Pamela Cruz. Peninsula 360 Press [P360P].
Researchers at Stanford University analyzed water consumption with the help of macro-data and Artificial Intelligence tools, which could change urbanization strategy and provide a better understanding of water use and supply in cities.
Notably, the research, published in Environmental Research Letters, is the first to demonstrate how real estate platforms can be used to obtain water use data for city planning, drought management and sustainability.
The report uses data from Zillow and other real estate websites to compile information on single-family homes, including lot size, home value and number of bedrooms in Redwood City, California, a fast-growing and economically diverse city with various home styles, lots and neighborhoods.
Then, using demographic information from the city's Census Bureau, they analyzed factors including average household size and income, along with the percentage occupied by renters, non-family, college-educated, and seniors.
By combining the data and applying machine learning methods, the researchers were able to identify five clusters of communities, which they also compared with billing data from the city's public works department, to identify water use trends, seasonal patterns, and conservation rates during California's 2014-2017 drought.
"Evolving development patterns may be the key to our success in becoming wiser in our use of water and building long-term water security," said the study's lead author, Newsha Ajami, director of urban water policy for Stanford's Water in the West program.
Researcher Kim Quesnel, for her part, said that with this method "we were able to develop more accurate community groupings, beyond simply grouping clients based on income and other socioeconomic qualities," resulting in some unexpected findings.
Contrary to previous studies, the researchers found that the two lowest income groups scored average in water use, despite having a greater number of people living in each household.
The middle-income group had high outdoor but low indoor water use, indicating the use of efficient appliances, such as low-flow faucets and toilets.
While, of the two groups with the highest incomes, that of younger residents with smaller lots, as well as newer homes in dense, compact developments, had the lowest water use citywide.
Meanwhile, the other high-income group, consisting of older homes built on larger lots and with fewer people, turned out to be the largest consumers of water.
"The finding runs counter to most previous research linking income and water use, and suggests that changing the way communities are built and developed. It may also change water use patterns, even for the wealthiest customers," the report adds.
Thus, this research sets the stage for the integration of big data and artificial intelligence into urban planning, providing more accurate usage expectations for different community configurations.