U.S. Drought Risk Wider than Previously Thought

New project research conducted as part of the Columbia Water Center’s “America’s Water Initiative” suggests that many more places in the United States are at risk of drought-induced water stress than is commonly thought, including dense metropolitan regions in the mid-Atlantic and Northeast such as New York City and Washington, D.C.

The findings are based on a new drought-induced water stress indicator developed by scientists from the Columbia Water Center, the City University of New York and Hohai University in China. Results were published in April’s Geophysical Research Letters, a journal of the American Geophysical Union. An earlier version of the research was published last year in a white paper from the Columbia Water Center.

According to the paper, most prior drought indices considered only precipitation and temperature averaged over a specific time window, without looking at demand, while water stress indices have mostly considered only available water supply, without considering potential drought severity and duration. The new index, by contrast, addresses both current water demand and climate scenarios.

Last year, the Columbia Water Center was awarded a three-year, $2,144,785 National Science Foundation grant to analyze the historical and present state of America’s water use, resources and infrastructure, and to develop a modeling environment to project and prepare for future water challenges. The grant was one of 26 awards, totaling $25 million that the NSF and the U.S.Department of Agriculture’s National Institute for Food and Agriculture awarded in 2014 as part of their joint Water Sustainability and Climate program.

The America’s Water Initiative aims to bring academic, industry, government and non-profit partners together to research all facets of America’s water challenges and develop solutions at a national scale. Researchers include Columbia University scientists and professors from a number of affiliated Earth Institute centers, including the Columbia Water Center, the Center for International Earth Science Information Network and the International Research Institute for Climate and Society, as well as from Columbia Law School and the Department of Economics. According to Columbia Water Center Director Upmanu Lall, a better understanding of the interplay between energy availability, price, agriculture, natural climate variability and climate change over time and in different regions will be crucial to managing water supply efficiently.

For their drought analysis, the researchers looked at 61 years of daily rainfall and temperature data from 1948 to 2009, interpolated to each of the 3,111 counties in the continental United States. They then computed county-by-county water demand, including agricultural water demand (based on average water demand per acre for 47 crops) as well as county-level industrial, livestock, mining, aquaculture, thermoelectric and domestic water withdrawals. The team pulled data and water requirement estimates from the National Agricultural Statistics Services, the Food and Agriculture Organization and the United State Geological Survey.

The team then developed two risk metrics to assess the impact of within-year dry periods (Normalized Deficit Index, or NDI) and over multiple years (Normalized Deficit Cumulated, or NDC). The indices are based on the “sequent peak algorithm,” which was originally developed for reservoirs to quantify water storage capacity needed to meet demand. An NDI or NDC of greater than 1 represents a location where the average rainfall is less than the average use.

Perhaps not surprisingly, Arizona and California are among the places that face the most severe water stress. As defined by the NDC, much of the water supply of California is based on imported water.

The difference between the NDI and NDC was significant. According to the paper, while most attention is paid to within-year drought and water stress, in the United States, multi-year, cumulative water stress is actually much more important. As maps from the study show, the multi-year, cumulative measurement suggests severe water stress over much of the Midwest agricultural belt, where much of the nation’s most heavily cropped land is, even where within-year drought risk is much lower.

In addition, metropolitan regions such as New York City and Washington face serious water stress at the county level. While these cities lie in wet climates, by restricting the definition of available supply to the rain that actually falls in the county, the study exposes the degree to which such areas can only meet their water needs by drawing on their rainy hinterlands, implicitly limiting the available water supply in those hinterlands and setting up potential conflicts between different users and uses.

For example, according to the America’s Water analysis, much of the West and Midwest are in a pattern of long-term, chronic deficit with regard to water use–even as some of those areas are net-exporters of high water-consuming crops. This suggests that by shifting water-intensive crop production to less water-stressed regions, overall national water stress could be substantially reduced. But in order to do so, policy makers will need a much more robust set of models that address the complex interplay of factors affecting water supply and use.