Skip to content Skip to navigation

Learning From The Crowd

Nov 7 2016 | Posted in: Research Highlight
By Stacey Solie

National Park visitation rates help managers make important decisions, and justify maintenance budgets by showing public use – yet traditional survey methods are expensive and time-consuming. A new, cheaper, easier method using Flickr data works just as well. Credit: Juan Arreguin / Unsplash

It’s not just millennials who turn to social media as a major source of information. Scientists are using it too.

But scientists are doing a lot more than just browsing social media for news. Increasingly, researchers are finding creative ways to harness a mostly untapped treasure trove of data. At NatCap, these methods continue to shed new light on people’s relationship with nature, from how it affects our moods, to pinpointing where we go to relax, why we decide to go there, and what it means for the economy.

Photos are like a trail of breadcrumbs

The photo-sharing site Flickr holds a wealth of information that NatCappers are using to help estimate visitation rates to public lands based on the number of photographs taken there and shared. These estimates shed light on how much people value those lands. In many cases the visitation data didn’t previously exist.

“Data on park visitation is fundamental to knowing how to manage it,” said Spencer Wood, who developed the method, and has used it as the base of NatCap’s recreation model, a component of the InVEST software package.

“Sometimes park managers wonder where they should build new infrastructure for people” Wood said. “And often managers want to know how ecological restoration or climate change might affect where people recreate.” Wood is part of the NatCap team and based in the College of the Environment at the University of Washington.

Wood’s idea uses a photo posted to Flickr from a particular place as a record of a single person having been there. Summed up over many photos uploaded by many people, he’s got a surprisingly reliable proxy for how many people visited those places. Wood and colleagues published a paper a few years ago introducing this idea and testing it with places where visitation rates are known.

 survey data

The National Park Services visitation survey data matches almost perfectly with visitation estimates obtained using Flickr photo data. (Click to enlarge)

A better understanding of U.S. National Park visitation

The method is potentially biased towards younger people or Westerners, who are seen as more likely to use Flickr than older generations. However, another recently-published study of U.S. National Park visitors shows that Flickr photo uploads are a remarkably accurate stand-in for visitation numbers, and that it captures older and international visitors as well.

Lead author Carrie Sessions, based at the University of Washington, compared the results she culled from Flickr to National Park Service-collected visitation data, and found a nearly perfect correlation. The Flickr method is cheaper and easier than the Park Service’s approach, which typically requires a person to stand at an entrance and count cars, Sessions said. Her research, published in the Journal of Environmental Management could help support funding for national parks, and if the method were adopted, it could save parks money.

“Their budget has been cut significantly,” Sessions said. “This is a time when the National Park Service needs to prove their worth to the federal government, and demonstrate that people actually value parks and are going to these places. And, they need to be more efficient in their spending.”

Valuing clean water in lakes

In Minnesota, NatCap’s Bonnie Keeler has also used the Flickr method to shed light on how much people value clean water by analyzing how far people will go to visit a clean lake. Her research, which used data about people’s home cities embedded in their Flickr profile, showed people will travel up to an hour further to visit a clean lake, bypassing dirtier water bodies along the way.

“We assume that tourists and other users prefer higher quality sites and pristine systems,” Keeler said, in a Discover article about the research. “But it’s tough to prove because surveying people about their vacations is expensive and hard to do on a large scale.”

“Clean water is also worth more than what we reveal through our recreational behavior,” Keeler said in an essay in Open Rivers magazine. But the research at least begins to show that clean water brings money into the economy, which could be used to support arguments that investing money in cleaning up polluted lakes will pay off for communities.

Keeler continued, “An estimated 40 percent of lakes and rivers in Minnesota are classified as “impaired” and unfit for…basic human uses. Efforts to restore watersheds are expensive and public dollars to support those investments are limited. Quantifying the value of clean lakes and rivers is critical in making the case that the potential benefits of clean water protection or restoration exceed monetary costs.”

Outdoor recreation in Vermont

Another study led by Laura Sonter from Vermont’s Gund Institute, took the Flickr research a step further by estimating the tourism revenue that parks and other public lands bring into the state.

The analysis, published in PLoS One showed that parks – and other conserved areas –contributed $1.8 billion to Vermont’s tourism industry from 2007-2014. The study is the first to use social media to assess the value of protected lands for outdoor recreation across an entire U.S. state. It’s also the first comprehensive tally of outdoor recreation on conserved lands in Vermont.

The study also revealed key differences in the recreation preferences of Vermonters and out-of-state tourists. For example, out-of-state visitors gravitated to areas with more trees, and avoided places with a lot of cutting. But forest loss had less of an effect on in-state tourists, who preferred locations with easy access to clean water and swimming.

Where to go from here

Wood says there are several projects underway to expand on this knowledge of how people are using public lands, including one with the U.S. Forest Service, in Washington state’s Mt Baker Snoqualmie Forest.

NatCap is helping them measure trail use by hikers. The Forest Service needs to know which trails are popular in order to decide where to invest money in maintenance, restoration, and erosion control and to predict the popularity of future sites.

“We’re being innovative, we’re solving problems that have been lingering in science for decades,” Wood said. “Social media allows us to do analytics that just aren’t possible without these types and volumes of data on people’s behaviors and preferences. We’re learning all sorts of things about outdoor recreation that we didn’t know before.”

Stacey Solie is the Communications Lead at The Natural Capital Project.


Research Area: