treatment Plant
System Dynamics Modeling
With the complexity involved in the water supply and wastewater treatment systems,
there is the need for evolving a tool that can capture the complexity of water production
and wastewater treatment and effluent reclaim variables. A System Dynamics modelling
technique was therefore adopted to capture the dynamics of the system.
System dynamics models are causal mathematical models (Barlas, 1996). In
system dynamics modelling (SDM) the underlying premise is that the structure of
a system gives rise to its observable and thus predictable behaviour (Forrester,
1968, 1987). The first step in any system dynamics modelling project is to
determine the system structure consisting of positive and negative relationships
between variables, feedback loops, system archetypes, and delays (Sterman, 2000;
Wolstenholme, 2004). This understanding of system structure requires a focus on
the system as a whole. Holistic system understanding is a necessary condition for
effective learning and management of complex systems as well as consensus
building. These are important goals in their own right. Additionally, systems
modelling and simulation supports policy analysis and evaluation (Morecroft,
1992). System dynamics allows simple ideas to be combined into models of
complex systems and processes; it makes the integration of modeling and
experimentation a simple matter (Adeniran, 2013). In particular, SDM involves:
(i) Defining problems dynamically, in terms of graphs over time;
(ii) Striving for an endogenous, behavioral view of the significant dynamics of a
system, a focus inward on the characteristics of a system that themselves
generate or exacerbate the perceived problem;
(iii) Thinking of all concepts in the real system as continuous quantities
interconnected in loops of information feedback and circular causality;
(iv) Stocks or accumulations (levels) in the system and their inflows and outflows
(rates);
(v) Formulating a behavioral model capable of reproducing, by itself, the
dynamics problem of concem. The model is usually a computer simulation
model expressed in nonlinear equations, but is occasionally left un- quantified
as a diagram capturing the stock-and-flow/causal feedback structure of the
system; Deriving understandings and applicable policy insights from the
resulting model; and
(vi) Implementing changes resulting from model-based understandings and
insights (Richardson and Andersen, 2010).
The principles of SD are well suited for modeling and application to water resources and
environmental problems (Fletcher et al., 1998, Ford 1999) and (Deaton and Winebrake,
2001). This fact is corroborated by Nirmalakhandan (2002). The behaviour in space can
also be simulated using SD framework (Tangirala et al, 2003). Huang and Chang (2003)
described SD as an emerging tool with great potentials for improved understanding of
environmental systems. It has also been successfully deployed to model the strategic
planning of the University of Ibadan water supply system (A deniran and Bamiro, 2010).
Modeling C oncept for the Water Supply and Effluent Reclaim System
A System Dynamics approach was adopted in the formulation of the simulation model.
System dynamics tools allow for an intuitive approach to the modeling of dynamical
systems from any field of knowledge where there exists, a feed-back situation. System
Dynamics modelling can easily be deployed even by young learners (Forrester, 1992).
According to Donella Meadows (1991), “System dynamics is a software-based technique
for thinking and computer modeling that helps its practitioners begin to understand
complex systems—systems such as the human body or the national economy or the earth's
climate. Systems tools help us keep track of multiple interconnections; they help us see
things as a whole.” System Dynamics modelling has attracted considerable attention over
the past few decades. In a particular sense, System Dynamics is concerned with the use of
models and modelling techniques to analyze complex systems and policy issues with a
view of getting the right combination of scenarios to accomplish an efficient strategic
planning for the system being modeled (A deniran, 2013).
The concept of the model consists of interconnectivities and feed back loops between the
major factors of population, water production, water distribution, wastewater production
and effluent or wastewater reclaimed (Figure 4).
Community
Population
: Water
Raw Water Treatment
System System
Constructed
Wetland Water
Wastewater it
Wetland System
System ‘
Figure 4: Concept of Water Supply and Reclaimed Water Model
Model Development
VENSIM platform was then used to capture the contributing stocks and other variables of
the system (Figure 5)
Municipal Supply
Quantity
fy Domestic and Reclaimed
Laboratory Use Wat ‘Municipal Supply
Water In
¢ Distribution Fen}
= ‘Network fit
Ground Water
Supply
8 A Fire Demand Laboratory] /Disodge
% jemani Backwash Water v
viphieeet Ground Water
Demand
onc: SO
ett inflow oN,
( — retirements and
cans
i disengagements
admissions new appointme siting scholars births expulsions
Figure 5: System Dynamics Model for the Reclaimed Water Model
APPLICATION OF THE MODEL AND RESULTS
The primary objective of the model is to investigate as to the quantity of water that can be
added to the existing water supply if the reclaimed water from the constructed wetland sewage
treatment plant is utilized and to examine the impact of the addition on the current water stress
on the campus. The following scenarios were investigated with the model.
Scenario 1: Current situation where the reclaimed is not utilized.
Scenario 2: A situation where the reclaimed water is utilized as additional source of raw
water.
The graph obtained from the VENSIM modeling platform is as shown in Figure 6. The result of
the scenario simulations is also shown in Table 1.
Figure 6: Output Graph of Scenario Investigation
(Scenario 1- No Reclaim Water; Scenario 2 - 100% Reclaimed Water)
Water Production
3M
2.25 M
g 15M
750,000
0
73 146 219 292 365
Time (Day)
Water Production : scenario2
Water Production : scenario!
Table 1: Water Supply Availability Before and After Reclaimed Water Re-Use
Scenario | Reclaimed | Annual Water Daily Water Current %
Water Production Production Water Demand Water
Cum. Cum. Availability
Cum.
1 0% 1,840,000 5,051 10,750 46.89
2 100% 2,990,000 8,192 10,750 76.20
It is seen from Figure 5 and Table 1.0 that the water stress would be reduced if the effluent
from the constructed wetland can be reclaimed as additional source of raw water.
