The digital revolution inspires a new research direction in the structural diversity of ecosystems

A special issue of the magazine Frontiers in ecology and environment lays the foundation for pursuing structural diversity as a new research direction in ecology. The issue also describes the digital data collection methods that enable the new research direction and the applications of the work in different ecosystems.

“Structural diversity means thinking about what elements occupy a space and how they were arranged in the space,” said Songlin Fei, editor-in-chief of the special issue, professor of forestry and natural resources and Dean’s chair of remote sensing at Purdue. “The hope is that we provide a framework that can be applied regardless of the system you’re working in, from terrestrial to aquatic.”

As Fei and three co-editors write in their review, the special issue’s papers provide “a new framework for structural diversity, new applications in ecological theory, and case studies.”

Co-editors were Purdue’s Brady Hardiman, Associate Professor of Urban Ecology in Purdue’s Department of Forestry and Natural Resources; Elizabeth LaRue, assistant professor of biological sciences at the University of Texas at El Paso; and Kyla Dahlin, associate professor of geography, environmental and spatial sciences at Michigan State University.

Six of the special issue’s seven lead authors are early-career scientists developing applications for the 3D technologies that could lead to new ecological theories. These technologies include light detection and ranging (lidar) and data sensors mounted on drones and satellites.

“The adoption of these emerging digital tools and technologies will enable the next generation of ecologists to operate a fleet of sensors to measure ecosystems and swim freely in the resulting ocean of data,” the editors wrote.

Such methods form the basis of Purdue’s new Center for Digital Forestry, which Fei directs. One of the five strategic investments in Purdue’s Next Moves, the center uses digital technology and multidisciplinary expertise to measure, monitor and manage urban and rural forests to maximize social, economic and environmental benefits.

“In the past, as scientists, we measured the Earth as a flat entity,” said LaRue, a former postdoctoral researcher supervised by Fei and Hardiman. “Part of that is because we didn’t have good technology to measure 3D aspects of the planet.”

These aspects include elevation changes and fine-scale features such as tree branching patterns. Previously, researchers had to make such measurements by hand.

“Technology is advancing rapidly. We need to catch up with the science and theory enabled by these 3D technologies,” she said.

The special issue notes that while important work has already begun on forest management in ecosystem types such as wetlands, grasslands and marine ecosystems, more needs to be done.

“Our knowledge of the structural diversity in different ecosystem types is still quite limited,” said LaRue.

Traditionally, scientists have attempted to measure biodiversity by counting species and assessing their genetic diversity.

“These existing measures come back to this fundamental question: How much of the available ecological space has been occupied by different organisms?” Hardiman said. “The more ecological space that was occupied by different species, the more stable the system could be, because the absence of a species would not lead to the collapse of the system.”

But with new 3D digital technologies, researchers can now quickly determine the stratified arrangement of species within an environment, along with their size and numbers. Such skills benefit both land managers and researchers. Managers can now often collect higher quality data to support their decision-making much faster and at less cost. Sometimes they can just use a cellphone app to take measurements that previously required a tape measure.

The editors and authors of the special issue highlight four challenges that researchers must address in order to realize the full potential of such digital advances in ecology.

The first challenge for ecologists and environmental scientists is to collaborate more fully with colleagues from other disciplines. The required expertise ranges from aeronautics, engineering and computer science to graphic design, information science and the social sciences.

The second challenge is to use supercomputers, cloud computing, machine learning and artificial intelligence to process the vast 3D data sets that digital technology is generating today.

“A lot of the data we work with is publicly available and available,” Fei said. But researchers sometimes lack the expertise to take advantage of it. “They don’t have the computing power or the right tools to deal with it,” he said.

The third challenge is to pursue new approaches to better assess the hundreds of variables in ecosystem structure that 3D datasets often contain today. Rather than relying on traditional hypothesis testing, the editorial team recommended that researchers adopt data-driven approaches or combine both.

Finally, the editors stressed the critical importance of training the next generation of ecologists in digital technology.

“New data-centric skills such as collection, visualization, analysis and management of large datasets must become essential parts of ecological education,” they wrote.