at SciVerse ScienceDirect

Journal of Environmental Psychology 33 (2013) 86e95

Contents lists available

Journal of Environmental Psychology

journal homepage: www.elsevier .com/locate/ jep

Urban and rural perceptions of ecological risks to water environments in southernand eastern Nevada

Tanju Kiriscioglu a,*, David M. Hassenzahl b, Bulent Turan c

a School of Environmental and Public Affairs, University of Nevada, Las Vegas, 4505 Maryland Parkway, Box 454030, Las Vegas, NV 89154-4030, USAb School of Sustainability and the Environment, Chatham University, USAcDepartment of Psychology, University of Alabama at Birmingham, USA

a r t i c l e i n f o

Article history:Available online 15 November 2012

Keywords:Ecological riskPerceptionInterbasin water transferDroughtWater-intensive landscaping

* Corresponding author. Tel.: þ1 702 339 2959.E-mail address: kiriscio@unlv.nevada.edu (T. Kirisc

0272-4944/$ e see front matter � 2012 Elsevier Ltd.http://dx.doi.org/10.1016/j.jenvp.2012.11.001

a b s t r a c t

In this multidisciplinary study, we used an Internet-based tool to investigate perception of ecologicalrisks to water environments due to most relevant hazards (urban development, drought, urban waterconsumption, interbasin water transfer from rural to urban areas, and water-intensive landscaping) inurban and rural Nevada. Rural participants’ perception of risk was higher than urban participants for only"interbasin water transfer from rural to urban areas" while for the other four hazards the effect ofresidence location was not significant. The principal component analysis on fourteen scales identifiedthree factors that we named Ecological Impact, Benefits & Equity due to Hazards, and Controllability ofHazards. Urban people perceived Ecological Impact due to the five hazards to water environments higherthan rural people while rural people perceived Benefits & Equity due to Hazards higher than urbanpeople. Participants’ ratings in the survey represent their judgments of benefits and equity due to thehazards to water environments in urban Nevada (not in rural Nevada). Therefore, rural people seem toperceive that urban people benefit from the risky human activities of urban development, urban waterconsumption, interbasin water transfer, and water-intensive landscaping, yet rural people incur the costs.The two groups’ risk judgments did not differ significantly in Controllability of Hazards. Participants whoperceived higher ecological impact due to risks to water environments had less water-intensive (moredesert-friendly) landscape in their gardens. And finally we found that rural laypeople perceived greaterneed to regulate risks to water environments than urban laypeople, urban experts, and rural experts, andthe latter three groups were not significantly different from each other.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction

As human population and world economies continue to grow,the demand for potable water increases, further stressing fresh-water systems (Gleick, 2004; Pimentel et al., 2004; Postel, Daily, &Ehrlich, 1996). Freshwater scarcity, exacerbated by climate changeand disturbed water cycles, is especially severe in arid south-western United States where the Colorado River’s normal flow isfully allocated and there is a water deficit in parts of the region(Global Freshwater Programme, 2007; Libecap, 2005; NationalResearch Council, 1992).

Heavy urbanization and increased demand for municipal watercompound the difficult task of water resource management in the

ioglu).

All rights reserved.

Colorado River Basin, giving rise to water transfers from ruralbasins to urban areas (Libecap, 2005; Libecap, Glennon, & Ker,2005; NRC, 1992). This practice further complicates the matter,especially while the region is in a longstanding drought for overa decade now (Dettinger, 2004; Kerr, 2007; Piechota, Timilsena,Tootle, & Hidalgo, 2004; SNWA, 2009a).

Due to “sustained and severe” water shortage in the region,Southern Nevada Water Authority (SNWA) has come up with theinterbasin water transfer plan to bring groundwater from neigh-boring rural basins in eastern Nevada to Las Vegas Valley in ClarkCounty (SNWA, 2009a). Concerned citizens, environmental groups,and scientists contend that SNWA’s interbasin water transfer planwill impact the water regimes and interests in Lincoln and WhitePine Counties in eastern Nevada, and in neighboring Millard, Juab,and Tooele Counties in western Utah (Fig. 1), resulting in adversesocial, economic, ethical, and ecological implications (Deacon,Williams, Deacon Williams, & Williams, 2007; Great Basin WaterNetwork, 2011; Sierra Club, 2006, 2008).

Fig. 1. Map of the Colorado River Basin showing the study area (within the oval arc in Nevada) of the proposed Clark, Lincoln, and White Pine Counties groundwater developmentproject.

T. Kiriscioglu et al. / Journal of Environmental Psychology 33 (2013) 86e95 87

This article examines urban and rural perceptions of ecologicalrisks to water environments in southern and eastern Nevada due towater-related hazards in general, and interbasin water transfer inparticular. In this study we adapted the psychometric paradigm onperception of risks based on cognitive risk studies dating back to1970s. One of the earliest studies in this field was the pioneeringwork of Fischhoff, Slovic, Lichtenstein, Read, and Combs (1978) thatused rating scales to assess people’s perceived risks and benefits of“risky activities” on several “attributes” such as controllability,reversibility, observability, and availability of alternatives. Thisapproach eventually led to the use of a variety of psychometricscaling methods to produce quantifiable and predictable measuresof perceived risk, perceived benefit, and other aspects of percep-tions for different types of hazards (Slovic, 1987, 1992, 2000; Slovic,Fischhoff, & Lichtenstein, 1985, 1986).

Fischoff (1985) explains why understanding perception of risksis important to risk management and risk communication.According to Fischoff, one cannot predict how people will respondto an issue without knowing how they perceive the issue to shapetheir opinions and attitudes. Since Fischoff et al. study in 1978,several risk scientists utilized the psychometric paradigm to

examine risk judgments in general, and expert and laypersonperceptions of risk in particular.

McDaniels, Axelrod, and Slovic (1995) study investigatingpeople’s perception of ecological risks, and later McDaniels,Axelrod, Cavanagh, and Slovic (1997) study investigating people’sperception of ecological risks to “water environments” were thetwo pioneering studies that examined and found differencesbetween lay and expert perceptions to ecological risks. McDanielset al. (1997) found that laypeople generally perceive ecologicalrisks higher than experts. Lazo, Kinnell, and Fisher (2000), too,confirmed that in general laypeople’s perception of ecological riskswere higher than the experts’. Current risk judgment studies focusmostly on identifying differences in expert and lay judgments, yetthere is a growing need to understand the differences in perceptionof risks between “urban” and “rural” people.

