CREATE Research Archive

Published Articles & Papers

2009

Understanding the Implications of CriticalInfrastructure Interdependencies for WaterRae ZimmermanNew York University, rae.zimmerman@nyu.edu

Follow this and additional works at: http://research.create.usc.edu/published_papers

This Article is brought to you for free and open access by CREATE Research Archive. It has been accepted for inclusion in Published Articles & Papersby an authorized administrator of CREATE Research Archive. For more information, please contact gribben@usc.edu.

Recommended CitationZimmerman, Rae, "Understanding the Implications of Critical Infrastructure Interdependencies for Water" (2009). Published Articles& Papers. Paper 7.http://research.create.usc.edu/published_papers/7

UNDERSTANDING THE IMPLICATIONSOF CRITICAL INFRASTRUCTUREINTERDEPENDENCIES FOR WATER

Rae ZimmermanInstitute for Civil Infrastructure Systems (ICIS), New York University, Wagner Graduate School ofPublic Service, New York

Abstract: Direct terror attacks on water infrastructure will have implications not onlyfor water or wastewater systems but also for other interdependent systems (e.g. fire pro-tection and other emergency services, manufacturing, and food production). At the sametime, attacks on other infrastructures (e.g. power production, chemicals production, trans-portation, and communications) may have an impact that would impair or debilitate theoperation of water and wastewater systems. Although broad system interdependenciesare covered earlier in the Handbook, this article will discuss the water-specific interde-pendencies and what infrastructure managers can do to understand, quantify, and henceplan for the implications of infrastructure failures beyond their direct control.

Keywords: water; interdependencies; critical infrastructure; natural hazards; security;energy; telecommunications; transportation

Water systems are dependent on and interdependent with many other infrastructures.This is an outcome of functional necessities, spatial proximity to other infrastructures,and economies of scale that have arisen over time. These relationships are growing withthe size of the population, generally increased demand for water resources [[1], p. 10]particularly for public supplies [[2], p. 39], population distribution that has promoted thetransmission of water over long distances, the geographic concentration of water-relatedinfrastructure components, and changes in technology for water control and delivery

Wiley Handbook of Science and Technology for Homeland Security, Edited by John G. VoellerCopyright © 2009 John Wiley & Sons, Inc.

1

2 CRITICAL INFRASTRUCTURE INTERDEPENDENCIES FOR WATER

systems. This article begins by introducing the concept of dependence and interdepen-dence, characteristics of water systems (covering water supply and wastewater treatment)essential to understanding the nature and impact of these relationships, and the relevanceof this area of inquiry for security policy, including the allocation of resources for riskmanagement and needs of emergency response. Finally, existing research organized bythe major infrastructure sectors to which water is interrelated, how interdependencies canbe measured, and recommendations for future research directions are discussed.

Dependence and interdependence as they pertain to infrastructure are usually con-sidered distinct concepts. Rinaldi et al. [[3], p. 14] define dependency as a relationshipbetween two infrastructures in a single direction, that is, one infrastructure influencesthe state of another, whereas interdependency is bidirectional (and implicitly multidirec-tional) with two (and implicitly more) infrastructures influencing each other. Spatial andfunctional concentration is a key element associated with interdependence.

Although interdependencies are often beneficial, they may also be disadvantageous ifthey potentially increase the vulnerability of water systems and the systems that dependon water to threats posed by natural hazards and terrorism. Disruptions in systems uponwhich water is dependent, whether from natural hazards, terrorism, or accidents, necessar-ily magnify the effects on water systems. Security strategies now emphasize an all-hazardsapproach encompassing natural hazards, terrorism, and other intentional attacks given thatoutcomes or consequences of these different events are often similar. Natural hazardsthat often drive infrastructure disruptions have been increasing, with the annual rise infederally declared major US disasters estimated at 2.7% per year from 1953 to 2005 [[4],p. 382]. Similarly, terrorist attacks disrupting the interdependent infrastructure can mag-nify the consequences. Although terrorist attacks directly on water infrastructure (asdistinct from vandalism or acts of sabotage) are rare in the United States, the threat forwater is real enough to prompt the US government to include it in the list of criticalinfrastructures slated for protection under federal homeland security programs. In theUnited States, water systems have been compromised in a manner analogous to a terror-ist attack, and internationally, terrorist attacks on water have been quite prevalent [[5],p. 528].

