Land-use, temperature, and single-family residential water use patterns in Portland, Oregon and Phoenix, Arizona

Abstract

Adaptation to climate change requires urban water providers to develop a complex understanding of how temperature affects water use patterns. We used a geographic information system and statistical analysis to compare the spatial relationships among single-family residential water use patterns, land use characteristics, and temperature in Portland, Oregon and Phoenix, Arizona. We developed mean water use patterns at the census block group level using data from 2002 to 2009 in Portland and from 2000 to 2008 in Phoenix. These mean values were used to estimate the localized temperature sensitivity of water use in each census block group through an ordinary least squares regression with summer average air temperature. Taking the slopes of regression estimates as our dependent variable, we examined spatial relationships among temperature-sensitive water use patterns, housing density, impervious surfaces, low vegetation, and tree canopy extent. Temperature sensitive water use was found to be positively correlated with low vegetation and negatively correlated with impervious surfaces in both cities. Tree canopy coverage tends to increase with sensitivity in Portland, while the reverse relationship is found for Phoenix. Regression analysis indicates that building density explained the most variation in the dependent variable in Portland whereas, in Phoenix, the strongest correlations related to vegetation patterns. A comparative approach highlights the complex, localized correlations that exist among local climate regimes, urban landscapes, and water use patterns. Census block group-level water use analyses equips water providers with detailed information on the sensitivity of local water use to temperature variation, which could prove valuable to developing a viable municipal climate change adaptation strategy.