Due to water-related hazards and concerns over the proposedinterbasin water transfer plan in the study area, we believe it isimportant to investigate people’s perception of ecological risks towater environments in both the urban and rural settings insouthern and eastern Nevada while taking into account thedifferences between expert and lay risk judgments. It is also

T. Kiriscioglu et al. / Journal of Environmental Psychology 33 (2013) 86e9588

essential to understand how people’s risk judgments affect waterconsumption and water resource management in both settings.This research is significant because it is on two communities (ruraland urban Nevada) with conflicting interests where each wants tosecure future water resources for itself in the face of uncertainty(Brean, 2009; Desert Beacon, 2008; Great Basin Water Network,2011; SNWA, 2009a; Witcher, 2008). While the urban watermanagers in the study area are concerned over the growing waterdemand in urban areas, the rural residents are afraid of losingtheir “livelihood”, something they heavily depend on in farmingand ranching (Desert Beacon, 2008). This research is also impor-tant because it investigates actual risk behavior (people’s choice ofresidential landscape type) and hazard adjustments in the face ofuncertainty e something we believe has not been researchedenough in arid regions. There is a need to investigate whetherperception of risks to water resources do indeed predict actualbehavior of residential outdoor water use. Therefore, we wanted toexamine if perceptions of risks to water environments are asso-ciated with the type of landscape people have in their residentialgardens.

In this research we hypothesized that rural people in easternNevada from where SNWA plans to bring groundwater (Fig. 1)perceive interbasin water transfer as an ecological risk higher thanurban people in Las Vegas Valley (Hypothesis 1). Our secondhypothesis was that rural people’s perception of benefits and equitythat urban people (Las Vegas Valley residents) will receive due tointerbasin water transfer is higher compared to the urban people’sperception of the same (Hypothesis 2). Third, we hypothesized thatpeople in the study area who perceive the impacts of the ecologicalrisks to water environments high have environmentally friendlylandscapes in their residential gardens (Hypothesis 3). And finally,we hypothesized that rural people’s mean attitude to regulatehazards to water environments is higher compared to urbanpeople’s (Hypothesis 4).

This study can aid the ongoing policy debate on interbasinwatertransfer plan from rural basins in eastern Nevada to Las VegasMetropolitan Area, so that (1) the stakeholders’ views on both sidesof the issue and the plan’s social, economic, fiscal, ethical andecological implications can be better understood, and (2) informeddecisions can be made to develop more effective risk managementand risk communication strategies that are critical to attain watersustainability in arid regions like southern Nevada.

2. Methods

2.1. Study area

Our study area was eastern and southern/southeastern Nevada(see Fig. 1). The County of Clark and Las Vegas Metropolitan Areaare located in southern Nevada, within the Mojave Desert andColorado River Basin. Las Vegas Valley receives on average 4.5inches (11.4 cm) of precipitation per year and has an average annualevapotranspiration rate of about 90 inches (228.6 cm) (James, 1984;Shevenell, 1996). The winter temperatures can dip to around 20 �F(�6.7 �C) in the valleys while the summer temperatures can reach125 �F (51.7 �C) (James, 1984; NOAA, 2011).

White Pine and Lincoln Counties (eastern Nevada) haveterrains in both the Great Basin and the Colorado River Basin,displaying typical basin and range topography. Both counties havenorth-trending, long, narrow, parallel mountain ranges, and broad,flat valleys in between. Average yearly precipitation in thesecounties is between 6 and 9 inches (15.2e22.9 cm) (James, 1984;NOAA, 2011), and average annual evapotranspiration rate isbetween 45 and 71 inches (114.3e180.3 cm) (Shevenell, 1996).Temperatures can be as low as �30 �F (�34.4 �C) in White Pine

and as high as 110 �F (43.3 �C) in Lincoln County (James, 1984;NOAA, 2011).

White Pine County’s population is 10,030 (1.1 persons/mile2)and Lincoln County’s population is 5,345 (0.5 person/mile2) whileClark County’s population is 1.95 million with a population densityof 246.7 persons/mile2 (U.S. Census Bureau, 2011). In Lincoln andWhite Pine Counties surface waters as well as groundwaterresources are mostly used for farming, ranching, mining, andmunicipal purposes. Water use in Clark County’s urban areas ismostly for municipal purposes. Water resource management in LasVegas Valley is a challenge, especially when 90% of the area’s waterneeds are met by the Colorado River (SNWA, 2009a), the flow ratesof which have been in decline in recent years (Cohen, 2011;Morrison, Postel, & Gleick, 1996). Las Vegas Valley currently usesabout 550,000 afy (acre-feet/year) water, 60 percent of which is forconsumptive outdoor use e mainly water-intensive landscaping(SNWA, 2009a, 2009b). The current level of water use in Las VegasValley [222 gal/(capita*day)] (SNWA, 2011) is fairly high comparedto the domestic national average of 98 gal/(capita*day) (Kennyet al., 2009).

2.2. Survey development

Due to the type of data we collected, aggregate-level, hazard-focused analysis based on the traditional psychometric paradigmwas the most suitable choice for our study. Our survey consisted ofquestions on perception of ecological risks to water environments,and a behavior question on the type of landscape the participantshave in their gardens.We expected the participants to complete thesurvey in about 20e30 min.

Our approach was similar to McDaniels et al. (1997) study thatinvestigated people’s perception of ecological risks to water envi-ronments in the Fraser River Basin, British Columbia, Canada.Current study was not as comprehensive as the McDaniels et al.(1997) study that analyzed 33 hazards (items) on 17 attributes(scales). In order to stay focused on those items and scales that aremost relevant to water environments in the study area, we askedthe subjects to rate only 5 hazard items on 14 scales.

Smith (2001) posits that risk is the probability of a hazardoccurring and resulting in loss. A hazard, on the other hand, isa naturally occurring or human-induced process that has thepotential to cause harm to humans and/or their welfare (Smith,2001). According to Smith (2001), a natural hazard can occur inany uninhabited region of the world and will remain a hazard, butthe same hazard becomes a risk if it occurs in an area where“humans and their possessions” exist. McDaniels et al. (1995, 1997)classify ecological hazards as a broad range of processes thatinclude natural events, technologies, human activities, and social,political, and economic beliefs and systems that pose a threat to the“health and productivity of species and natural environmentalsystems”. The term “hazard”we used in this article is comparable tothe term McDaniels et al. (1995, 1997) used in their studies.