CONCLUSION
A system dynamics model has been developed to integrate the water supply system with
the sewage system of the University of Lagos, Nigeria. The versatility of the SD model
was utilized to examine the impact that the reuse of reclaimed effluent water would have
on the current water stress on the campus. It is easily seen that System Dynamics can be
use to suggest alternative solutions to water resources problems. The model developed is
capable of other simulation investigations including water in the distribution, population
and wastewater treatment.
References
1;
Adeniran A. E. (2013): System Dynamics as a Viable Tool for Engineering
Education, Proceedings of African Regional Conference on Engineering Education
(ARCEE 2013), University of Lagos, Nigeria, 9-12 September, 2013.
Adeniran A. E. and Bamiro O. A. (2010): A System Dynamics Strategic Planning
Model for a Municipal Water Supply Scheme, International Conference of the
System Dynamics Society, Soul Korea, 2010.
Adeniran A. E., Aina A.T. and Oshunrinade O. O. (2013): Investigation of
Reclaimed Effluent Water from a Constructed Wetland Domestic Sewage
Treatment Plant as Additional Source of Raw Water at the University of Lagos,
Nigeria. Int. J. Environ. Sci.Toxic.. Vol. 1(1): 1-11, 2013.
Adeniran AE (2011): The effect of Water Hyacinth in the Treatment of Domestic
Sewage in an African University. Proceedings of the AWRA Annual Conference,
Albuquerque, NM, USA November, 7-10. American Water Works Association
(2012).
Amy G, Asano T, Gerba C, Haas C, Rose J (2005): Chemical and Microbial
concerns in Wastewater Reclamation/Reuse. Presentation at a strategy workshop of
the Global Water Research Coalition (GWRC) on the Role and Status of Water
Reuse-An International Review. Utrecht, the Netherlands.
Deaton, M. L., and J. I. Winebrake (1999): Dynamic Modeling of Environmental
Systems, Springer-Verlag, New Y ork, USA.
Fishman C (2011): The Big Thirst - The Secret Life and Turbulent Future of
Water. Free Press, New Y ork.
Fletcher, EJ. (1998): The use of system dynamics as a decision support tool for the
management of surface water resources. First Intemational Conference on New
Information Technologies for Decision Making in Civil Engineering Montreal,
Canada: 909-920.
Ford, A. (1999): Modeling the environment- an introduction to system dynamics
modeling of environmental systems. Island press, Washington, DC, 1999.
. Forrester J.W. (1968): Principles of Systems. Productivity Press, Portland, Oregon,
USA, 1968.
. Forrester J.W. (1987): Lessons from system dynamics modelling. System Dynamics
Review, 3(2):136-149, Summer 1987.
. Forrester, J. W. (1992): Road Map 1 - System Dynamics and Learner-C entered-
Learning in Kindergarten through 12th Grade Education. MIT System Dynamics
in Education Project, 1992
1
14,
15.
1
1
1
19.
2
21.
22.
i)
2
2
w
oOo NS
S
= oO
Gregory A (2000): Strategic direction of Water Recycling in Sydney. In Water
Recycling Australia. Proceedings of the First Symposium Water Recycling.
Adelaide, Australia 19-20, October, 2000.
Huang, G. H. and N. B. Chang, (2003): Perspectives of Environmental Informatics
and Systems Analysis, Journal of Environmental Informatics, 1, 1-6, 2003.
Meadows, D. H. (1991): System Dynamics meets the press. in The Global Citizen.
Island Press, Washington, 1991.
Menge J (2005): Treatment of wastewater for reuse in the drinking water system of
Windhoek. City of Windhoek, PO Box 59, Windhoek, Namibia.
Morecroft J.D.W (1992): Executive knowledge, models and learning. European
Journal of Operational Research, 59:9-27, 1992.
National Water Research Institute (1995): Water from Water: Recycling at
Fountain Valley, California, 1995.
Richardson, G.P. and Andersen, D. F. (2010): Systems Thinking, Mapping, and
Modeling for Group Decision and Negotiation, Handbook for Group Decision and
Negotiation, C Eden and DN Kilgour, eds. Dordrecht: Springer, 2010, pp. 313-324,
2010.
Sterman J. D. (2000): Business Dynamics: Systems Thinking and
Modelling for a Complex World. McGraw-Hill, Boston, MA, 2000.
Twort, A.C., F.M. Law, F.W. Crowley and D.D. Ratnayama (1994): Water
Supply (4" edn), Edward Arnold, Hodder Headline Group, London, UK.
U.S. EPA, Offices of Water and Wastewater and compliance (1992): Guidelines for
water reuse. U.S. EPA, Washington.
UNEP-IETC (1999): Innovative Ways of Finding Water, Proceedings of the
International Synposium on Efficient Water Use in Urban Areas, Kobe Japan 1999.
Wolstenholme E. F (2004): Using generic system archetypes to support
thinking and modelling. System Dynamics Review, 20(4):341-356, 2004.
doi: 10.1002/ sdr.302.
Recycled Content Products and Imported Building Materials
A leading concern for San Francisco‘s recovery is how and where construction material will be
acquired following a disaster, when building materials will be in great demand. It is assumed that
San Francisco traditionally purchases imported construction material, either from outside the
country or outside the locale, for its building construction needs. Contrarily, recycled content
products (RCPs) are assumed to be locally processed, generating local revenues and local
demand for housing construction. Causal loop diagrams explain the notions and assumptions of
RCPs versus imported products (Figures 5-6).
Local Material
Zz Processing \ » wtted —\
- Material Demand
( BA Time to Construct
\ pe Sy
\ Local Benefits
\
“Local Profit
Figure 5 — Causal Loop Diagram, Imported Materials
RCP Use Policy
Ss
%
Local Material __—»Imported
Processing = + a + Materials —~
fo \
if \
| \
Material Demand ~ ay
43) Time to Construct
Local Suppliers \ h- Sine j
weal Soil \ Imports )
\ /
5 a
‘Local Profi New Housing
Figure 6 — Causal Loop Diagram, Recycled Content Products
In the CLDs above, the reinforcing cycle show that the use of imported material goods decrease
the time to construct, since this is a familiar method of acquiring building products, and
eventually increase new housing and material demand. However, it prevents damaged local
economies from benefiting through economic stimulus by means of production and supply
chains manufactured regionally. Therefore, decreasing the amount of imported materials and
increasing the RCP inventory shifts power to the local economy, but not without the cost of a
slower recovery. The benefit for providing more jobs, being eco-responsible and stimulating
growth in a damaged city may provide enough incentive to tolerate a longer recovery period.
