New Research Shows Water Use Impacted by the Shape of Our Cities

aerial view of suburban homes

For Center for Geospatial Analytics doctoral student Georgina Sanchez (Dept. of Forestry and Environmental Resources), studying urban water use is both a personal and professional passion. “From availability to use––I’ve always been interested in water,” she says. For her dissertation at NC State, she is exploring water demand in the Southeast, and how to “create useful tools that facilitate decision making.” Her latest analyses, published in Water Resources Research, show that the shape of urban development at small scales influences water use more than the non-structural variables, like education and precipitation, commonly used in past studies. Essentially, her research finds, “Developed landscapes that promote simple, cohesive spatial patterns show potential for more efficient use of water.”

Why, you may ask, is shape so influential?

“We need to understand one simple connection,” Georgina says. “People make de facto water use decisions as they make land use decisions.” Compact, simply shaped patches of development represent landscapes in which people make different resource-use decisions than in sprawling, irregularly shaped patches. “We have to think about the whole picture,” Georgina says: “Behavior, process, and pattern.”

In simply shaped urban areas, for example, there are fewer lawns, amenities tend to be closer together, and the land footprint per person is smaller. Conversely, in more complicated-shaped patches of development, such as suburban housing, residents tend to use more water (e.g., for their lawns and gardens, backyard pools, etc.), as well as more energy and gas for their cars. “Shape, in relation to the way we design urban spaces, affects not only how we use water, but also it has been shown to influence energy consumption, carbon footprint and physical activity,” Georgina says.

suburban housing vs. urban grid of buildings
Water use depends in part on the spatial pattern of developments, due to human decision-making. Developments with more complicated shapes (left) have higher water demand, due to the choices of people who live there. Developments with more compact, simple shapes (right) exhibit lower water demand.

For her study, Georgina examined several thousand census tracts in more than 100 counties across North and South Carolina and built statistical models to predict their water use using landscape, socio-economic and environmental variables. The best performing model included all three categories of variable, but a metric called Shape Index proved to be most important. Shape Index measures the complexity of developed land patches on a landscape; patches with the most simple shapes have low values and patches with more complicated shapes have higher values. “Overall,” Georgina and her co-authors say, “increases in the geometric complexity of spatial patterns of development (measured by the Shape Index) was associated with higher water use of both domestic water use and total water use.”

Her study is the first to show this across a large geographic region. What, then, are the implications of these findings?

“This paper serves as proof of concept of a method,” Georgina says. “Our results indicate that water consumption rates in developed areas across the Carolinas are sensitive to patterns of urbanization. Now, can we use this information to guide how future development choices might play a role in local and regional water demand?” Her ongoing work now “couples water use modeling, land change modeling and advanced geospatial analytics” to forecast urban growth and the associated total water use nearly fifty years into the future. “This is quite novel,” she says, “projecting water demand based on spatial patterns of urbanization and anticipated changes in climate.” A status quo simulation she is running projects urban development based on observed growth trends over the past 20 years, while a “WaterSmart” simulation optimizes water efficiency, including through incentives that promote “infill” near existing development rather than sprawl. Her presentation of a poster based on this research won third place at the 2018 College of Natural Resources Graduate Research Symposium.

For Georgina, the accolades are secondary to making a difference with her research. “All of my work is behind the computer, modeling, but I try as much as I can to connect with our partners and stakeholders in this project––municipalities, water-suppliers, state regulatory agencies, to name some of them,” she says. Her collaborators include researchers from the U.S. Geological Survey’s South Atlantic Water Science Center, Department of the Interior Southeast Climate Science Center and U.S. Forest Service. Georgina is also using only publicly available datasets and open source software for this research, ensuring that it is replicable and scalable to the entire nation. “At the end of the day,” she says, “we want to create a product that people will use.”


Georgina is co-advised by Ross Meentemeyer, director of the Center for Geospatial Analytics, and Jordan Smith, director of the Institute of Outdoor Recreation and Tourism at Utah State University.

NSF Grant Funds Smart City Research to Tackle Regional Problems with Serious Games

Interacting with a digital map

Smart City researchers at North Carolina State University are uniting leaders from government, industry, academia and non-profit groups across the Triangle to design an online game called TomorrowNow that will crowdsource perspectives on stormwater management in the region and inform policy decisions. The work is supported by a four-year, $499,847 grant from the National Science Foundation through its Smart and Connected Communities initiative.

“Our aim is to transform engagement of urban and rural residents and policymakers in region-wide problems like stormwater management,” says Ross Meentemeyer, director of NC State’s Center for Geospatial Analytics, professor in the College of Natural Resources and leader of the project. “Solutions to large-scale problems in rapidly urbanizing regions require input from a variety of stakeholders and coordination between nearby cities and towns. We expect that an online game, as a collaborative platform for knowledge-sharing, could enhance smart connections among communities and break down barriers to finding optimum solutions.” The game will use innovative geospatial analytics to enable communities connected by shared water networks––and their associated problems of flooding and pollution––to collaboratively design management alternatives. Publicly available, the game would allow anyone to contribute.