The attributes (scales) we picked for this study were“Controllability”, “Alternatives”, “Impacts on animals/plants”,“Benefits”, “Equity”, “People affected”, “Area affected”, “Humanhealth”, “Experts’ knowledge”, “Immediacy of impacts”, “Revers-ibility of impacts”, “Observability”, and “Risks to water resources”;as well as “Need to regulate” as a dependent variable (see Table 1).In risk judgment studies a hazard can be a natural event (likedrought or earthquake) or it can be an anthropogenic activity (likeurban water consumption or agricultural pesticide use). Thehazard items we chose were “Urban development”, “Drought”,“Urban water consumption”, “Interbasin water transfer from ruralto urban areas”, and “Water-intensive landscaping (residential,commercial, and public e including golf courses)”. Drought

Table 1Description of scales and response categories as presented in the survey.

Scale Description of scale Scale endpoints

Low (1) High (7)

Controllability Using the following scale please click the appropriate circle torate how controllable each of the following is, in terms of ourability to control its impact on water resource systems insouthern/southeastern Nevada

Not at all controllable Very controllable

Alternatives Using the following scale please click the appropriate circle torate the extent to which there are alternatives to each of thefollowing that may have less impact on water resource systemsin southern/southeastern Nevada

Alternatives notavailable

Alternatives available

Impacts onanimals/plants

Using the following scale please click the appropriate circle torate how much suffering there would be by animals and plantsas a result of each of the following in southern/southeasternNevada’s water resource systems

No impact Great impact

Benefits Using the following scale please click the appropriate circle torate how you think each of the following provides benefits tosociety in southern/southeastern Nevada

No benefit Great benefit

Equity Using the following scale please click the appropriate circle torate the equity of each of the following taking place insouthern/southeastern Nevada in terms of whether those whoreceive the benefits from the event are the same people whoincur the costs

Not equitable Very equitable

People affected Using the following scale please click the appropriate circle torate how many people are, or could be, affected by the impacteach of the following may have on southern/southeasternNevada’s water resource systems

Very few people Great number of people

Area affected Using the following scale please click the appropriate circle torate the scope of the impacts of each of the following in termsof the area of water resources potentially affected withinsouthern/southeastern Nevada

Small area Large area

Human health Using the following scale please click the appropriate circle torate the extent to which each of the following and its associatedeffects on southern/southeastern Nevada’s water resourcesystems pose a risk to human health

Poses no risk tohuman health

Poses great risk tohuman health

Experts’ knowledge Using the following scale please rate how much you feel theexperts in southern Nevada know about the harmful impactsof each of the following on southern/southeastern Nevada’swater environments

Very little known Great amount known

Immediacy of impacts Using the following scale please click the appropriate circle torate the immediacy of each of the following, in terms of howsoon possible harmful effects may occur in southern/southeasternNevada’s water resource systems

Occurs immediately Occurs far in the future

Reversibility of impacts Using the following scale please click the appropriate circle torate the extent to which impacts on southern/southeasternNevada’s water resource systems are reversible (i.e., the conditionsof the water resource systems will return to pre-event conditions)

Completely irreversible Completely reversible

Observability Using the following scale please click the appropriate circle torate how observable the impacts of each of the following are onwater resource systems in southern/southeastern Nevada

Not at all observable Very observable

Risks to water resources Using the following scale, please rate how risky you think each ofthe following is in terms of its impacts on southern Nevada’ssurface waters and groundwater

Poses no risk towater resources

Poses great risk towater resources

Need to regulate Using the following scale, please rate the extent to which you feeleach of the following requires regulation due to its potentialadverse impacts on southern/southeastern Nevada’s surface watersand groundwater

Requires no regulation Requires strict regulation

T. Kiriscioglu et al. / Journal of Environmental Psychology 33 (2013) 86e95 89

hazard is the only natural event in the current study while theother four hazards are anthropogenic. For each hazard item, theparticipants used a 7-point semantic differential scale to rate theirperception of risks to water environments in the study area(Table 2).

Wealsoasked theparticipantswhatkindof landscape theyhaveattheir residence as a behavior question. For this question, we assignedthe highest score of “4” for the zero-scape/native desert landscape thatused the least amount of water, a score of “3” for xeric landscape,a score of “2” formesic landscape, and the lowest score of “1” for thehydric landscape that used the most amount of water. Participantswhodidnot knowtheir landscape typeordidnothave any residentialgarden (n ¼ 9) were excluded from the analyses using this variable.

2.3. Participants and data collection

The majority of data we collected belonged to people in ClarkCounty (urban and rural participants), and in White Pine andLincoln Counties (rural participants). Rural people elsewhere inNevada (especially in Nye County), and in neighboring counties inwestern Utah (n ¼ 3; laypeople) who have direct ties to easternNevada also participated in this study.

Data collection for our cross-sectional study was through theonline survey tool SurveyMonkey�, the web address of which wemailed to urban laypeople, and e-mailed to rural laypeople, andurban and rural experts. The participants took the survey at theirconvenience during the online survey’s duration e from early May

T. Kiriscioglu et al. / Journal of Environmental Psychology 33 (2013) 86e9590

in 2010 until mid-June in 2010 e and we did not compensate theparticipants for taking the survey.

The laypeople sample group consisted of (1) randomly selected(via proportionate stratified random sampling) urban residents inLas Vegas Valley, and (2) non-randomly selected (via conveniencesampling) rural residents in eastern Nevada, from where SNWA isplanning to bring water to Las Vegas Valley. The urban laypeoplestrata in Las Vegas Valley were the United States Postal Service’s ZIP(Zone Improvement Plan) code areas. The sampling size in eachstratum was proportional to the ZIP code area’s population:Densely populated areas received higher rate of random samplingin comparison to sparsely populated areas. The random sampling inLas Vegas Valley resulted in each ZIP code area receiving fairrepresentation. The rural laypeople in the sample group were themembers of the water networks and associations living in WhitePine and Lincoln Counties, and people living in neighboring Nevadaand Utah counties who have direct ties to the study area.

The expert sample group consisted of non-randomly selected(convenience sampling) University of Nevada, Las Vegas (UNLV)faculty at the School of Environmental and Public Affairs, Civil andEnvironmental Engineering Department, Geo-science Department,Water Resources Management Program, and School of LifeSciences; Harry Reid Center for Environmental Studies scientists;Desert Research Institute (DRI) scientists; Southern Nevada WaterAuthority (SNWA), Las Vegas Valley Water District, and othersouthern Nevada water district officials and scientists; and scien-tists from other relevant organizations at the local, state, regional,and federal level. The experts in the current study have expertise inat least one of the following: regional ecosystems, human-naturesystems, urban management, water resource management, andrisk studies. Urban experts are those that reside in urban areas, andrural experts are those that reside in rural areas. We believe thatmost of these experts in the study have expertise in both the urbanand rural parts of the study area.