It should be clarified that not all disaster debris is recyclable, meaning that some percentage will
always be either unrecyclable or lost in processing or maneuvering. For this research, a
maximum percentage of recyclability is calculated to be 72.5% (Appendix A). Material that is
not salvaged or reprocessed is then dumped to landfill. Also, care must be taken to separate
materials containing household hazardous waste, asbestos, treated wood and lead-based paint for
reasons of contaminating mixed and recyclable debris material. In the simplified models used for
this study, debris handling is described to have two immediate end-life options, either these are
sent to landfill or sent to be reprocessed as new material. However, some debris material can be
used as fuel, which has historically been a viable alternative of waste management. Urban woody
debris is oftentimes chipped and used as biofuel, creating opportunity for waste-to-energy
streams. For the purposes of a simplified study, this option for waste management is defined to
be beyond the scope of research.
System Dynamics Analysis of San Francisco
A conceptual understanding of the simulated model begins with a CLD that links end-of-life
housing units to debris to new housing in virtuous or reinforcing loop. Adding the exogenous
factor of the 7.2 magnitude earthquake can accelerate these trends increasing new housing,
contingent on several other factors further detailed in the driving system dynamics models. The
simplified CLD shows potential for positive growth of housing using recycled content products
for building materials to arrive at pre-disaster habitation levels. Each variable within has its own
set of influencing variables. For example, New Housing” is also affected by construction rates,
construction delays, and contractor availability, for example. These inherent factors impede the
-virtuosity” of this reinforcing loop, causing delays and complexities to the system at large
(Figure 7).
Materials ——,"
Consumption New Housing
/ 4
R
Building Materials te) %
+ Debris Recycling V :
for New Construction End of Life Earthquake
rs
Debris Generated
Figure 7 — Causal Loop Diagram of Simulated Systems
Driving Models
Two driving System Dynamics models exist in simulating the hypothesis; housing units and construction and demolition waste. These
two streams are essential in understanding debris removal and material use conditions, as well as residential housing recovery.
Boundaries have been established in defining the model, and the perceived critical endogenous and exogenous variables have been
included. The following is a visual and verbal description of the model with its various components. Using these models, a pre-
earthquake equilibrium set of data is examined, along with a system impulse by an earthquake disruption. In order that results are
comparable, control variables are set with values and explanations illustrated here for the base case models.
Pulse Quantity 1
Pulse Time Y D B
Earthquake
Zo
Local RCP use Units destroyed by Units to be Demolished Deconstructed
Policy Unit/Ton Earthquake
\ {
= \ Surveying — |
~ \
| ae ne ee,
r Construction noeg ea emolishing ae 4 " ‘truction Rate
SF Housit ie Demolish Rate
sewing | | Cansiteson | TTT nae
-
Stn 7 hac sac
{ Sone Delay in
Percent Deconstructing,
Demolished. Policy for
___Deconstruetion
Demolished Units Deconstructed Units
A ic
Figure 8 — Base Case Model 1, Housing Unit Driving Model
. 1 Unit Lookup f¢
Units Collapsed by Transter Station
Capacity Inerease
‘Tons/Month of C&D. ‘Normal Fractional
ource C&D from EQ Growth Rate
Effect of Capacity on
ED from ) Growth Rate ee
Desirueted Buildin r
C&D on Site C&D MRF Capacity
Rate of C&D |___+ Fraction Used Capacity Growth Decline
fect on — Z Delay in C&D. iia
| Decline Rate
Road Gees PiransiQ®— Effector TS,
gee Capacity — =
G Se Maximum TS
C&D to be Capacity
‘Transfer Delay——_, |
toLF >
FIM To Lanafit \ oa
Rate of
‘ Percent C&D to ‘Separations nent SbO Reaver
De/Accolerati pel _—____——— (Mandated)
Ble eee
Recovered Materials Gareratbte o
a¥ tae
D
Delay in :
proce Rate of Recycling Recyclable (Ave)
Policy for Amount
=~ Processed Per Montt
Landfill Accumulation [Recycled Building Materials
‘Sold to Market
Figure 9 — Base Case Model 2, Construction and Demolition Waste Streams Driving Model
Housing Units Base Case Model Description (Figure 8)
A. Representation of housing construction, with variable labeled -Eocal Recycled Content
Policy Use Policy” for using specific amounts of imported material versus recycled
content building products. Construction halt is the stopping and slow progression of
residential construction following the earthquake.
B. End-of-life streams for housing, whether caused by -act of God” or old housing age.
C. Representation of -destruction” stream. These include a surveying process by which a
unit is deemed safe or unsafe, and two consequential flows for destruction — that of
demolition and that of deconstruction. Policy for deconstruction is a percentage of units
that are deconstructed, the rest assumed to be demolished.
D. The earthquake pulse is an 85,000 housing unit decrease from 330,000 units at month 12.
-Local RCP Policy” is the construction flow representing the lever that adjusts between
imported, virgin materials (traditional construction) and Recycled Content Building Materials
(debris reprocessing). The latter variable is possibly the most essential in analyzing the
hypothetical situation, that of sorting and processing debris as new building materials for housing
refurbishment. -kocal RCP Policy” is set as a percentage representative of the amount of
material that is imported for building construction. In providing a control mode that other
variables can be tested and compared against, an RCP policy measure of 25% is maintained,
meaning 25% of the construction material is imported and 75% is recycled content from debris
matter. Assuming that some amount of imported material is required in all cases, 25% represents
that control value of imported building construction material, with trials of higher and lower
values in additional simulation runs. The upper and lower bounds observed for —kocal RCP
Policy” are 75% and 10%, respectively.
