Effects of local drought condition on public opinions about water supply and future climate change
Jason M. Evans1 & Jon Calabria2 & Tatiana Borisova3 &
Diane E. Boellstorf4 & Nicki Sochacka5 &
Michael D. Smolen6 & Robert L. Mahler7 &
L. Mark Risse8
Received: 12 February 2014 /Accepted: 28 April 2015 # Springer Science+Business Media Dordrecht 2015
Abstract A growing body of research indicates that opinions about long-term climate change and other natural resource issues can be significantly affected by current weather conditions (e.g., outside air temperature) and other highly contingent environmental cues. Although increased severity and frequency of droughts is regarded as a likely consequence of anthropogenic climate change, little previous research has attempted to relate the experience of drought with public attitudes about water supply or water-related climate change issues. For this study, a large set (n=3,163) of public survey data collected across nine states of the
Electronic supplementary material The online version of this article (doi:10.1007/s10584-015-1425-z) contains supplementary material, which is available to authorized users. * Jason M. Evans email@example.com 1 Department of Environmental Science and Studies, Stetson University, 421 N. Woodland Blvd., Unit 8401, DeLand, FL 32723, USA 2 College of Environment and Design, University of Georgia, 285 S. Jackson St., Athens, GA 30602,
USA 3 Food and Resource Economics Department, University of Florida, 1097 MCCB, PO Box 110240
IFAS, Gainesville, FL 32611, USA 4 Department of Soil and Crop Sciences, Texas A&M University, 349B Heep Center, College Station,
TX 77843, USA 5 College of Engineering, University of Georgia, 206 Driftmier Center (Annex), Athens, GA 30602,
USA 6 Department of Biosystems and Agricultural Engineering, Oklahoma State University, 218 Ag Hall,
Stillwater, OK 74074, USA 7 College of Agriculture and Life Science, University of Idaho, PO Box 442339, Moscow, ID 83844,
USA 8 Marine Extension Service, University of Georgia, 1180 E. Broad St., Athens, GA 30602, USA southern United States was spatio-temporally linked with records of short-term (~12 weeks) and long-term (~5 years) drought condition at the level of each respondent’s zip code.
Multivariate ordinal logistic regression models that included numerous other independent variables (environmental ideology, age, gender, education, community size, residency duration, and local annual precipitation) indicated highly significant interactions with long-term drought condition, but showed no significant effect from short-term drought condition. Conversely, attitudes about water-related climate change showed highly significant interactions with shortterm drought, with weaker to no effects from long-term drought. While the finding of significant effects from short-term drought condition on opinions about future drought is broadly consistent with previous public opinion research on climate change, the finding of water supply attitudes being more responsive to longer term drought condition is, to our knowledge, a novel result. This study more generally demonstrates the methodological feasibility and applied importance of accounting for local drought condition when public opinion information is used to evaluate outreach programs for water conservation and climate change. 1 Introduction
One of the most serious expected consequences of climate change is increases in the frequency and severity of droughts across many regional and local areas. Communication of future water supply risks associated with climate change and promotion of increased water conservation as a climate adaptation strategy is therefore a standard components of many water outreach and planning programs (e.g., Bjorkland and Pringle 2001; Cohen et al. 2006; Rosenzweig et al. 2007). Because such outreach programs are specifically intended to educate the public in ways that change attitudes and, ultimately, behaviors, they often are assessed through broad-based survey instruments designed to gauge longitudinal effectiveness of program activities and outreach materials on public opinions (see e.g., Shepard 2002; Lorenzoni and Pidgeon 2006;
Jackson-Smith and McEvoy 2011).
Qualitative observations generally support the intuitive notion that the onset of severe drought conditions tends to increase public interest and receptivity to water conservation programs (Jensen 1996; Delorme et al. 2003; Casagrande et al. 2007). Although not directly focused on drought experience, a more recent body of psychological literature more specifically indicates that contingent environment factors such as current outdoor temperature and condition of indoor vegetation (Gueguen 2012) can have significant impacts on public opinions about climate change (Joireman et al. 2010; Li et al 2011; Weber and Stern 2011;
Akerlof et al. 2013). However, surprisingly little previous quantitative research has attempted to link spatio-temporally specific measures of drought condition and severity to public attitudes about water resources or climate change phenomena.
Previous studies that have examined effects of precipitation conditions on public opinions include the work of Diggs (1991), Trumbo et al. (1999), and, more recently, Safi et al. (2012).
Workingwith Great Plains farmers, Diggs (1991) found that certainty about the existence of longterm climatic change was stronger among farmers in North Dakota, which had experienced severe drought conditions throughout much of the 1980s, as compared to farmers in an area of northeastern Colorado that had been generally wet over this same time period. Trumbo et al. (1999) conducted a post hoc analysis of water conservation attitudes collected from two surveys in Reno, NV, and found that public attitudes were significantly more favorable to water conservation programs during a drought period as compared to a comparatively wet period.
Work by Safi et al. (2012) with Nevada farmers found that higher livelihood dependence on agriculture and higher education both were associated with an increased concern for climate change, but that differences in relative local water stress was not a significant predictor of climate change concern. However, we note that the water stress metric used by Safi et al. (2012) did not explicitly include direct measures of local drought condition, but was instead was based on calculations of water availability, water demand, and population at a zip code level.