We contacted 1150 people by mail, and about 500 people by e-mail. The survey was a one-wave survey: We contacted eachparticipant once, and wemade no attempt to ensure the delivery ofthe e-mail messages. 127 survey letters were returned to us, andwedon’t know how many e-mails failed to reach their destination.Overall, 181 people (11% of the sample pool) participated, and only115 of them (62.2% of the participants) completed the survey withthe following breakdown: urban laypeople (n¼ 32), rural laypeople(n¼ 33), urban experts (n¼ 27), and rural experts (n¼ 23). 64.6% ofthe participants were male and 35.4% of the participants werefemale; 69% were laypeople and 41% were experts; 13.8% of theparticipants had a high school degree, 36.2% had some collegeeducation or a junior college/bachelor’s degree, and 64.3% had

Table 2The ecological hazards to water environments in the survey.

Risks to water resources

Using the scale below, please rate how risky you think each of the following is in terms oPoses no risk to water resources

1 2 3

1. Urban developmentB 1 B 2 B 3

2. DroughtB 1 B 2 B 3

3. Urban water consumptionB 1 B 2 B 3

4. Interbasin water transfer from rural to urban areasB 1 B 2 B 3

5. Water-intensive landscaping (residential, commercial, and public eincluding golf coB 1 B 2 B 3

Note. The environmental hazards that we used are Urban development, Drought, Urban wintensive landscaping (residential, commercial, and public eincluding golf courses). Her

a graduate degree; 14.9% of the participants were between the agesof 18e34, 39.5% were between 35 and 54, and 45.6% were 55 orolder; and 96% of the participants were Caucasian.

3. Results

3.1. Analysis of urban and rural ratings of ecological risks to waterenvironments

We averaged all 14 scales for each of the five hazard items forboth the urban and rural people; the results are shown in Table 3. Inorder to examine whether or not differences between urban andrural people’s risk perceptions differ by type of ecological hazardwe conducted a repeated measures generalized linear modelanalysis (GLM). The type of hazard (Urban development, Drought,Urban water consumption, Interbasin water transfer, and Water-intensive landscaping) was the within-subjects variable while thebetween-subject variables were: (a) residence location (urban vs.rural) and (b) respondent type (expert vs. lay). The interactionterms (residence location by type of hazard and respondent type bytype of hazard) were also entered as predictors. The interactionbetween respondent type and type of hazard did not reach signif-icance: F (4, 109) ¼ 2.21, p ¼ .07, partial eta2 ¼ 0.08. Importantly,and as expected, the interaction between residence location andtype of hazard was a significant predictor of perceptions of risk; F(4, 109) ¼ 5.40, p < .01, partial eta2 ¼ 0.17. Fig. 2 shows this inter-action effect, suggesting that rural people perceive the interbasinwater transfer as riskier than urban people, whereas rural peopledo not perceive the other four types of hazards to water environ-ments differently than their urban counterparts.

Follow up univariate GLM analyses where one by one each of thefive hazard types was used as the dependent variable confirmedthis result: Rural (M ¼ 5.40, SD ¼ 0.97) and urban participants(M ¼ 4.76, SD ¼ 0.92) differed in their perception of risk for onlyinterbasin water transfer; F (1, 115) ¼ 13.27, p < .01, partialeta2 ¼ 0.11 (again controlling for respondent type). For the otherfour hazards the effect of residence location was not significant.That is, rural people do not perceive interbasin water transferriskier compared to urban people only because they have higherrisk perceptions in general.

3.2. Principal component analysis

To ascertain whether a pattern of underlying dimensions exists,we conducted a principal component analysis on the means of thefive hazards averaged for each of the 13 attribute scales. Just like theMcDaniels et al. (1997) study we did not include “Need to regulate”

f its impacts on southern/southeastern Nevada’s surface waters and groundwater:Poses great risk to waterresources

4 5 6 7

B 4 B 5 B 6 B 7

B 4 B 5 B 6 B 7

B 4 B 5 B 6 B 7

B 4 B 5 B 6 B 7urses)

B 4 B 5 B 6 B 7

ater consumption, Interbasin water transfer from rural to urban areas, and Water-e we show the scale “Risks to water resources” as an example.

Fig. 2. Mean levels of perceived ecological risk to water environments for the fivetypes of hazards for urban and rural people.

Table 3Mean perception of risks due to each hazard across 14 scales for urban and ruralpeople.

Sample subgroup n Mean Std.deviation

MeanURBANDEVa Urban laypeople 32 5.13 0.51Rural laypeople 33 5.28 0.96Urban experts 27 5.06 0.80Rural experts 23 5.14 0.71Urban people 59 5.10 0.65Rural people 56 5.22 0.86Urban & ruralcombined

115 5.16 0.76

MeanDROUGHTa Urban laypeople 32 5.36 0.48Rural laypeople 33 5.26 0.63Urban experts 27 5.17 0.55Rural experts 23 5.06 0.59Urban people 59 5.27 0.52Rural people 56 5.18 0.62Urban & ruralcombined

115 5.23 0.57

MeanURBWATERCONa Urban laypeople 32 5.11 0.54Rural laypeople 33 5.32 0.81Urban experts 27 4.87 0.72Rural experts 23 5.00 0.71Urban people 59 5.00 0.64Rural people 56 5.19 0.78Urban & ruralcombined

115 5.09 0.71

MeanINTBASWATERTRANSa Urban laypeople 32 4.64 0.67Rural laypeople 33 5.52 1.02Urban experts 27 4.90 1.16Rural experts 23 5.25 0.89Urban people 59 4.76 0.92Rural people 56 5.40 0.97Urban & ruralcombined

115 5.07 1.00

MeanWATERINTSVLANDSCPa Urban laypeople 32 4.88 0.66Rural laypeople 33 5.30 0.86Urban experts 27 4.80 0.76Rural experts 23 4.52 0.77Urban people 59 4.85 0.70Rural people 56 4.98 0.91Urban & ruralcombined

115 4.91 0.81

Note. a MeanURBANDEV stands for the mean of “Urban development”, Mean-DROUGHT for the mean of “Drought”, MeanURBWATERCON for the mean of “Urbanwater consumption”, MeanINTBASWATERTRANS for the mean of “Interbasin watertransfer from rural to urban areas”, and MeanWATERINTSVLANDSCP for the mean of“Water-intensive landscaping (residential, commercial, and public eincluding golfcourses)”.

T. Kiriscioglu et al. / Journal of Environmental Psychology 33 (2013) 86e95 91

in the principal component analysis because it was used asa dependent variable. The goal in doing the principal componentanalysis was to reduce the information in several variables into a setof weighted linear combinations of those variables.