Policy for Deconstruction” is also a percentage representing the amount of units to be
deconstructed, versus those that will be demolished. This is set at a control mode of 40%,
indicating that more units will typically be demolished. This value is an approximation based on
the density of San Franciscan neighborhoods and city demolition requirements for compacted
and careful tear-downs of buildings [23]. These tend to resemble deconstruction techniques more
than traditional, demolition-ball destruction methods.
Construction and Demolition Waste Stream Base Model Description (Figure 9)
E. Accumulation of Construction and Demolition waste on site. Within this micro-stock and
flow are included construction and demolition generated from building collapse debris
and that from demolition and deconstruction of units. A constant flow of construction and
demolition from source is estimated to be 5,760 tons per month [24], and is varied post-
earthquake as one assumes that normal C&D generating processes will be stymied or
slowed in an extreme condition after a disaster.
F. Capacity growth and total Construction and Demolition/Materials Recycling Facility
Capacity stock and flow stream for five transfer stations of concern. The amount of C&D
Capacity varies as the model is affected by pulses and policies, and is useful in
determining necessary capacity and processing requirements for debris removal. As can
be noted, an increase in C&D Capacity increases the amount of debris transfer since it is
understood that debris transfer possibilities are accelerated as space for debris staging and
processing, as well as labor, is increased. This effectively speeds up recovery time, but
my nominal amounts.
G. This portion of the model describes the transfer and processing of debris and its ultimate
destination — landfill or to recycled content material supply chains. The RCP materials
are then used as construction material in the Housing Units model, and behave as a nexus
between the two driving models.
Two important elements within the site-to-sorting station transfer rate are -Road
Clearance” and -Fransfer Delay”. Road Clearance is the amount of road impediment due
to the debris generated after the seismic disruption. In its equilibrium state, the Road
Clearance = 1. Based off of GIS calculations and Associated Bay Area Governments
(ABAG) data, the total road mileage that is affected due to the earthquake nears 25%
[25]. From the Loma Prieta earthquake, it is noted that on average, it took 134 days to
clear the roads. However, since the 7.2 magnitude San Andreas earthquake is expected to
generate far more debris than the Loma Prieta, this value has been increased to a year‘s
worth of clearance time.
Transfer Delay is the amount of time it takes for trucks to deliver debris to
staging/material recycling facilities from sites of construction or demolition and debris. It
is estimated that it takes about two months to transfer 20,000 tons of material [26].
Also detailed in this segment is the fact that all recycled materials are not inherently
recyclable. Calculations show that about 72.5% [27] of recovered construction and
demolition debris can be recycled as building material, the rest having potential as
biofuels or for landscaping and siting purposes.
Policy for Amount Processed per Month” in the C&D waste stream model accounts for the
percentage of recovered material that can be processed per month. This is a function of the kinds
of eco-industrial businesses suitable for reprocessing in the San Francisco Bay Area, and is set at
a control value of 40% for the base earthquake scenario. This assumed value is based off the
City‘s existing high material diverting capability.
Another influential exogenous factor is Percent C&D Recovered (Mandated),” which denotes a
65% required recovery rate for all construction and demolition waste streams as per San
Francisco Environmental Code Chapter 14 and C&D Debris Recovery Ordinance [28]. In order
that the hypothesis maintains any validity, this mandate must remain true throughout the time
frame in question. Without preserving or increasing the 65% diversion rate, no recycled content
processing is possible, disabling the premise of the research question. This leads to an important
point on the relaxing of various ordinances, policies, and norms in a post-disaster setting, which
greatly influences how recovery is managed and the city is rebuilt.
The time frame for simulating both models is 120 months, with the pulse occurring at month 12.
Not included in the models are lookup tables that provide conditional output based on a specified
input. For example, in the Housing Units model, -ookup for Transfer Station Capacity
Increase” outputs a value between 0 and 16 when the -Fraction of Construction and Demolition
waste to Transfer Station/MRF Capacity” reaches a specific value. This output number behaves
as a multiplier captured in -Effect of Capacity on Growth Rate,” which is factored into the
growth rate of the Transfer Station/MRF total capacity, thereby affecting the aforementioned
fraction of debris to capacity. As the growth rate increases, the fraction used decreases, providing
an embedded balancing loop within the model. This loop provides information to the Transfer
Station/MRF Capacity required to process the influx of debris for five transfer stations of
concern. It also allows simulation of recovery times if such capacity is locked to a certain
number of tons if, for example, capacity growth is considered unrealistic [29].
Results and Recovery Forecast
Performing numerous simulations of the base model with policy alternatives described earlier,
graphs and descriptions are provided to quantify the effects of possible scenarios. To reiterate,
complete recovery is described in terms of reaching the pre-earthquake housing state of 330,000
units, compared by the time for such recovery. Also evaluated is the amount of landfilled
material versus recovered material, which will consequently service as building material
following processing. The overall results indicate 6.8 years of recovery following a 7.2
magnitude earthquake, with the benefit of 1.5 million tons of debris being diverted from landfill.
Comparing the extreme cases, a larger percentage of locally supplied recycled material for
construction slows total recovery by two years while saving more than three years of landfill
space [30] and upwards of 1.6 million tons of potential usable debris from being disposed.
Base Case Model Experiencing No Earthquake
Under normal conditions, San Francisco would experience a normal growth rate of housing and
near stable transfer station/MRF capacity (Figure 10). A growth of about 0.89% in housing units
occurs over the 10 year period examined, reaching 332,960 residential units. However, due to the
in-built balancing loop formed, the transfer station capacity is shown to decrease significantly.
Realistically, however, square footage of the material recycling facilities would not be decreased,
but would rather stay constant or increase slightly given reasonable economic and space
circumstances. The equilibrium level for five transfer stations is estimated to be 367,500 tons of
storage and processing capacity per month (Figure 11).
HOUSING STOCK
340,000.
320,000
310,000
300,000.