Interdependencies between the water sector and many other kinds of infrastructureand especially those that comprise emergency services have been identified as a crit-ical element of federal security policy, including resource allocation for risk manage-ment. Interdependencies are a centerpiece of the National Infrastructure Protection Plan(NIPP), and are a component of assessing risk to the water sector. For energy and waterinterdependencies alone, the House and Senate Subcommittees on Energy and WaterDevelopment Appropriations requested “a report on energy and water interdependen-cies, focusing on threats to national energy production that might result from limitedwater supplies” [[6], p. 9]. The US Department of Homeland Security (DHS) issuedsector-specific plans (SSPs) to implement the NIPP with input from other agencies. Thewater sector is one of 17 critical infrastructure sectors to which the plan applies. Thewater SSP is the longest of the SSPs issued by the US DHS [7]. The water SSP definesdependency and interdependencies in the following way: “Reliance on another asset orsector for the functioning of certain assets is called a dependency; if two assets dependon one another, they are called interdependent” [[7], p. 50]. Interdependencies betweenthe communications sector and the water sector are included as an important element inthe Communications SSP [8]. The water sector is mentioned in and regulated by over adozen security and environmental laws combined and addressed in a half dozen federaldirectives and executive orders [7].

WATER SYSTEM COMPONENTS AND INTERDEPENDENCY 3

Public supply11%

Thermoelectric48%

Irrigation34%

Other*4%

Industrial5%

*Other is an aggregate total ofdomestic, livestock, aquaculture,and mining, each less than 1%.

FIGURE 1 Total water use by type of user, US, 2000. Note: This reflects total water use, and doesnot take into account return flow. (Source: diagrammed from US Geological Survey Estimated Useof Water in the United States in 2000, Figure 1, May 2004. http://pubs.usgs.gov/circ/2004/circ1268/htdocs/figure01.html.)

1 WATER SYSTEM COMPONENTS AND INTERDEPENDENCY

Water usage patterns provide a context for understanding water infrastructure and itsrelationship with other sectors. Figures 1 and 2 show the distribution of water use fortotal water and freshwater, respectively, in the United States (not including return flows,i.e. water consumption).

The type, extent, and impact of dependencies and interdependencies associated withwater and other infrastructure vary depending on the water component and the type oftechnology used for each. Technologies for the provision of water and size of facilities arelikely to dramatically alter the way in which other infrastructures are used to provide ser-vices for water infrastructure; for example, the Electric Power Research Institute (EPRI)[[9], p. 3–5] estimates that unit electricity consumption for surface water treatment andwastewater treatment declines with the size of plant and for a given plant size, the varia-tion in energy consumption for wastewater treatment can vary depending on the type oftechnology by one and a half to three times. The water-supply sector consists of a verycomplex system of interconnected resources, facilities, and services. Water sources existboth above and below ground. Large transmission systems, called aqueducts , primarilybring surface water supplies to the points of consumption connecting to the holding orstorage reservoirs and extensive distribution lines. O’Rourke [[10], p. 23] identifies waterdistribution lines as a type of “lifeline system” interdependent with other infrastructurelifelines noting that during the 2001 World Trade Center (WTC) catastrophe, water linebreakages affected other infrastructure lifelines, flooding transit arteries and fiber-opticlines [[10], p. 24; [11]]. Underground water resources, accounting for about one-third ofthe US public water supply [2], serve both large systems and individual users relying on

4 CRITICAL INFRASTRUCTURE INTERDEPENDENCIES FOR WATER

Public supply13%

Thermoelectric39% Irrigation

40%

Other*4%

Industrial5%

*Other is an aggregate total ofdomestic, livestock, aquaculture,and mining, each less than 1%.

FIGURE 2 Freshwater use by type of user, US, 2000. Note: This reflects total fresh water use, anddoes not take into account return flow. (Source: diagrammed from US Geological Survey, EstimatedUse of Water in the United States in 2000, Table 2, May 2004. http://pubs.usgs.gov/circ/2004/circ1268/htdocs/table02.html.)

wells usually associated with electricity-driven pumps for supply. Water storage repre-sents another set of infrastructure facilities that interface with and are usually connectedwith the transmission and distribution systems.

Attributes of a couple of the key components—water and wastewater treatment plantsand dams—are highlighted here because of their special significance for interdependen-cies and their consequences.

1.1 Water and Wastewater Treatment Plants

Water-supply plants are extensively distributed or localized throughout the US watersupplies or community water supplies, defined under the Safe Drinking Water Act asserving 25 persons or more or having 15 service connections, as of 2004 numberedapproximately 161,201 [[7], p. 16] and serve 84% of the US population [[7], p. 1]. Inspite of the extensive geographic coverage of community water-supply facilities in theUnited States, they are concentrated, reflecting the fact that 45% of the US populationis served by only 6.8% of the water-supply facilities [[5], p. 531]. Wastewater utilities,regulated under the Clean Water Act, are far more concentrated than water systems, giventheir generally larger size and urban orientation, and the number of wastewater facilitiesis about one-tenth the number of water supplies. There are 16,255 regulated publiclyowned treatment works [[7], p. 19], serving 75% of the US population [[7], p. 1]. Therelatively greater degree of concentration of wastewater facilities is not accounted forby the lower percentage of people served, and has to do with economies of scale intreatment technology. The degree of concentration is even greater in both the sectorswhen one considers that relatively few of these utilities serve the bulk of the population.