Our factor analysis (N ¼ 115) produced 3 factors. With Varimaxfactor rotation we identified those orthogonal factors that hadeigenvalues greater than 1.0, and then we verified these results onthe scree plot. The scree plot illustrated a definite break after thethird factor and the plot flattened thereafter. Table 4 shows thefactors that we identified at the conclusion of the principalcomponent analysis and the total variance explained. McDanielset al. (1997) study identified four factors, explaining 90% of thevariance while our analysis yielded only three factors, explaining57.2% of the variance. In our analysis Impacts on animals and plants,People affected, Area affected, Immediacy of impacts, Observabilityand Risks to water resources loaded on factor 1, and this factorexplained 24.6% of the variance. We named this factor “EcologicalImpact”. Benefits, Equity, and Experts’ Knowledge loaded on factor 2,explaining 18.9% of the variance, and we named this factor “Benefits

& Equity due to Hazards”. The scale Benefits measures whether anyof the five hazards in the study provides benefits to society insouthern/southeastern Nevada while the scale Equity measureswhether those who receive the benefits due to any of the hazardsare the same people who incur the costs. The scales Controllabilityand Alternatives loaded on factor 3, explaining 13.7% of the variance.We called this factor “Controllability of Hazards”. The scales Humanhealth and Reversibility did not load in any of the three factors in ourstudy (see Table 4).

3.3. Comparison of the three factor scores among urban and ruralsamples

Next, we created weighted factor scores for each participant andexamined the effects of residence location on the three factor scores.We conducted a repeated measures generalized linear model anal-ysis (GLM). The type of factor (factors 1e3) was the within-subjectsvariable. The between-subject variableswere: (a) residence location(urban vs. rural) and (b) respondent type (expert vs. lay). Theinteraction terms (residence location by type of factor and respon-dent type by type of factor) were also entered as predictors. Theinteraction between respondent type (expert vs. lay) and type offactorwasnot significant; F (2,106)¼0.41, n. s., partial eta2¼0.01. Asexpected, however, the interaction between residence location(urban vs. rural) and type of factor was a significant predictor ofperceptions of risk; F (2, 106) ¼ 11.70, p < .01, partial eta2 ¼ 0.18.

Follow-up univariate GLM analyses where one by one each ofthe three factors was used as the dependent variable revealed thaturban participants’ perception of risks to water environments(M ¼ 0.22, SD ¼ 0.85) was higher for factor 1 (Ecological Impact)compared to rural participants (M ¼ �0.10, SD ¼ 1.03); F (1,110) ¼ 4.32, p < .05, partial eta2 ¼ 0.04. (Respondent type and theother two factors were also entered as covariates.) For factor 2(Benefits & Equity due to Hazards), however, urban participants’perception of risks to water environments (M ¼ �0.39, SD ¼ .81)was lower than rural participants’ (M ¼ 0.41, SD ¼ 1.07); F (1,110) ¼ 19.97, p < .01, partial eta2 ¼ 0.16. That is, the urban peopleperceive the ecological impacts due to the five hazards to water

Table 4Rotated factor loadings for risk characteristic scales.

Characteristic Factor 1EcologicalImpact

Factor 2Benefits &Equity dueto Hazards

Factor 3Controllabilityof Hazards

McDaniels et al.(1997) studya

Impacts on animals & plants 0.68 1People affected 0.76 1Area affected 0.66 1Immediacy of impacts 0.64 3Observability 0.65 3Risks to water resourcesb 0.68Benefits 0.76 2Equity 0.74 2Experts’ knowledge 0.74 3Controllability �0.81 4Alternatives 0.76 4Human healthc 1Reversibilityc 1Rights of nonhuman speciesd 1Relevanced 1Predictabilityd 3

Factor performance Factor 1 Factor 2 Factor 3

Explained variance 24.56 18.95 13.73Eigenvalue 3.19 2.46 1.79

Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization.Only loadings greater than 0.60 in absolute value are shown.

a The column labeled McDaniels et al. (1997) shows on which factor the same variable loaded in McDaniels et al. (1997) study.b McDaniels et al. (1997) did not have an independent variable named Risks to water resources like we did; we therefore had to leave it blank in McDaniels et al. (1997)

study’s column.c The scales Human health and Reversibility did not load on any of the three factors in our study.d We did not use the scales Rights of nonhuman species, Relevance, and Predictability in our study; these scales loaded on Factor 1 in McDaniels et al. (1997).

T. Kiriscioglu et al. / Journal of Environmental Psychology 33 (2013) 86e9592

environments higher than the rural people while the rural peopleperceive the benefits and equity due to the five hazards to waterenvironments higher than the urban people in the study area. Thetwo groups did not differ significantly on factor 3 (Controllability ofHazards); F (1, 110) ¼ 0.23, n. s., partial eta2 ¼ 0.00.

3.4. Comparison of landscape type and need to regulate amongurban and rural samples

There is a scarcity of research on the association betweenperception of ecological risks to water resources and actualbehavior. We wanted to test (1) whether factors on ecological riskperception to water environments do indeed predict actual choicesconcerning landscape type in residential gardens, and (2) whetherthe association between people’s risk perceptions and landscapetype in their gardens is different for urban and rural people. Weconducted a univariate GLM analysis where the landscape typeserved as the dependent variable. The predictors were: residencelocation, respondent type, scores on the three factors, and the threeinteraction terms between residence location and each factor score.Factor 1 (Ecological Impact) was indeed a significant predictor ofresidential landscape type; F (1, 96) ¼ 5.20, p < .05, partialeta2 ¼ 0.06: People in the study area who perceived higherecological impacts due to ecological risks to water environmentshad less water-intensive (more desert-friendly) gardens. Thoseparticipants who thought that the five hazards in the study wouldhave adverse effects on humans, animals, plants, and the produc-tivity of the water resources that support the local human-naturesystems exhibited environmentally friendly behavior by havinglandscape types in their residential gardens that use less water. Thismeans that they have adjusted themselves to negative conse-quences of potential hazards to water resources. The interactionbetween factor 1 (Ecological Impact) and residence location alsoapproached significance, F (1, 96)¼ 3.19, p¼ .08, partial eta2 ¼ 0.04.Even though the interaction did not reach significance, we explored

the association between factor 1 scores and landscape type sepa-rately for rural and urban participants. Factor 1 was a significantpredictor of residential landscape type for rural people witha moderate to large effect size; F (1, 41) ¼ 7.92, p < .01, partialeta2 ¼ 0.18. However, factor 1 was not a significant predictor ofresidential landscape type for urban people; F (1, 55)¼ 0.15, p¼ .70,partial eta2 ¼ 0.00. It is a tentative conclusion that, even thoughurban people perceive the impacts of the ecological risks to waterenvironments due to the hazardous activities/events in the studyhigher, this is not necessarily translated into their choices forresidential landscape type in their gardens.