The first step to producing the game is to build a network of regional stakeholders that can provide input on the idea and frame its development. Meentemeyer and Helena Mitasova, professor of marine, earth and atmospheric sciences at NC State, will collaborate with Kevin Foy of North Carolina Central University, Emily Bernhard of Duke University and Todd BenDor of the University of North Carolina at Chapel Hill to recruit representative groups from across the Triangle to planning workshops, create spatial models of urbanization and stormwater flow, and develop technical specifications for the game.

The research team will co-host symposia, in-person discussions and online forums to identify obstacles that have previously hindered decision-making efforts regarding stormwater. These dialogues will also define the project’s research agenda and the design of a game that is functional, engaging, realistic and easy to use. This model for engagement in interdisciplinary citizen science will also establish an innovative platform for further Smart City initiatives between area universities, local governments and the public.

“This unprecedented collaboration draws on the strengths of four powerhouse universities in the region,” Meentemeyer says, “and ultimately, we expect that our approach will serve as a model to empower smart and connected communities elsewhere.”


For more information, see the National Science Foundation press release.

Getting in Touch with the Designer in All of Us

Payam Tabrizian looks on as another researcher experiments with Tangible Landscape

Payam Tabrizian is a landscape architect and Ph.D. student in Design, but he firmly believes that designing landscapes shouldn’t be limited to the professionals. “All people are designers by nature,” he says, “and they should be designers”––they need only the chance to shine. At the Center for Geospatial Analytics, Payam keeps this motto in mind as he develops high-tech tools that bridge real and virtual worlds. His research also bridges disciplinary gaps between design and the geospatial and cognitive sciences. His central aim, he explains, is “to unleash people’s creativity.”

Now, he and his colleagues Anna Petrasova and Brendan Harmon are one step closer to liberating the imaginations of professional and amateur designers alike, as they bring landscape design into the fast lane and make it something you can literally put your hands on.

Traditionally, professional landscape design plans take months to develop. From the original concept sketch to final approval, these plans require a range of software expertise to produce, and they pass through the hands of many people who have the skillsets to help visualize and interpret the initial ideas, as well as assess how the landscape might function ecologically. Thanks to Payam and his colleagues, now the entire process of hand-sketching, 3D modeling, and assessing landscape pattern takes just minutes, and anyone can do it.

Their solution is a high-tech but intuitive one: a person simply arranges cloth pieces on the surface of a landscape model to represent different vegetation types, and as that arrangement is scanned and fed into a computer, the user receives immediate feedback about ecological indices, such as landscape complexity and biodiversity, and can see their creation rendered as a photorealistic 3D model in a virtual reality display. What used to take months can now take minutes, and designers can refine their plans over and over in response to the near-real-time feedback.

Payam’s approach uses the Center for Geospatial Analytics’ celebrated Tangible Landscape system, developed by Dr. Helena Mitasova and her Ph.D. students, Anna Petrasova, Vaclav Petras, and Brendan Harmon. Tangible Landscape runs on open source GRASS GIS software coupled with a scanner and projector; Payam has connected the system with the open source 3D modeling and rendering software Blender, enabling traditional bird’s-eye views of landscapes to be visualized at ground-level, from the human perspective, in virtual reality.

Any person regardless of their expertise can sit down with a Tangible Landscape setup, equipped with a set of colorful cloth pieces, and arrange those pieces to design a landscape from scratch. And Payam believes it’s important that they do. Everyone has a sense of beauty, balance, and composition, he explains, and all they need to design landscapes are the tools that can help bring their visions to life and help them understand what their designs mean for the surrounding ecosystem.

The new technology produces a dynamic dashboard that reports the size, shape, number, and diversity of vegetation patches the designer chooses—variables important to consider from an ecological perspective. “This idea of landscape metrics is intimidating for landscape ecologists, let alone designers,” Payam says, but the dashboard breaks down barriers for considering how designed landscapes could affect biodiversity, habitat suitability, and ecosystem function.

landscape design with felt and VRPayam and his collaborators unveiled the new technology at the April 2017 annual meeting of the US regional association of the International Association for Landscape Ecology (US-IALE) in Baltimore, MD. One of the many users of the demo was Joan Iverson Nassauer from the University of Michigan, co-editor-in-chief of Landscape & Urban Planning and a distinguished scholar in the fields of landscape architecture and landscape ecology. “It was great fun to use this technology,” she told Payam by email after the conference, “and the results of each trial could be very informative for designers.”

So, what’s next? Payam and his team will soon invite faculty from NC State’s Department of Landscape Architecture to experiment with the prototype and offer their feedback. Equipped with tracing paper, markers, and a variety of colored felt pieces, these faculty will have the enviable task of designing landscapes in a way that is both fast and fun, and which brings them immediately into the places they create. Perhaps soon, all of our inner designers will be able to do the same.