WATER SYSTEM COMPONENTS AND INTERDEPENDENCY 5

These characteristics do not take into account private bottled water providers, organizedand regulated differently, and is beyond the scope of this article.

1.2 Dams

Dams are another area where interdependencies can occur, since the provision of water(excluding individual water systems) usually begins with the use of dams, and the opera-tion and control of dams depends on many other infrastructures. The National Inventoryof Dams (NID) records close to 80,000 dams in the United States. The spatial distribu-tion of dams is potentially significant for interdependencies and the vulnerabilities theymay pose. At the state level, the number of dams and to a greater extent capacity (totalmaximum capacity) is highly concentrated: about half of all dams is located in onlyeight states, and about half of the total maximum dam capacity is located in only fivestates. The results of an initial analysis of the distribution of the number of dams andtotal maximum dam capacity as distributed by state in the United States are shown inFigures 3 and 4 [12]. People’s dependency on storage of water by dams can in a grossway be portrayed in terms of where dams are located relative to population. The locationof dams relative to population and population density is portrayed in Figure 5, indicatinga modest relationship with a low, even though not significant, correlation. Numerousactivities depend on the water supply that dams provide, reflecting the purpose thatthese dams serve (Fig. 6). Many dams serve multiple purposes. Analysis of data on thenumber of dams by primary purpose from the NID indicates that the following activ-ities are dependent upon dams: recreation (33.4%), flood protection including stormwater management (15.5%), and fire protection including stock and small pond farms(13.6%). In addition, 9.3% of the dams are used for water supply (as the primarypurpose).

Number of dams

Value below median (1006 dams)

1007–1312 (IN, KY, AR, IL, WY)

1313–1651 (PA, CA, VA, MA, CO)

1652–2339 (OH, NY, NE, AL, SC)

2340–3302 (SD, IA, MT, NC, MS)

3303–6951 (MO, GA, OK, KS, TX)

FIGURE 3 Number of dams by state in the United States, 2006. (Source: mapped from TheStanford National Program on Dam Performance Database as of 2007 By Sara A. Clark, GraduateResearch Assistant, NYU-Wagner, Institute for Civil Infrastructure Systems.)

6 CRITICAL INFRASTRUCTURE INTERDEPENDENCIES FOR WATER

Population density(persons per square mile)

1–33

67–138

34–66

139–277278–9317

Total dam capacity15.7–18.3 mil. acre-ft. (ME, OR, WY, ID, KS)

18.3–24.2 mil. acre-ft. (TN, GA, MO, MS, NY)

24.2–30.4 mil. acre-ft. (AR, KY, SC, ND, WA)

30.4–42.1 mil. acre-ft. (MN, NV, SD, OK, AZ)

42.1 mil. to 9.7 bil. acre-ft (MT, CA, TX, FL, MI)

FIGURE 4 Total dam capacity and state population density in the United States, 2006. (Source:mapped from The Stanford National Program on Dam Performance Database as of 2007; 2000US Census; US UASI, DHS, 2006 By Sara A. Clark, Graduate Research Assistant, NYU-Wagner,Institute for Civil Infrastructure Systems.)

2 TYPES OF INTERDEPENDENCY

Conceptual literature in the infrastructure interdependency area emphasizes functional andgeographic interdependencies as major types of interdependency, though other typologieshave expanded or refined the number of categories [3, 13].

2.1 Geographic Interdependencies: Co-location

Physical interconnections often called utility bundling or utilidors [14] are enhanced inutility distribution systems by economies of co-location. Transportation is one infras-tructure that has important physical linkages to water distribution systems. The watersupply for the city of Paris, France, uses bridges to link water from the Left Bank tothe Right Bank. A town in New Jersey shares wastewater treatment services with atown in Pennsylvania which involves transporting wastewater across a bridge. During adrought period, New York City constructed a temporary water-supply line, which traveledacross the George Washington Bridge to supply water to New Jersey if required. Spatiallinkages between water distribution systems and electric power and telecommunicationlines are also common. Although these interdependencies provide many advantages andinnovations, the proximity of water distribution lines to other infrastructures potentiallymagnifies vulnerabilities to disruption.