290,000 —Housing No EQ
its
a
280,000
270,000
260,000
© 1 8 18 28 38 48 S58 68 78 88 98 108
Mont
Figure 10 — Housing Stock in Equilibrium Case
MATERIAL GENERATED & TRANSFER STATION CAPACITY
2 250 .
3 —tandfill
2 200 ——Recovered
ry —T Capacity
5 150 —Constant
0 1 8 18 28 38 48 58 68 78 88 98 108
Months
Figure 11 — Landfill and Recovered Material versus Storage and Processing Capacity
Base Case Model Experiencing Earthquake
With an earthquake pulse resulting in a deficit of 85,000 housing units, a 25% imported material
rate results in a 6.8 year, or 82 month recovery period, as indicated by the blue line on Figure 12.
Varying the imported building material rate to a higher and lower value presents differing
recovery times. As imported material rate is increased, a faster housing refurbishment time is
observed since it is a conventional method of acquiring construction materials. It is assumed that
local processing of material is limited in and near San Francisco, and phase of learning and
implementation by local producers following the earthquake will slow the RCP supply chain,
further escalating the recovery period. Figure 13 shows -Construction Rate” as a flow from the
Housing Units model, indicating the rate of change between housing starts to completed housing
units. As can be noted, an increased import rate intensifies construction rates, as well as further
decreases total recovery time.
HOUSING STOCK
340,000
320,000.
310,000
2 300,000 10% Imported
5
290,000: —25% Imported
75% Imported
280,000
270,000
260,000
250,
o 1 8 18 28 38 48 58 68 78 88 98 108
Months
Figure 12 — Housing Stock with Effects of Earthquake
CONSTRUCTION RATE
2,500:
2,000:
= 1,500:
2 10% Imported
Z —25% Imported
5 1,000. —75% Imported
0 1 6 15 24 33 42 51 60 69 78 87 96 105
Months
Figure 13 — Construction Rate with Effects of Earthquake
Effects on the internal balancing loop for transfer station capacity outputs an increase of 70,000
tons/month of transfer station capacity required to achieve the subsequent results. Therefore at
the point of recovery, the throughput capacity reaches a value of 445,500 tons per month of
processing function in order that a 6.8 year period is realized for the control case, a 21% increase
from the original capacity. It is noticed in the Transfer Station graph (Figure 14) that the amount
of imported versus RCP material used does not affect the total MRF capacity requirements. This
is due to the delay in the processing of RCP to its actual implementation to the construction
stream. In addition, the decision to deconstruct versus demolish affects the landfill, recovered
material and transfer station streams directly. Transfer station capacity value will change based
on which factors are variegated.
TRANSFER STATION CAPACITY
—TS Capacity for
10%, 25%, 75%
200 Imported Material
Tons (Thousands)
0 1 6 415 24 33 42 51 60 69 78 87 96 105
Figure 14 — Transfer Station Capacity with Effects of Earthquake
The usage of recycled content products diverts nearly 1.5 million tons of debris from landfill at
the month of recovery, as shown in Figure 15. Much material still enters the landfill since
bounded processing capacity and delays limit total divertability, totaling about 1.1 million tons
of debris as refuse. The tradeoff for a greater recovery period comes with the benefit of nearly 3
years of landfill space that is conserved with RCP methods [31], saving nearly $6.6 million
dollars in landfill contracting [32][33]. In addition, local markets of recovered content products
will serve to generate income in order to lessen the economic impact on the City after disaster
strikes, while simultaneously providing materials for recovery. An empirical justification of
-building-back-better” is shown in Figure 12. To ensure environmental protection in the
recovery phase alongside the rebuilding of quality housing stock, a compromise in recovery time
must occur to allow necessary time for preparation and planning of reconstruction.
LANDFILLED VS. RECOVERED MATERIAL
1,000; —Landfill Accumulation
800 Recovered Material
Tons (Thousands)
0 1 6 15 24 33 42 51 60 69 78 87 96 105
Months
Figure 15 — Landfilled Material and Recovered Material following Earthquake
Furthermore, the inertia of recovery within the first several months can be attributed to the
learning and implementation of a new means of acquiring construction material via debris
reprocessing. This is so that local suppliers and waste managers can begin producing
construction materials within their respective industry, which is assumed to be non-conventional
in ordinary circumstances. If most material today is being imported into the region, then a shift in
processing RCP material in the Bay Area requires a steepened learning curve before results of
higher RCP content can be witnessed.
Conclusions
Results from the control case described previously show clear incentive for harnessing the
potential of debris as material for new construction. Comparing the extreme cases of 75% import
rate to a 10% import rate shows an increase in recovery of nearly two years (5.6 years versus 7.6
years, respectively). In spite of this, the compromising housing recovery delay is befitted with
the enormous tonnage of debris that is recovered for use, reaching upwards of 1.6 million tons of
material diversion.
An important caveat exists in achieving any material recovery, which relies upon the mandated
ordinance for landfill diversion. San Francisco‘s Ordinance 27-06 requires contractors to recover
at least 65% of materials created on construction and demolition sites. This research envisions
that this and similar directives are kept in place, or optimistically increased, in times of post-
disaster recovery, wherein they may otherwise be relaxed or jettisoned entirely. Without such a
regulation, hopes for landfill diversion are dismal and possibility for material extraction from
debris is difficult.
It is also important to comment on the economic, social and environmental benefits of
reprocessing debris for building construction materials. As is noted by the Community Action
Plan for Seismic Safety [34], economic impacts from a 7.2 magnitude earthquake will result in
direct costs of nearly $14 billion dollars in housing, property, material damage and loss.
Secondary economic hardships would also ensue, resulting in many residents being out of work
until relocation and business restoration is managed. Though difficult to estimate explicit
secondary losses in terms of employment deficits, instating a holistic materials recovery program
will assist in boosting economic conditions and reviving local industry.