Thus, the degree to which urban people are willing to sacrificepersonally does not seem to be related to their perceptions of risk.This may mean that efforts to increase awareness about possiblerisks to water environments may not be enough to change urbanlaypeople’s behaviors related to their water use: It may be neces-sary to have stricter regulations on water use, especially in urbanareas. We examined rural and urban participants’ views on theneed to regulate hazards to water environments. A GLM analysiswas conducted where (a) residence location (urban vs. rural), (b)respondent type (expert vs. lay), and (c) the interaction of these twovariables served as independent variables. The dependent variablewas the perception of the need to regulate. The interaction termwas significant; F (1, 115) ¼ 4.26; p < .05, partial eta2 ¼ 0.04. Asshown in Fig. 3, tests of simple main effects revealed that rurallaypeople perceived greater need to regulate hazards to waterenvironments than urban laypeople F (1, 111)¼ 5.44, p< .05, partialeta2 ¼ 0.05. However, urban and rural experts did not differ fromeach other significantly; F (1, 115) ¼ 0.49; n. s., partial eta2 ¼ 0.00.Post-hoc tests also revealed that rural laypeople perceived greaterneed to regulate risks to water environments than urban laypeople,urban experts, and rural experts, and the latter three groups werenot significantly different from each other.

To summarize, Hypothesis 1 was supported: We found that, onaverage, rural people in eastern Nevada fromwhere SNWA plans to

Fig. 3. Mean levels of perceived need to regulate the five hazards to water environ-ments by urban and rural experts and laypeople.

T. Kiriscioglu et al. / Journal of Environmental Psychology 33 (2013) 86e95 93

bring groundwater perceive interbasin water transfer as anecological risk higher than people in Las Vegas Valley. Hypothesis 2was supported as well because we found that, on average, ruralpeople’s perception of benefits and equity that urban people (LasVegas Valley residents) will receive due to interbasinwater transferis higher compared to the urban people’s perception of the same.The results indicate that, on average, those people in the study areawho perceive the environmental impacts of ecological risks towater environments higher do indeed have environmentallyfriendlier landscapes in their residential gardens. Therefore,Hypothesis 3 was supported. And finally, Hypothesis 4 was onlypartially supported because, on average, only the rural laypeople’smean attitude to regulate hazards to water environments washigher compared to urban laypeople’s. The same hypothesis did nothold for experts: On average, the urban experts’ mean attitude toregulate hazards to water environments was not different than therural experts’ mean attitude to regulate the same.

4. Discussion and conclusion

There is an increasing concern in public debate about ecologicalrisks. Many human activities that involve environmental risksadversely impact human-nature systems. In recent years, thesituation has become worse with some very large, interconnected,and complex environmental changes that have taken place e suchas biodiversity loss, stratospheric ozone depletion, and globalclimate change. Due to fast-paced advances in science and tech-nology humanity now faces more human-induced risks than anyother time in recorded history. In today’s world humans live ina risk society (Adams, 1995).

One of the potential human-induced environmental risks is theworldwide trend to bring water from ecologically and socio-economically sensitive rural areas to growing, water-deprivedurban centers (Celio, Scott, & Giordano, 2010; Cohen, 2011). Citiesthat have traditionally secured water for their growth from nearbysources (rivers, lakes, aquifers, etc.) are now increasingly turning todiverted water from distant rural basins. This trend, however,creates conflict between urban and rural areas where eachwants tosecure water to meet its current and future needs (Swyngedouw,2009) e especially in areas where climate change, drought, ordisturbed water cycles are a major concern.

People living in eastern Nevada’s rural areas assert that the stateshould not allow Southern Nevada Water Authority to take

groundwater out of their basins because excessive pumping canadversely impact existing water supplies and the socioeconomicstructure in these basins. Whenwe designed this study, we figuredit would beworth to investigate the differences between urban andrural, as well as laypeople and expert perceptions of ecological risksto water environments in southern and eastern Nevada.

The study presented some complications in the early stages.The time frame of the study inadvertently coincided with theharvesting period in most of the rural areas where we surveyed. Inorder to avoid respondent fatigue and ensure high participation inrural areas, we had to make some changes to shorten the survey,yet in spite of this effort the completion rate of the survey waslower than our expectations. Like any other survey study, this onehad its limitations as well. Surveys can establish whether or nota relationship exists between two variables, but they are notsufficient to determine causality; the current study did notattempt it either. Second, surveys are subject to bias, and we don’tknow how biased the respondents’ answers were. Third, onlinesurveys may be an inconvenience to some potential survey takerswho may not be able to take the survey because of the lack ofcomputer and/or Internet connection. We do not know how manyof the contacted subjects failed to take the online survey becauseof these reasons. And last, if conducted correctly, surveys can berepresentative of the opinions and judgments of a sample groupwithin the population of interest, yet we can never be sure howrepresentative the sample group’s opinions and judgments wereof the population in the study area.

In this study we were interested in investigating the relevantattributes of those hazardous activities/events that pose a risk towater environments in the study area, and finding out what kind offactor structure the dataset would generate in principal componentanalysis. The results in this study provide some empirical supportfor the factor structure found in McDaniels et al. (1997). We endedup with a three-factor structure of how people in the study areaperceive risks to water environments, and just like in McDanielset al. (1997) study, explained variance was highest for theperceived risk to water environments due to ecological impacts.

In the current study the interaction between residence locationand type of hazard was a significant predictor of perception ofecological risk. Rural people perceive the “interbasin water trans-fer” as riskier than urban people, whereas they do not perceive theother four types of hazards (“Urban development”, “Drought”,“Urbanwater consumption”, and “Water-intensive landscaping”) towater environments noticeably different compared to their urbancounterparts.

