Data Fusion to FEWSION: Mapping the U.S. food, energy, water nexus

Food, energy and water intersect at many levels; food and water are essential to human life, and both water and energy are essential to producing food at the scale required to sustain the world’s growing population.

These crucial dependencies are usually understood in simplistic terms by policy-makers and the general public, but researchers have identified the need to provide information which better explains food, energy and water supply chains, and demonstrates the vulnerabilities and dependencies of specific geographic areas. Not all cities, states or, in some cases, countries, can produce their own energy and agriculture, and they are therefore reliant upon resources from other areas to support their communities.

In the United States, these linkages are the focus of the FEWSION project, which aims to make big data visualisation of the food, energy, water system available to the public. The FEWSION project is funded through the INFEWS program, which is sponsored by the U.S. National Science Foundation and the U.S. Department of Agriculture.

Professor Richard Rushforth, Associate Research Professor with the Northern Arizona University School of Informatics, Computing and Cyber Systems, is the lead research scientist for the FEWSION project. He spoke about the intent, achievements and aspirations of the initiative at the UNSW Global Water Institute’s latest seminar on Thursday 24 May.

The FEWSION project is funded by the United States National Science Foundation and hosted by Northern Arizona University (NAU), which is based in the city of Flagstaff. Professor Rushforth used Flagstaff as a case study during his May presentation, pointing out that while the city has a cool climate and a snow pack, it has no agriculture, limited water resources and little to no local energy production. This means that the city is highly dependent on food from Phoenix, California, Washington or the Midwest; energy from southern Arizona and New Mexico; and water from other parts of Arizona as well as California and the Southwest to produce goods.

Prof Rushforth explained that Flagstaff is not diversifying its water sources – concentrating its water supply chain on the south-west of the United States which is getting hotter and dryer.  This leaves the city vulnerable to the impacts of climate change and natural disasters, whether they occur at home or in the areas where the food, energy and water is coming from.

While building and evaluating Flagstaff’s water footprint, Rushforth and his team decided that they wanted to understand potential hotspots for risk within the city’s water footprint. To do this required a model that recognised the systems-based nature of the problem, as well as the fact that multiple cities and regions are simultaneously dependent on the same virtual water resources.

 “We realised we needed to look at the United States as a whole as so many geographic areas are all connected,” said Prof Rushforth.

“We had to take a step back and approach this problem from a more holistic, systems-level perspective”.

The FEWSION project is described as a ‘mesoscale big data effort for the United States food, energy and water system’. In essence, it is a project that uses big data to define food, energy and water relationships at the county scale and make them easy for all to access and understand.

“We want to enable communities to understand where their food, energy and water come from, and how sustainable and resilient their food, energy and resources may or may not be,” said Professor Rushforth.

We want to enable communities to understand where their food, energy and water come from, and how sustainable and resilient their food, energy and resources may or may not be

Prof Richard Rushforth, Associate Research Professor, Northern Arizona University

The FEWSION visualisation product was born of the notion that data products are notoriously messy, and it was recognised that an approach was needed to visualise data sets so that they made sense to people not immersed in the data.

“We have a complex data product that is difficult for researchers to deal with sometimes, and we are wanting to make it accessible to the general public,” said Rushforth.

FEWSION 1.0 uses a unique dataset, made up of raw data from a wide range of publicly-available sources such as census data, labour statistics, water withdrawals, agriculture statistics, commodity flows and freight, which is then analysed and downscaled into county-scale data on food, energy and water flows for the years 2010-2012. The tool, which will be released to the public in late September or early October this year, will enable a user to build their own scenario by selecting their boundary (city or county), type of flow (inflow or outflow), regions of interest, commodity (e.g. power, food, etc), and metric (e.g. tonnages of food, value in dollars, etc). FEWSION will then map these aspects and enable them to be downloaded as summary graphs, data tables or graphics.

According to Prof Rushforth, there are some challenges in working with such data, including inconsistencies with the availability of data at the county level and differences with timescales.

“We must develop methods to relate, downscale, and upscale data to produce a new data layer at the right geographic scale,” said Prof Rushforth.

FEWSION 2.0 will be released in 2019, and will interpret data from earlier than 2010, allow for future projections, and have the capacity to compare two geographic regions simultaneously.

“We have much of this data at hand but we’re in the constant process of tweaking and adding new data into the system,” said Prof Rushforth.

“There has been a lot of work in the back end to produce this so that when we do make it available to the public, there is no lag and you’ll be able to understand food, energy and water linkages in the United States in near real-time.”

Because metropolitan areas or cities are sometimes part of multiple counties, the next step in the FEWSION project is to collect data which delves below the county level. This is being done through a data-driven downscaling initiative by NAU and West Virginia University, and through a citizen-science downscaling project which involves interviewing stakeholders to collect additional data and to learn more about how they engage with their local food, energy and water system.

“The citizen science process is a way of ground-proofing and validating county level data produced through the FEWSION project while also providing outreach,” said Prof Rushforth.

While it was anticipated that FEWSION would be of most interest to policy-makers and the private sector, emergency management groups have also expressed a keen interest in the project for its ability to predict how a natural disaster in a specific part of the United States may affect the availability of food, water and energy in other areas.

NAU has also been working to connect ecosystem impacts to the FEWSION project – looking at aspects such as how crops contribute to stress on river ecosystems. Eventually, it is hoped that the FEWSION project will become the first complete footprint database for every city in the United States—covering food, energy, water, greenhouse gases, industrial goods, air pollutants, nutrient flows, primary productivity and policy.