Large cities routinely experience water distribution disruptions, and causes vary. InNew York City, environmental factors are a major factor contributing to the average

TYPES OF INTERDEPENDENCY 7

800

600

1000

1200

080007000600050004000

Number of dams per state

Pop

ulat

ion

dens

ity (

peop

le p

er s

quar

e m

ile,

excl

udin

g D

C, U

S C

ensu

s 20

00)

3000200010000

200

400

30,000,000

20,000,000

25,000,000

35,000,000

40,000,000

080007000600050004000

Number of dams per state

Tot

al s

tate

pop

ulat

ion

(US

Cen

sus

2000

)

3000200010000

5,000,000

10,000,000

15,000,000

FIGURE 5 Relationship of number of dams to state population and state population density inthe United States, 2006. (Source: graphed from the national inventory of dams as of fall 2006by Sara A. Clark, Graduate Research Assistant, NYU-Wagner, Institute for Civil InfrastructureSystems.)

of 500–600 water main breakages annually. Water main breakages can disrupt otherinfrastructure and vice versa. Ways that water disruptions affect other infrastructuresinclude undermining or washing out of street surfaces by water releases, shorting outof electrical lines, and undermining support of gas lines. Ways that water disruptionsare caused by other infrastructures include proximity to roads and transit systems [15],vibration, weakening of lines from the undermining of soil support due to construction,being hit by construction equipment, and electrical conductance created by proximityto electrical lines and voltages from trains. An analysis of about 100 cases of multipleinfrastructure failures involving water main and other infrastructure breakages foundthat water main breakages are more commonly initiators of outages in other nearbyinfrastructures, such as gas main breaks and roadway washouts, than vice versa, butthese findings are sensitive to the types of cases in the database [16].

8 CRITICAL INFRASTRUCTURE INTERDEPENDENCIES FOR WATER

20(%)

5(%)

10(%)

15(%)

25(%)

30(%)

35(%)

40(%)

Recreation

Flooding

FireIrrigation

Water Supply

OtherNone

Hydroelectric

Wildlife

Tailings

“No Value”

Debris

Navigation

Wastewater

FIGURE 6 Distribution of dams by purpose in the United States, 2006. (Source: graphed fromThe Stanford National Program on Dam Performance Database as of 2007.)

2.2 Functional Interdependencies by Infrastructure Sector

2.2.1 The Energy Sector: Water and Energy Interdependencies Water for energyproduction . As shown in Figures 1 and 2, 48% of total water usage and 39% of freshwaterwithdrawals in the United States in 2000 were accounted for by thermoelectric power[[2], p. 35], though when consumption is considered, most of that water is returned andthermoelectric power production in 1995 accounted for 3.3% of consumption [6]. TheU.S. Department of Energy (DOE) notes that “of the 132 billion gallons per day offreshwater withdrawn for thermoelectric power plants in 1995, all but about 3.3 billiongallons per day (3%) was returned to the source. While this water was returned at a highertemperature and with other changes in water quality, it was available for further use”[[6], p. 17]. Thermoelectric generating plants using open-loop cooling in turn produce31% of US energy generation [[6], p. 18].

Energy for water production and wastewater treatment . The dependency of water onelectric power has been underscored by a number of very large power outages that threat-ened water services or actually did bring water production and wastewater treatment to ahalt. Electric power outages in California in 2001 nearly stopped major water pumps [[3],p. 11; [17]]. The August 2003 US and Canada electric power outage stopped wastewaterpumps in New York City resulting in untreated water discharges to New York water-ways. The same outage disrupted water-supply plants in major cities such as Clevelandand Detroit, and it took those two cities over two times as long to restore their watersystems as relative to the amount of time it took to restore electricity (see Section 3).Electric power outages have been increasing at the rate of 7.2% in the United Statesbetween 1990 and 2004 [18]. Weather-related events have been dominating other con-ditions as causes of electric power outages, contributing to an increasing overall outageduration rate of 14% between 1990 and 2004 in the United States and higher rates sincethe late 1990s [19]. Both these trends are likely to affect water systems. Although waterproduction and movement (for both treatment and supply) account for a small portion ofenergy produced in the United States, estimated at 4% across all functions [[6], p. 25; [9],pp. 1–2], from the perspective of the individual water company, energy figures promi-nently in water production, accounting for an estimated 80% of the costs for processing

TYPES OF INTERDEPENDENCY 9

Energy inputs

Onsite power generation2.6 MW

Western area power authority4.16 MW

Pacific gas and electric

Oxygenerationplant27%

Activatedsludge mixing

22%

Headworks

18%

Lighting,losses, misc.

12%

Solidshandling

10%

Activatedsludge pumping

7%

Othermotor loads

5%

Energy outputs

EBMUD

FIGURE 7 Example of energy use in a water supply and wastewater treatment plant: East BayMunicipal Utility District, California, 2004. (Source: diagrammed from information in Hake, J.M., Gray, D. M. D., and KaIlal, S. (2004). Power Diet. California’s energy crisis prompts onetreatment plant to reevaluate its power intake. Water Environ. Technol ., May, 37–40.)

and distribution of municipal water supplies [[9], pp. 1–2]. The estimated electricityconsumption for fresh water supply in 2000 provided by public water-supply agencieswas 30.6 billion kWh per year, and this was estimated to increase up to 45.7 billionkWh per year by 2050, an increase of about 50%, largely driven by population growth[[9], p. A-3]. In the water production process, most of the energy is used for pumpingand treatment. For example, the East Bay Municipal Utilities District, a water companythat provides both water supply and wastewater treatment, uses half of its energy forpumping and treatment (Fig. 7). Its use of energy to acquire raw water is lowered by thefact that it obtains its water resources via gravity.