Environmental benefits from landfill diversion include reduction in caustic methane emissions
from landfill sites, decrease in space required for landfilling, provision of added building
allotment for commercial or housing needs, and greater utilization of resource value salvaged
from solid waste. Based on the results of this research, calculations show the economic benefits
of reliance on MRF for C&D waste management versus resorting to landfilling as the only
means of debris removal. A 70,000 tons/month MRF capacity growth is suggested for material
recovery; this results in a savings of about $4.4 million dollars as compared to contractual costs
for increasing ears of landfill space [35].
Another important measure of landfilling is the amount of CO. emissions resulting in truck
transport to Altamont landfill in Livermore, CA, a nearly 100 mile round trip over the Bay
Bridge from San Francisco City. More than 73,000 truckloads would be necessitated to move all
debris to landfill if recovery is implemented, about 1000,000 truckloads less than if diversion is
ignored. This diversion not only avoids overall tipping costs of about $120 million (estimated at
$80/ton), but also controls the amount of CO, emitted by a near 63,000 ton reduction in noxious
truck fumes, which is 23 times more emissions than a material recovery scenario [36]. These
drastic increases in cost trickle down to the resident level while emissions affect the aggregate
health of citizens, proving an unsustainable recovery. Though increased practice of material
recovery following a disaster will not entirely remove the described negative impacts, it does
afford greater economic and societal advantage when compared to total landfilling of debris.
Insights
The results provide useful information on prevalent variables considered endogenous to the
system. Some of these aspects should be interpreted with detail, including the transfer stations
and retrofit policies.
Results from the transfer station sub-system modeled within the larger stock-flow structure allow
for interpretation of values to realistic suggestions. Transfer station capacity requires a 21%
increase to stage and process construction and demolition debris only. Since construction and
demolition waste comprises only 12% of California‘s landfill content, the projected capacity
increase will not be sufficient for the other types of debris that will be generated. Therefore,
discussions about regional provisions for staging and processing are critical for material that is to
be recovered, and for those deemed harmful and must be disposed of in other ways.
Additionally, the fact that a relatively large delay (nine months) for transfer station capacity
growth has not severely delayed the overall recovery results of the model is a point of interest.
This means that even with a slow and steady transfer station capacity growth, a great amount of
material can be processed for reuse. The reason for this is that capacity increase is assumed to be
coupled with increased labor and machinery for processing. Adding physical capacity alone will
not progress the movement and recycling of debris; rather, additional variables of labor and
machinery are influential in waste management.
A retrofit policy analysis also attests to the large benefits from retrofitting housing prior to a
hazard event (Appendix B). Retrofitting has been a foregoing and prominent means of
community resiliency, but is faced with financial and societal complexities which hinder its
widespread adoption. Testing the feasibility of retrofitting within the system bounds again prove
its viability and necessity in a community that is prone to adverse environmental threats.
Retrofitting is shown to cut overall recovery time by nearly two years, which allows potential for
greater life safety prior to a disaster event, as well as maintenance of citizens following an
earthquake, both indicative of a resilient city.
Model Critique
As with any System Dynamic model, the decision of bounding the model must always be
questioned and pushed such that feasible insights are not excluded. In this case, the model
boundary treats as exogenous the aspects of stakeholders, public decisions and societal concerns,
land use and design and supply chain understandings of recycled content materials. For example,
the decision of housing is left uncomplicated in this model, but can have large implications in a
future recovery scenario. Also, notions of environmental equity are also left exclusive to the
model; questions of which communities will suffer from new landfills that must be formed in or
near their locales if debris is not diverted must be considered when forming disaster management
plans. If intentions to recover material from damaged housing exist, then notions of designing for
deconstruction should be studied and perhaps implemented in new construction of housing. This
method of design includes parameters of end-life housing removal or destruction, and entails
guidelines on how to best recover and recycle material from a home that is no longer of service.
An additional iteration would consider these and other variables as inclusive of the model
bounds, as they have clear influence for the overall built environment and city recovery scenarios
as defined by the scope of research.
Additionally, some results show sharp or precipitous growth/decline rates that may not be fully
be realistic. For example, housing construction shows a sharp decline after all homes are
refurbished. In the real world, contractors would slow down momentum as they anticipate
housing reconstruction nearing its end, and perhaps send labor force to other tasks. In this case,
and again for the sake of simplicity, the results are shown to be abridged where they can be
contrasted to other variables to understand relatedness. This method of resolution allows for
comparison and linkage to other aspects of the model since relative results per variable are the
same.
Tradeoff for such exclusions and simplifications are the levels of clarity and focus that the model
can bring for early comprehension of the variables of interest and the hypothesis in question.
This allows for broader discussions about influential aspects of post-disaster recovery within the
regional community. The intention for such model simplification is to reach multiple audiences
that are able to add more foreseeable variables and contexts to the impending issue of post-
disaster recovery. Thus far, the model adequately provides the results and processes needed for
the discussion the author set to simulate. It also speaks to the need of additional research on a
broader level, with components that must be linked to the existing model to understand the vast
interconnections of variables affecting hazard mitigation and disaster management. The hope is
that future iterations constitute vantages beyond debris handling and housing stock
refurbishment.
Appendix A — Material Recyclability
Percent Recyclable
Material (%)
Remainder/
Ci ite C&D
Rock soil and fines
Table 2 — Simplified Recyclability Rates per Material
Note: Materials with recyclability values are used for this research model. For an expanded list of
material recyclability, see Table 3.
Material
Second-life Uses
Timber/ Wood/ Lumber Waste}
-Wood waste generated during site work can be ground up and recycled with greenwaste
-Wood from the demolition process requires more labor-intesive disassembly of materials to
remove fasteners and finishes and should be screened for lead paint
-Recycled wood can be ground into wood chips or wood flour and used to make composite
or engineered lumber products, mulch or composted
-Unseparated waste wood is sometiems burned to produce waste electricity
Clean wood waste can be more easily used as feedstock for engineered lumber
-Lumber and other wood products can be directly reused or gound and used for boiler fuel,
mulch and engineered lumber, Care should be taken to separate leadbased paint coated
wood and chemically treated lumber
-Large timbers and dimensional lumber removed from demolition operations can be reused
or recut for consruction projects. However, in many cases, the lumber will need to be
regraded by a certified grader if it is used for anything other than ornamental purposes
-Composite wood based thermoplastic products
Brick
95%
~ Brick has a salvage value of $400 per ton, clean and stacked on a pallet.
