New tool will help conservation planners strategize around development risks

Don't it always seem to go     

That you don't know what you've got

'Till it's gone

You can probably hear the next lyric of Joni Mitchell’s song Big Yellow Taxi in your head, and if you work in the field of conservation, you might take it to heart.

But thanks to a new tool developed by the Designing Sustainable Landscapes (DSL) Project at the University of Massachusetts Amherst (UMass), conservation practitioners in the Northeast can start to change their tunes. 

“Our ‘Sprawl’ model simulates urban growth 70 years into the future, producing maps that show likely patterns of future development and probability of development across the entire 13-state region,” explained Ethan Plunkett, a Research Associate at UMass. “These maps are directly useful to conservation planners who might wish to strategize around the risk of development.”

Sprawl is just one component of a suite of models developed by DSL to assess ecological value across the region and predict how the landscape changes that have been rolled into the Nature’s Network conservation design.

“Looking at the big picture, the idea behind landscape conservation design is to bring together information at large spatial and temporal scales to inform local conservations decisions,” said Plunkett, pointing out that like climate, development is a major driver of environmental change on the landscape scale. But while urban growth is probably easier for many people to envision than say, the greenhouse effect, until now, it hadn’t been depicted as well at a regional scale by models.

“There are a lot of existing models for climate change in the Northeast, but for urban development, there wasn’t very much available to draw from that covered the entire region at a fine resolution,” Plunkett said. “That was the need we wanted to address in developing Sprawl.”

To develop a robust approach for modeling development across the entire Northeast region, the team looked to various other urban growth models that have been developed in the past, as well as to the past itself. “Mostly, we tried to let historical patterns drive the model, rather than try to model the processes that drive development themselves,” he explained.

Instead of analyzing real estate trends, they analyzed land-cover maps developed by National Oceanic and Atmospheric Administration to show landscape characteristics like development intensity, open water intensity, tree canopy, and impervious surface.

By comparing land-cover images from different time periods, they captured information about changes in land-cover over time, and trained the model to apply historical patterns to today’s landscape in bite-sized chunks. The result is an educated guess about where development will happen based on where it has happened in the past.

They also took into account geophysical factors like slope, and proximity to influencing features like open water, existing secured lands, roads, and urban centers. The resolution of the model is 30 meters -- about a quarter of an acre -- so practitioners can use the tool at the local level to look for important areas that might be vulnerable to development.

To complement the probability of development map, the DSL team produced a set of additional maps that combine results from Sprawl with two other models: the Index of Ecological Integrity, and Regional and Local Vulnerability. The resulting maps can help practitioners look for areas of high value for ecological integrity or connectivity that are likely to be negatively impacted by urban growth.

Although Plunkett cautions that the Sprawl model is not a crystal ball -- the results should be considered potential scenarios, not prophesies -- it gives planners working at any scale a valuable tool that can help support sustainable growth of human and natural communities across our region.

“We know that over the next 70 years the region will be drastically altered by development,” he said. “But we hope that by understanding how that development might play out we will be able to mitigate the worst of its impacts.”

In the coming decades, we’re are probably going to need more parking lots, but with better information about where that need will arise, we can plan to leave room for paradise too. 

Plunkett will lead a Science Seminar on the Sprawl model on Thursday, December 14th. Learn more.