The transportation of water itself is dependent upon energy in most cases unlesstransport occurs via gravity. Although no comprehensive information exists on changesin the acquisition of water, it is often cited that water is being accessed from longer andlonger distances to meet water needs, especially for urban areas, which will inevitablyinvolve increases in the use of electricity.

2.2.2 The Transportation Sector: Water, the Chemical Industry, and TransportationThe reliance of the water industry on transportation for the provision of chemicals to treatwater is a potential vulnerability point, and has contributed to changes in the choice ofchemicals. The viability of providing water services by centralized water utilities to denseurban areas is dependent on quality controls, which, in turn, is dependent on chemicals,in particular, chlorine gas for disinfection. Potential attacks on chlorine storage tanks andtransport vehicles and accidents involving the transport of chlorine by truck or rail haveunderscored this as a distinct vulnerability. The water industry performs conversions fromchlorine gas to the less vulnerable sodium hypochlorite and ultraviolet disinfection, andthis conversion has been estimated to be $647,000 to $13.1 million per plant at abouttwo dozen selected larger plants [20]. An analysis of the conversion cost data reveals amoderate but positive correlation between the cost of conversion and plant size as shownin Figure 8 [12].

10 CRITICAL INFRASTRUCTURE INTERDEPENDENCIES FOR WATER

14,000,000

12,000,000

10,000,000

8,000,000

6,000,000

4,000,000

2,000,000

0

Dis

infe

ctio

n co

nver

sion

cos

t ($)

0 200 350300250150

Wasterwater facility size (mgd)

10050

FIGURE 8 Relationship between wastewater disinfection cost and wastewater facility size forsample facilities, 2007. (Source: analyzed from: US GAO (2007). Securing Wastewater Facilities:Costs of Vulnerability Assessments, Risk Management Plans, and Alternative Disinfection MethodsVary Widely . Report to the Chairman, Committee on Environment and Public Works, US Senate,GAO-07-480, March, Table 1, p. 14.)

2.2.3 Water, Communications, and Information Technology In the water sector,communication and information technologies increasingly control water quality, distri-bution, and customer interfaces. Information technologies are not only linked to watersystems but also provide connections between water and other systems. The dependencyof water on communications and information technology is identified in the Communi-cations SSP [[8], p. 41]. The effect of disruptions of computerized control systems, suchas supervisory control and data acquisition (SCADA), on water systems is noteworthy.For example, in 2001, a hacker disabled the SCADA system operating the wastewatertreatment system, Queensland, Australia, causing extensive discharge of sewage [[21],p. 9].

Information technology, particularly as used in water applications, has been revolution-ized by nanotechnology enabling detection of water chemicals to achieve extraordinarysensitivity. Wireless communication technologies further revolutionized water measure-ment. In the mid twentieth century, water-supply and wastewater quality standards werelargely based on qualitative measures of chemical and biological material, for example,appearance, and by the late twentieth century quantitative standards gradually emerged,for example, expressed in parts per thousand and parts per million. In the twenty-firstcentury those measures often went into the parts per trillion levels. These increasinglymore stringent standards were made possible by newer detection technologies [[22], p.80]. As a result of the increase in the quantification of and limits of detection for waterquality measures, the water industry has become more dependent on information tech-nologies that are usually very specialized [22, 23]. In 2006, American Society of CivilEngineers (ASCE) and American Water Works Association (AWWA) draft guidelinesfor water utility security outlined an extensive set of criteria for sensor-based detection,which reflect the greater use of, and hence dependency upon information technologiesfor water infrastructure [[24], Section 9].

RESEARCH DIRECTIONS 11

3 MEASURING FUNCTIONAL INTERDEPENDENCY

When attention was first drawn to the importance and centrality of infrastructure interde-pendency, it was at a more conceptual and scenario-based level. Over the past decade ormore, quantified measures of interdependence have emerged, potentially providing inputsfor some of the modeling efforts underway in the area of infrastructure interdependencies.For example, Zimmerman and Restrepo [[25], p. 223] applied the ratio of the amountof time it took for electric power to be restored and the time it took for water servicesto be restored after the August 2003 blackout, finding that the restoration time for theCleveland water supply and Detroit system was at least two three times, respectively,as long as the time it took for electric power to be restored in those cities, assuming anaverage electricity outage of 24 h. Dependency on electricity-driven pumps (rather thanreliance on gravity systems) accounted for most of the delay in these areas. In othercases, backup power enables water systems to be restored more quickly than relying onthe restoration of the general electric power systems [[25], p. 226].