The process of cleaning mortar from brick, however, can be labor intensive, removing much
of the profit from this process.
-Brick remains, however, a very recyclable CD material that recyclers will often accept at no
cost, Non-salvagebable brick can be crushed and used as aggregate base or backfill material
-Bricks can be recycled through a crushing process, creating “brick chips.” Those brick chips
can be used as a landscape material, or can be reground through the manufacturing process
to create new, quality brick,
Asphalt Paving
75%
-Asphalt is often ground up and used as road-base under new roadways or parking lots. On
larger projects this recycling of asphalt can be accomplished on-site utilizing mobile
grinding equipment. This can yield substantial savings by eliminating transportation costs
and tipping fees while providing raw materials and road-base that would have needed to be
purchased.
Asphalt Roofing
75%
-Recycling of Asphalt composition roofing results in aggregate base, asphalt pavement and
pavement cold patch
“Asphalt shingles can be reycled into new asphalt pavement mixes. They can also serve two
purposes at a cement kiln: conbustion of the shingles provided energy in the kiln and the
remaining mineral components containing the limestone granules, serve as raw material for
cement
Concrete
80%
-Crushed, screened and used as road base. Aggregates can be recovered from this process
and used in the production of new cnocrete, in regions where aggregates are not readily
available
-Recycled concrete must be used with caution due to strength parameters
-preferably for foundation and site work
Gypsum
-Gypsum dry wall can be reycled into new drywall, cement and agricultural uses.
-Drywall gypsum can be reycled back into new drywall if most of the paper is removed.
The paper limits the amount of recycled gypsum allowed in new drywall, because the
paper content affects its fire rating
-Potential markets for drywall waste: cement plants use large quantities of virgin
gypsum to clinker, stucco additive, ; drywall wastes from demolition waste can be
reyclced for nonagricultural markets;
Steel
85%
“Steel C&D is very recyclable due to is lack of contaimnation by dissimilar materials.
-85% of CRD steel is currently recycled by recyclers
-Good markets exist for ferrous metals such as iron and steel, as well as other
non-ferrous metals such as copper, brass and aluminum.
-Metal is almost always recycled back into other metal products and recycling
opporunities are avialable in virtually every area around the country
Treated Wood
25&
Treated wood should be handled separately from vegetative debris being
recycled
-Besides wooden utility poles, other lumber that may be chemically treated
includes decks, fences landscaping materials, wood bridges and railroad ties.
Treated wood contains chemical perservatives that can contaminate recycled
wood produccts; these woods can be combused in waste to energy facilities.
Table 3 — Material Recyclability Methods
Appendix B - Retrofit Policy
This particular policy scenario differs from the four described previously in that it applies a pre-
earthquake, mitigative effort to retrofit housing prior to assessing effects from the earthquake
shock. The causal loop diagram in Figure 16 demonstrates how retrofitting homes becomes
dually beneficial if enough homes are retrofitted to at least a minimum standard [37]. This
retrofit scheme is a minimal approach intended to reduce harm to those who live or frequent the
building. Collapse would be prevented, and occupants should be able to escape the building
safely, but the building might not be repairable or fit for occupancy following the earthquake
event [38]. This is determined as the least costly method of retrofit, at an average of $6.60/ sq.
ft., adding to approximately $11,000 [39] per housing unit [40]. The retrofitting would result in a
57% reduction of damaged housing; a drop to 49,000 damaged or collapsed residential units
versus 85,000 units.
The diagram shows that as retrofit policy is implemented, uninhabitable units are consequently
decreased following an earthquake. If more homes are rendered habitable, either green-tagged or
yellow-tagged, more shelter-in-place is possible. Shelter-in-place is described as a -resident‘s
ability to remain in his or her home while it is being repaired after an earthquake — not just for
hours or days after an event, but for the months it may take to get back to normal. For a building
to have shelter-in-place capacity, it must be strong enough to withstand a major earthquake
without substantial structural damage. This is a different standard than that employed by the
current building code, which promises only that a building meets Life Safety standards (i.e., the
building will not collapse but may be so damaged as to be unusable)” [41].
The San Francisco Planning and Urban Research Association (SPUR) estimates that only 75% of
the City‘s current housing stock will provide adequate shelter for residents after a large
earthquake, slowing overall recovery. SPUR‘s projected goal for resilience is that the housing
stock reaches a 95% shelter-in-place standard [42]. This goal is augmented by substantial
retrofitting, which helps to retain the San Franciscan in the city after an earthquake. A resilient
city can facilitate recovery and increase housing construction to restore uninhabitable residences
to livable standards so as to regain any displaced residents. This is shown by the right-hand
reinforcing loop in Figure 16.
Retrofit Policy
ee > shelter-in-Place
a <v
Destructioi Sy \
f '
Uninhabitable
/ Housing =¥
/ - 4. Displacement
| R
| (@} | Ae
Debris Impediments \ Shelter-in-Place
\
Reconstruction *
Debris Clearance Sig Recovery
—— _-
Figure 16 — Causal Loop Diagram, Retrofitting Policy
Appendix C — Emissions
The following calculations estimate the number of round trips trucks would need to make from
landfill, and the total emission that would be released from these trips. It is assumed that trucks start
in San Francisco and return back to San Francisco. Note that these are very simplified results that may
exclude other variables that also generate emissions.
Altamont Landfill is about 60 miles one way. To calculate emissions, reduce 10 miles for transfer that
is near San Francisco. Therefore, the round trip for one truck 100 miles (5Omi x 2)
CNG like LNG has 90% less air pollution than diesel gas, and is what Recology trucks are using in San
Francisco.