The interaction between residence location (urban vs. rural) andtype of factor was also a significant predictor of perception of risks.Urban participants’ perception of risks to water environments washigher for factor 1 (Ecological Impact) compared to rural partici-pants. For factor 2 (Benefits & Equity due to Risks), however, ruralparticipants’ perception of risks to water environments was highercompared to urban participants. That is, the urban people perceivethe ecological impacts due to the five hazards to water environ-ments higher than the rural people while the rural people perceivethe benefits and equity due to the five hazards to water environ-ments higher than the urban people. It is important to note,however, that rural participants’ ratings in the survey representtheir judgments of benefits and equity due to the five hazards towater environments in southern (urban) Nevada e not in eastern(rural) Nevada. Rural participants assert that (1) urban residents insouthern Nevada benefit from the risky human activities of “Urbandevelopment”, “Urban water consumption”, “Interbasin watertransfer”, and “Water-intensive landscaping”, and (2) those whoreceive the benefits from any of these risky activities are not thesame people who incur the costs. In other words, the rural people

T. Kiriscioglu et al. / Journal of Environmental Psychology 33 (2013) 86e9594

claim that the urban people get all the benefits caused by theserisky activities, yet it is the rural people who end up paying theprice.

There is a growing need to investigate actual behavior of peoplewhile studying their perception of risks to water environments.O’Connor, Bord, and Fisher (1999) showed that risk perceptions andknowledge generally increase people’s willingness to take steps toaddress environmental problems. In the current study, however, wefound that the urban residents in the sample group know the waterresource availability and drought conditions in the region, havefairly strong attitudes toward water resource related issues, yet theresults tentatively imply that the urban residential landscape typesare not necessarily desert-friendly enough for an arid region thathas awater deficit and is in a longstanding drought. Future researchusing intensive interviews and qualitative research methods mightclarify the processes involved in this interesting finding.

The sampled urban group may be aware of the hazards to waterenvironments as well as environmental phenomena such aschanged weather patterns, disturbed water cycles, drought, andglobal climate change, yet they may not be fully aware of thepotential implications. Corollary, the sampled urban group may beunder the impression that they are not facing any immediatethreats to their water resources. They may continue with theirwater use habits until the time when the changes in behavior willbe necessary e a reactive as opposed to a proactive approach. Andfinally, the sampled urban group may think that they don’t need tomake any sacrifices and changes in their gardens to reduce theirconsumptive water use. Therefore, it is important to investigatewhether urban people in the study area fully understand theimplications of the ecological hazards to water environments and ifthere is a need to improve the efficaciousness of risk communica-tion in regards to these hazards.

As McDaniels et al. (1997) explain, the extent to which a hazardis believed to be controllable is likely to influence the society’sjudgments for the need to regulate that hazard. If a communityperceives a hazard as “not controllable”, or denies that a certainhazard exists, this may lead to “communal inaction” (McDanielset al., 1997). On a case like this, risk managers may have to inter-vene to communicate effective and efficient strategies, andformulate policies that will require community participation andthus minimize potential risks associated with that hazard. In thestudy area, “Drought” is a natural hazard while the other fourhazards (“Urban development”, “Urban water consumption”,“Interbasin water transfer”, and “Water-intensive landscaping”) areanthropogenic. Obviously one has no control over drought, yet wecan control the anthropogenic hazards to avoid or mitigate adverseimplications on human-nature systems in the study area. Samplegroup agrees that there is a need to regulate the anthropogenichazards to water environments in the study area, and the rurallaypeople have the strongest attitude among the four subgroups toregulate these hazards: Rural laypeople perceive greater need toregulate risks to water environments than urban laypeople, urbanexperts, and rural experts, and the latter three groups were notsignificantly different from each other (Fig. 3).

An interesting finding in the current study was the level of riskjudgment for the urban residents in regard to interbasin watertransfer. Before the current study, we knew how strongly the ruralpeople, environmentalists, and certain special interest groups havebeen opposing to the interbasin water transfer plan from easternNevada to Las Vegas Valley, yet we did not know how the urbanpeople felt about this. Although significantly lower than their ruralcounterparts, urban people’s perception of risks due to the inter-basin water transfer was high enough (slightly less than 4.8 ona scale 1e7) to indicate that they see at least a moderate level ofriskiness in this plan. We need to further investigate this attitude,

so that the decision-makers can better understand the will of theurban people in regards to the interbasinwater transfer. In addition,future studies using intensive interviews and field visits might helpus understand the implications of the current findings morecompletely.

Municipal water use is the fastest growing sector in the Colo-rado River Basin, placing pressure on a river system that is alreadyover-allocated (Fig. 1) and is facing the prospect of long-termdeclines in run-off due to climate change (Cohen, 2011). Sinceconsumptive water use is a problem in most urban areas of theColorado River Basin, we recommend more comprehensive studiesthat will investigate ways to cut down per capita urbanwater use inthe region.

We believe that better water conservation practices will lead tofaster water sustainability in the region. If Las Vegas Valley resi-dents use water more efficiently, especially outdoors where wateruse is mostly consumptive, there may not even be a need to bringwater from distant rural basins where people perceive the associ-ated risks due to interbasin water transfer higher. Formulation ofeffective and efficient policies for sustainable water use andresource management in Las Vegas Valley will ensure that therights of stakeholders and natural systems are not violated, andscarce resources are conserved. This way Las Vegas Valley will bea sustainable model not only in the Colorado River Basin, but also inother arid regions of the world.

Acknowledgments

We thank James E. Deacon, Paul Slovic, and Nick Allum for theirexcellent contributions to this work. We are particularly thankful toSusanna Hornig Priest, Helen R. Neill, and Thomas C. Piechota fortheir informational support and invaluable input. And we finallythank to School of Environmental and Public Affairs, University ofNevada, Las Vegas; and Marianne Stewart Carpenter for makingthis research project possible.

References

Adams, J. (1995). Risk. London: UCL Press Limited.Brean, H. (2009, August 21). Pipeline receives go-ahead. Las Vegas Review Journal,

A1.Celio, M., Scott, C. A., & Giordano, M. (2010). Urbaneagricultural water appropria-

tion: The Hyderabad, India case. The Geographical Journal, 176(1), 39e57.Cohen, M. J. (2011). Municipal deliveries of the Colorado River Basin water. Oakland,

CA: Pacific Institute, Retrieved from. www.pacinst.org/reports/co_river_municipal_deliveries/crb_water.pdf.

Deacon, J. E., Williams, A. E., Deacon Williams, C., & Williams, J. E. (2007). Fuelingpopulation growth in Las Vegas: How large-scale groundwater withdrawalcould burn regional biodiversity. BioScience, 57(8), 688e698.

Desert Beacon. (2008). Plan alert on latest Las Vegas water grab. Retrieved from.http://desertbeacon.wordpress.com/2008/02/05/plan-alert-on-latest-las-vegas-water-grab/.

Dettinger, M. D. (2004). Coping with severe and sustained drought in the Southwest.Retrieved from. http://geochange.er.usgs.gov/sw/changes/natural/codrought.