4 GLOBAL CONSIDERATIONS

Considerable attention has been paid to the dependence of population growth and eco-nomic development on resource capacity, and water and the infrastructure that supportsit is a key component of the resource base. A concept capturing the resource capacityand usage relationship is the “ecological footprint”, defined as utilization of resourcesby a population or economy relative to the availability or production of that resource[26]. The footprint or imbalance cited by the World Wildlife Fund (WWF) is that theuse of resources globally by 2006 has exceeded the ability to regenerate those resourcesby approximately 25% and the footprint has increased more than three times what itwas in 1961 [[26], p. 1]. Globally, water withdrawal per capita and the ratio of with-drawals to resources (water stress) vary dramatically from country to country. Globally,the correlation between water stress and the ecological footprint (defined at the countrylevel) is positive at 0.4 and statistically significant; however, if the four countries (in theMiddle East and Africa) with extreme values of water stress are eliminated, the corre-lation between these two factors approaches zero [12]. The ecological footprint appearsunrelated to water consumption, however, it seems to be qualified by two factors thatare likely to influence water consumption: a country’s income and availability of water.With respect to income, the WWF [[26], Table 2], for example, notes that high-incomecountries withdraw almost double the amount of water per capita (957,000 m3 per year)than middle- or low-income communities do (552,000 and 550,000 m3 per year, respec-tively). However, “water stress” is the same in high- and low-income countries (10%of total resources), whereas it is half of that (5% of total resources) in middle-incomecountries.

5 RESEARCH DIRECTIONS

Water systems depend upon other infrastructures, in particular, electric power, informa-tion technologies, and transportation, and indications are that this dependency is likelyto increase with technological changes in water production and delivery. Other infras-tructures in turn require water to function. These relationships imply that a disruption

12 CRITICAL INFRASTRUCTURE INTERDEPENDENCIES FOR WATER

in water systems or the infrastructures upon which water is interdependent creates a farmore complex system of impacts than is typically portrayed by a direct or single disrup-tion of one infrastructure, and in some cases can magnify the costs to human life, health,and welfare. Thus, a deeper understanding of the ramifications of these interdependenciesand their consequences should disruptions occur is needed. A means to quantify theseinterdependencies and their consequences is a necessary prerequisite to comparing thenature and magnitude of consequences across different types of interdependencies.

Vulnerabilities from interdependencies exist on top of vulnerabilities posed by anycondition or performance problems that water infrastructure may experience. Water andwastewater infrastructures nationwide were rated D–, the lowest grade given to anyinfrastructure area, by the ASCE in its 2005 infrastructure scorecard on the basis of con-dition alone from noncatastrophic sources [27]. Research is needed on how such ratingsand other assessments can incorporate interdependencies and the natural hazard and ter-rorism context. This will ultimately affect the performance and viability of infrastructuresdependent on water.

Global perspectives on water usage are important in addressing a new set of dimen-sions about dependency and interdependency that impact infrastructure and the peoplewho depend on these systems. Water usage is a key component of the global resourcebase. An understanding of how patterns of water usage by different kinds of infras-tructures influences resource capacity as measured by such concepts as the ecologicalfootprint is critical to linking security with sustainability.

Interdependencies among infrastructures have a special significance in times of emer-gencies, since emergency response is heavily dependent upon infrastructure. In fact, itis likely that the impact of such interdependencies may become magnified given thecompressed time frame necessary for emergency response.

Thus, the scope of the concept of infrastructure interdependencies is expanding andundergoing a transformation to adapt to the needs of security. The water sector representsa key part of that picture.

ACKNOWLEDGMENTS

The author wishes to acknowledge the assistance of Dr Carlos E. Restrepo, Wendy E.Remington, and Sara A. Clark (for her graphing and mapping of data on dams) ofthe Institute for Civil Infrastructure Systems at New York University (NYU)’s WagnerSchool. This research was supported by the US DHS through the Center for Risk andEconomic Analysis of Terrorism Events (CREATE), Grant number EMW-2004-GR-0112.This research was also supported in part by the US DHS through the Center for Catas-trophe Preparedness and Response at NYU, Grant number 2004-GT-TX-0001 for theproject titled Public Infrastructure Support for Protective Emergency Services . However,any opinions, findings, and conclusions or recommendations in this document are thoseof the author and do not necessarily reflect views of the US DHS.

REFERENCES

1. Zimmerman, R., and Horan, T. (2004). What are digital infrastructures? In DigitalInfrastructures: Enabling Civil and Environmental Systems through Information Technology ,R. Zimmerman, and T. Horan, Eds. Routledge, London, pp. 3–18.

REFERENCES 13

2. Hutson, S. S., Barber, N. L., Kenny, J. F., Linsey, K. S., Lumia, D. S., and Maupin, M. A.(2004). Estimated Use of Water in the United States in 2000, Circular 1268 , U.S. GeologicalSurvey, Reston, VA.

3. Rinaldi, S. M., Peerenboom, J. P., and Kelly, T. K. (2001). Identifying, understanding andanalyzing critical infrastructure interdependencies. IEEE Control Syst. Mag. 21(6), 11–25.