Diesel produces 22.3 Ibs of CO2/gallon (EPA)
Calculate the CO2 emissions for diesel, then reduce per CNG ratings:
The average garbage truck travels about 25,000 annually, gets about 3 miles/gallon, and uses 8,600
gallons of fuel.
For 100 mile round trip to/from Altamont Landfill in Livermore
100 miles/3 miles/gallon = 34 gallons per truck
34 gallons per truck X 22.3 lbs CO2/gallon = 758.2 Ibs CO,/truck
Using the base case scenario for RCP 25%, we get about 1.096M tons of landfill:
From information from Richard Valle, each truck can carry 10-15 tons. Using the higher range:
1.096M tons LANDFILLED / 15 tons/truck = 73,066 trucks
73,066 trucks x 758.2 Ibs CO2/truck = 55,399,147 Ibs CO2 = 27,700 tons CO2 for diesel fuel
Reduce by 90% for CNG estimate:
2770 tons CO, WITH RECOVERY
Account for recovered material added to landfill material:
1.096M+1.5M = 2.596M
2.596M tons / 15 tons/truck = 173066.67 trucks
17,3067 trucks x 758.2 Ibs CO,/truck = 131219399.4 Ibs = 65609.6997 tons
Reduce by 90% = 65,610 tons CO, if NO RECOVERY
65,610-2770 CO, = 62,839 tons CO, SAVED if recovery implemented over 6.8 years
Source: www.informinc.org/fact_get.php
AS5- Hauling and CO, Emissions
References
[1] Association of Bay Area Governments. 2011. Shaken Awake. ABAG: 1-2.
[2] Phillips, Brenda D. Disaster Recovery. 2009. Boca Raton, London, New York: CRC Press: 35-37
[3] United States Geological Survey. 2008. Bay Area Earthquake Probabilities. Retrieved May, 10, 2012, from
http://earthquake.usgs. ucerf/
City and County of San Francisco, Public Works and Engineering Annex. 2010. Appendix B: Disaster
Debris Management Plan. City and County of San Francisco: 9-12.
[5] Applied Technology Council (ATC). 2010. Here Today- Here Tomorrow: The Road to Earthquake
Resilience in San Francisco: Report 52-1 and 52-1A.
[6] Ibid, 2010.
[7] Ibid, 2010.
[8] Ibid, 2010.
[9] Phillips, Brenda D. 2009. Disaster Recovery. Boca Raton, London, New York: CRC Press.
[10] Solis, Gabriela Y., H.C. Hightower, J. Sussex, and J. Kawaguchi. 1996. Disaster Debris Management. The
Disaster Preparedness Resources Center, The University of British Columbia.
[11] See Appendix A
[12] Tukker and Gielen, 1994, as cited by Woolley, G.R., Goumans, J.J.J.M and P.J. Wainwright; Waste
Materials in Construction, Science and EDgmrens of Recycling for Environmental Protection, 2000.
[13] Addis, William. 2006. Building with Reclai r and Materials: A Design Handbook for Reuse
and Recycling. 9 London Sterling, VA: E ailscan,
[14] See Appendix A
[15] Addis, William. 2006. Building with Reclaimed Components and Materials: A Design Handbook for
Reuse and Recycling. 9 London Sterling, VA: Earthscan.
[16] Collins and Sherwood, 1995, Collins et al., 1998; BRE, 1998 as cited by Addis
[17] Lauritzen, E.K. 1998. Emergency Construction Waste Management. Safety Science 30(1-2): 45-53.
[18] Applied Technology Council (ATC). 2010. Here Today- Here Tomorrow: The Road to Earthquake
Resilience in San Francisco: Report 52-1 and 52-1A.
[19] Phillips, Brenda D. 2009. Disaster Recovery. Boca Raton, London, New York: CRC Press. Image adapted
from Figure 21.
[20] Ramezankhani, Atefe and Najafi Yazdi, Mostafa. 2008. A System Dynamics Approach on Post-Disaster
Management: A Case Study of Bam Earthquake. International Conference of the System Dynamics Society.
Athens, Greece: 1-34
[21] Ho, Yufeng, Chienhao Lu and Hsiao-Lin Wang. 2006. Dynamic model for earthquake disaster prevention
system: A case study of Taichung City, Taiwan. Thesis, Taichung, Taiwan: Graduate School of Architecture
and Urban Design.
[22] Quinn, David. 2008. Modeling the Resource Consumption of Housing in New Orleans using System
Dynamics. Cambridge: M | Institute of Technol
[23] 1600 tons per day, 12% of which is C&D Debris, CalReycle, 1997
[24] Association of Bay Area Governments. 2011. Shaken Awake. ABAG: 1-2.
[25] Conversation with Richard Valle, CEO Tri-Ced Recycling, Union City, CA, 2011.
[26] See Appendix A
[27] City and County of San Francisco. 2006. San Francisco Environmental Code Ch 14.
[28] See Appendix A
[29] Compared to 5 years of landfill space used if no recovery is instated.
[30] Compared to 5 years of landfill space used if no recovery is instated.
[31] San Francisco Chronicle. 2010. Altamont Landfill. Retreived Jan. 4 2012 from http://www.sfgate.com/cgi-
bin/article.cgi? f=/c/a/2010/08/06/MNLNIELPE1. DTL.
[32] Assuming 500,000 tons per year landfill capacity
[33] Applied Technology Council (ATC). 2010. Here Today- Here Tomorrow: The Road to Earthquake
Resilience in San Francisco: Report 52-1 and $2-1A
[34] MRF cost comparison based on construction costs only; variable costs have not been included
[35] See Appendix C
[36] See Appendix B
[37] Applied Technical Council. 2009. Here Today—Here Tomorrow: The Road to Earthquake Resilience
Earthquake Safety for Soft-Story Buildings 52-3. Redwood City, ATC.
[4
[38] Ibid,24-25
[39] Ibid, 28
[40] San Francisco Planning and Urban Research Association. 2012. Safe Enough to Stay. San Francisco,
SPUR: 1-5.
[41] Ibid, 2.