Fischhoff, B. (1985). Managing risk perceptions. Issues in Science and Technology,II(1), 83e96.

Fischhoff, B., Slovic, P., Lichtenstein, S., Read, S., & Combs, B. (1978). How safe is safeenough? A psychometric study of attitudes toward technological risks andbenefits. Policy Sciences, 9, 127e152.

Gleick, P. H. (2004). Global freshwater resources: Soft-path solutions for the 21stcentury. Science, 302, 1524e1528.

Global Freshwater Programme, World Wildlife Federation. (2007). Pipedreams?Interbasin water transfers and water shortages. Global Freshwater Programme,Retrieved from. http://assets.panda.org/downloads/pipedreams_ibts_final_report_27_june_2007_1.pdf.

Great Basin Water Network. (2011). Don’t drain Nevada. Retrieved from. http://greatbasinwaternetwork.org/.

James, J. W. (1984). Climate of Nevada (Paper No. 84-12). Reno, Nevada: Bureau ofBusiness and Economic Research, University of Nevada.

Kenny, J. F., Barber, N. L., Hutson, S. S., Linsey, K. S., Lovelace, J. K., & Maupin, M. A.(2009). Estimated use of water in the United States in 2005. Retrieved from.http://pubs.usgs.gov/circ/1344/.

T. Kiriscioglu et al. / Journal of Environmental Psychology 33 (2013) 86e95 95

Kerr, R. I. (2007). Climate change: Global warming is changing the world. Science,316, 188e190.

Lazo, J. K., Kinnell, J. C., & Fisher, A. (2000). Expert and layperson perceptions ofecosystem risk. Risk Analysis, 20(2), 179e193.

Libecap, G. D. (2005). The problem of water. Tucson: University of Arizona, NationalBureau of Economic Research.

Libecap, G. D., Glennon, R., & Ker, A. P. (2005). The implications of legal and politicalinstitutions on water transfers in the West. Arizona Review, 3, 3e4.

McDaniels, T. L., Axelrod, L. J., Cavanagh, N. S., & Slovic, P. (1997). Perception ofecological risk to water environments. Risk Analysis, 17(3), 341e352.

McDaniels, T. L., Axelrod, L. J., & Slovic, P. (1995). Characterizing perception ofecological risk. Risk Analysis, 15(5), 575e588.

Morrison, J. I., Postel, S. L., & Gleick, P. H. (1996). The sustainable use of water in theLower Colorado River Basin: A joint report of the Pacific Institute for Studies inDevelopment, Environment, and Security & the Global Water Project prepared withthe support of the United Nations Environment Programme and the Turner Foun-dation. Oakland, CA: Pacific Institute for Studies in Development, Environment,and Security [PISDES], Retrieved from. http://www.pacinst.org/reports/sustainable_co_river/sustainable_co_river_report.pdf.

National Oceanic and Atmospheric Administration (NOAA). (2011). NationalWeather Service Forecast Office. Retrieved from. http://www.weather.gov/climate.

National Research Council (NRC). (1992). Commission on Geosciences, Environment,and Resources. Water transfers in the west: Efficiency, equity, and the environment.Washington (DC): National Academy Press.

O’Connor, R. E., Bord, R. J., & Fisher, A. (1999). Risk perceptions, general environ-mental beliefs, and willingness to address climate change. Risk Analysis, 19(3),461e471.

Piechota, T., Timilsena, J., Tootle, G., & Hidalgo, H. (2004). The Western U.S. drought:How bad is it? Eos Transactions, American Geophysical Union, 85(32), 301e308.

Pimentel, D., Berger, B., Filiberto, D., Newton, M., Wolfe, B., Karabinakis, E., et al.(2004). Water resources: Agricultural and environmental issues. BioScience,54(10), 909e918.

Postel, S. L., Daily, G. C., & Ehrlich, P. R. (1996). Human appropriation of renewablefresh water. Science, 271, 785e787.

Shevenell, L. (1996). Statewide potential evapotranspiration maps for Nevada (NevadaBureau of Mines and Geology Report 48). Reno, Nevada. Reno: Mackay School ofMines, University of Nevada, Retrieved from. http://www.nbmg.unr.edu/dox/r48.pdf.

Sierra Club. (2006). Exporting eastern Nevada groundwater to Las Vegas. Retrievedfrom. http://nevada.sierraclub.org/conservation/pipelines/index.html.

Sierra Club. (2008). Letter to Nevada State Engineer Tracy Taylor. Retrieved from. http://nevada.sierraclub.org/conservation/pipelines/SE-SpringVly/SCwrittencommentsSE1006.pdf.

Slovic, P. (1987). Perception of risk. Science, 236, 280e285.Slovic, P. (1992). Perception of risk: Reflections on the psychometric paradigm. In

S. Krimsky, & D. Golding (Eds.), Social theories of risk (pp. 117e152). Westport,CT: Praeger.

Slovic, P. (2000). The perception of risk. London: Earthscan Publications Ltd.Slovic, P., Fischhoff, B., & Lichtenstein, S. (1985). Characterizing perceived risk. In

R. W. Kates, C. Hohenemser, & J. X. Kasperson (Eds.), Perilous progress: Managingthe hazards of technology (pp. 91e125). Boulder, CO: Westview.

Slovic, P., Fischhoff, B., & Lichtenstein, S. (1986). The psychometric study of riskperception. In V. T. Covello, J. Menkes, & J. Mumpower (Eds.), Risk evaluation andmanagement (pp. 3e24). New York: Praeger.

Smith, K. (2001). Environmental hazards: Assessing risk and reducing disaster (3rded.). New York: Routledge.

Southern Nevada Water Authority (SNWA). (2009a). Water resource plan 09.Retrieved from. http://www.snwa.com/assets/pdf/wr_plan.pdf.

Southern Nevada Water Authority (SNWA). (2009b). Water use facts. Retrievedfrom. http://www.snwa.com/consv/goals_facts.html.

Southern Nevada Water Authority (SNWA). (2011). SNWA goals & facts. Retrievedfrom. http://www.snwa.com/consv/goals.html.

Swyngedouw, E. (2009). The political economy and political ecology of the hydro-social cycle. Journal of Contemporary Water Research & Education, 142, 56e60.

U.S. Census Bureau. (2011). State and county quick facts. Retrieved from. http://quickfacts.census.gov/qfd/states/32000.html.

Witcher, T. R. (2008). Water wise: In the middle of a drought, in the middle of thedesert . where do we stand? Las Vegas Weekly, Retrieved from. http://www.lasvegasweekly.com/news/2008/jul/16/water-wise/.