4. Simonoff, J. S., Restrepo, C. E., Zimmerman, R., and Naphtali, Z. S. (2008). Analysisof electric power and oil and gas pipeline failures. In Critical Infrastructure Protection,E. D. Goetz, and S. Shenoi, Eds. Springer, New York, pp. 381–394.

5. Zimmerman, R. (2006). Critical infrastructure and interdependency. In D. G. Kamien, Ed. TheMcGraw-Hill Homeland Security Handbook , The McGraw-Hill Companies, Inc., New York,pp. 523–545.

6. U.S. Department of Energy (2006). Energy Demands on Water Resources. Report to Congresson the Interdependency of Energy and Water , December. Online. Available at http://www.sandia.gov/energy-water/docs/121-RptToCongress-EWwEIAcomments-FINAL.pdf. Accessed6 July, 2007.

7. Water Sector Coordinating Council (WSSC), U.S. Department of Homeland Security andthe U.S. Environmental Protection Agency (2007). Water. Critical Infrastructure and KeyResources Sector-Specific Plan as input to the National Infrastructure Protection Plan , U.S.DHS, Washington, DC.

8. U.S. Department of Homeland Security (2007). Communications Sector Specific Plan, U.S.DHS, Washington, DC.

9. Electric Power Research Institute (2002). Water & Sustainability (Volume 4): U.S. ElectricityConsumption for Water Supply & Treatment—The Next Half Century , EPRI. Online. Availableat http://www.epriweb.com/public/000000000001006787.pdf. Accessed 5 July, 2007.

10. O’Rourke, T. D. (2007). Critical infrastructure, interdependencies, and resilience. The Bridge37(1), 22–29.

11. O’Rourke, T. D., Lembo, A. J., and Nozick, L. K. (2003). Lessons learned from the worldtrade center disaster about critical utility systems. In Natural Hazards Research & ApplicationsInformation Center, Public Entity Risk Institute, and Institute for Civil Infrastructure Systems ,Beyond September 11th: An Account of Post-Disaster Research. University of Colorado, Boul-der, CO, pp. 269–292.

12. New York University, Wagner Graduate School of Public Service (NYU-Wagner), Institute forCivil Infrastructure Systems (2006–2007). Resource Allocation Based on Critical Infrastruc-ture, Research project for the Center for Risk and Economic Analysis of Terrorism Events(CREATE) at the University of Southern California, New York University, Wagner GraduateSchool of Public Service, New York, NY.

13. Zimmerman, R. (2005). Social implications of infrastructure network interactions. In O.Coutard, R. Hanley, R. Zimmerman, Eds. Sustaining Urban Networks: The Social Diffusion ofLarge Technical Systems , Routledge, London, UK, pp. 67–85.

14. U.S. Departments of the Army and the Air Force (1987). Chapter 8: Utilidors in ARMY TM5-852-5. AIR FORCE AFR 88-19 , 5. Technical Manual.

15. Vanrenterghem-Raven, A. (2003). Modeling of the Structural Degradation of an Urban WaterDistribution System, Polytechnic University of New York, Brooklyn, NY, p. 264.

16. Zimmerman, R. (2004). Decision-making and the vulnerability of critical infrastructure. InProceedings of IEEE International Conference on Systems, Man and Cybernetics , W. Thissen,P. Wieringa, M. Pantic, M. Ludema, Eds. Delft University of Technology, The Hague.

17. Thompson, M. (2001). Much of Northern California in the Dark: Water Pumps Stopped , 17 Jan2001. Online. Available at http://www.cnn.com/2001/US/01/17/power.woes.03/index.html.

18. Simonoff, J. S., Restrepo, C. E., and Zimmerman, R. (2007). Risk management and riskanalysis-based decision tools for attacks on electric power. Risk Anal . 27(3), 547–570.

14 CRITICAL INFRASTRUCTURE INTERDEPENDENCIES FOR WATER

19. Simonoff, J. S., Zimmerman, R., Restrepo, C. E., Dooskin, N. J., Hartwell, R. V., Miller, J. I.,Remington, W., Lave, L. B., and Schuler, R. E. (2005). Electricity Case: Statistical Analysisof Electric Power Outages , CREATE Report No. 3, NYU-Wagner, New York.

20. U.S. General Accountability Office (2007). Securing Wastewater Facilities. Costs of Vulnera-bility Assessments, Risk Management Plans, and Alternative Disinfection Methods Vary Widely ,GAO-07-480, USGAO, Washington, DC.

21. Energetics, Inc (2006). Roadmap to Secure Control Systems in the Energy Sector , sponsoredby, U.S. DOE and U.S. DHS, Energetics, Inc, Columbia, MD.

22. Zimmerman, R. (2004). Water. In R. Zimmerman, and T. Horan, Eds. Digital Infrastruc-tures: Enabling Civil and Environmental Systems through Information Technology , Routledge,London, pp. 75–95.

23. Synchrony (2001). Trends in SCADA for Automated Water Systems .

24. American Society of Civil Engineers (ASCE) and American Water Works Association(AWWA) (2006). Guidelines for the Physical Security of Water Utilities , ASCE, Reston, VA.

25. Zimmerman, R., and Restrepo, C. E. (2006). The next step: Quantifying infrastructure inter-dependencies to improve security. Int. J. Crit. Infrastruct . 2(2/3), 215–230.

26. World Wildlife Fund (WWF) International, Zoological Society of London, and Global FootprintNetwork (2006). Living Planet Report 2006 , WWF, Gland.

27. ASCE (2005). Report Card for America’s Infrastructure, Online. Available at http://www.asce.org/reportcard/2005/index.cfm. Accessed 7 November, 2005.

FURTHER READING

Apostolakis, G. E., and Lemon, D. M. (2005). A Screening Methodology for the Identification andRanking of Infrastructure Vulnerabilities Due to Terrorism. Risk Anal . 25(2), 361–376.

Charles, J. (2007). Neighborhood Report: New York Up High; Longtime Emblems of City Roofs,Still Going Strong , NY Times. 3 June, 2007, p. 9.

East Bay Municipal Utilities District (2007). EBMUD Mission Statement , Online. Available athttp://www.ebmud.com/about ebmud/mission statement/. Accessed 3 July, 2007.

Electric Power Research Institute (2002). Water & Sustainability (Volume 3): U.S. Water Consump-tion for Power Production—The Next Half Century , EPRI. Online. Available at http://www.epriweb.com/public/000000000001006786.pdf. Accessed 5 July, 2007.

Ezell, B., Farr, J. V., and Wiese, I. (2000). Infrastructure risk analysis model. J. Infrastruct. Syst .6(3), 114 –117.

Ezell, B., Farr, J. V., and Wiese, I. (2000). Infrastructure risk analysis of municipal water distri-bution systems. J. Infrastruct. Syst . 6(3), 118 –122.

Hake, J. M., Gray, D. M. D., and Kallal, S. (2004). Power diet. California’s energy crisis promptsone treatment plant to re-evaluate its power intake. Water Environ. Technol . 16(5), 36–40.

National Research Council (2002). Making the Nation Safer, The National Academies Press, Wash-ington, DC.

Vanrenterghem-Raven, A. (2007). Risk factors of structural degradation of an urban water distri-bution system. J. Infrastruct. Syst. ASCE 13(1), 55–64.

Restrepo, C. E. (2004). Infrastructure and IT dimensions in the developing world. In DigitalInfrastructures: Enabling Civil and Environmental Systems through Information Technology , R.Zimmerman, and Horan T., Eds. Routledge, London, pp. 179–202.

Restrepo, C. E., Simonoff, J. S., and Zimmerman, R. Analyzing vulnerabilities in the oil andgas sector from incident data. Paper presented at the Los Alamos National Laboratories RiskSymposium 2006—Risk Analysis for Homeland Security and Defense: Theory and Application .Sante Fe, New Mexico.

CROSS-REFERENCES 15

Restrepo, C. E., Simonoff, J. S., and Zimmerman, R. (2006). Unraveling geographic interdepen-dencies in electric power infrastructure. In Proceedings of the Hawaii International Confer-ence on System Sciences , Wiley-IEEE Computer Society Press, 248a, Digital Object Identifier10.1109/HICSS.2006518. http://ieeexplore.ieee.org/xpl/freeabs all.jsp?arnumber=1579808.

U.S. Geological Survey (2004). Estimated Use of Water in the United States in 2000 , Online.Available at http://water.usgs.gov/pubs/circ/2004/circ1268/htdocs/text-total.html. Accessed 6July 6, 2007.

Wackernagle, M., and Rees, W. (1996). Our Ecological Footprint , New Society Publishers, Gabri-ola Island.

Zimmerman, R (2003) Public Infrastructure Service Flexibility for Response and Recovery inthe September 11th, 2001 Attacks at the World Trade Center. In Beyond September 11th:An Account of Post-Disaster Research, Natural Hazards Research & Applications InformationCenter, Public Entity Risk Institute, and Institute for Civil Infrastructure Systems, Universityof Colorado, Boulder, CO, pp. 241–268.

Zimmerman, R., Restrepo, C. E., Simonoff, J. S., and Lave, L. B. (2007). Risk and economiccost of a terrorist attack on the electric system. In The Economic Costs and Consequences ofTerrorism, H. W. Richardson, P. Gordon, and J. E. Moore II, Eds. Edward Elgar Publishers,Cheltenham, pp. 273–290.

CROSS-REFERENCES

Protecting Water Infrastructure in the United States

Implications of Physical and Cyber Attacks on Water and Wastewater Infrastructure