Dipping and weaving through the crisp mountain air, a fuzzy dot of orange and black flitters back and forth on a sunny fall morning on Mount Lemmon, just north of Tucson. Lofted by a soft breeze, it touches down on a marigold plant dotted with freshly opened, nectar-filled flowers.

"That's a Painted Lady," says Katy Prudic, not taking her eyes off the skittish butterfly. Prudic makes a swift swipe with her net, and a second later the butterfly has disappeared. Like a magician reaching for a white rabbit, Prudic's fingers probe the mesh, until they expose the two-inch insect, gently holding it by its wings.

"This species migrates to as far north as Canada," says Prudic, an assistant research professor in the University of Arizona's College of Agriculture and Life Sciences who studies wild pollinators, with a focus on butterflies. "Painted Ladies only come to southern Arizona to spend the winter — kind of like snowbirds."

Carefully prying apart the insect's wings, Prudic takes a photo with her phone, then lets the butterfly go. It flutters up and onward in search of the next flower, unencumbered by its brief moment in captivity. (In contrast to popular belief, touching the wings does not harm the insect.)

Together with a team of ecologists and conservation scientists, Prudic has big plans for the colorful creatures: Enlisting hikers and nature enthusiasts as volunteers, she wants to collect much-needed field data about the biodiversity, ecology and health of pollinators from remote backcountry locations.

"About 85 percent of the data collected in citizen science projects comes from major urban areas," she says. "Once you drive out for more than two hours, to the more remote parts, our data get really sketchy, especially in western North America.

"How many pollinators are there? What are they doing? What plants do they interact with, and at what times of the year? Those are the kinds of questions we need to answer if we want to inform continental policy in conservation and management. Unfortunately, we don't have any data for most of the continent currently."

According to Prudic, ecology is really about space, and how interactions between organisms, such as pollinators and plants, scale up over space, from a patch of flowers to a forest to a whole biome. And not only that. Depending on where ecological interactions among species are taking place, they may be very different because of local, unpredictable processes that may have higher-level effects that ecologists don't know about — a real-world example of the famous "butterfly effect."

For example, a certain assembly of species in a certain habitat might have different biological responses or ability to deal with change based on their past history. Over time, populations incur small changes in their DNA — too subtle to justify splitting them into separate species, but enough to cause them to respond differently to factors that vary for each location, such as competition, water availability, temperature, climate and predation.

"For those of us working in conservation, it's these big data gaps that pose a problem," Prudic says, "especially in remote areas. We may know how a certain butterfly species interacts with a certain flowering plant in the Tucson area, but we may not know how it interacts with that plant in Yosemite National Park. The differences may or may not matter, but you don't know until you actually start looking. That's why we need people to go out there and check if our assumptions and predictions are actually correct."

By teaming with the Montana-based nonprofit Adventure Scientists to engage a team of volunteer data collectors, Prudic and her collaborators hope to generate an impressive amount of new information. The resulting dataset will close large gaps in the current understanding of pollinators in remote areas, and will help inform conservation and management efforts. Currently focused on five Western states — Washington, California, Montana, Utah and Arizona — the researchers hope to eventually cover many landscapes spanning North America.

Because of technologies that weren't available even five years ago, Prudic says researchers can now tackle challenges that require massive datasets and powerful computing platforms. For the first time, ecologists have the means to scale up their research from the local to a much larger level.

"The question is, what do we need to do that?" Prudic says. "How do I go from four study sites to 4,000, 40,000, the whole continent? We have the technologies that can handle the data storage and collection of data, but how do you implement that over space and time?"

Adventure Scientists' volunteers will provide the ground-truthing necessary to improve the algorithms that Prudic and Ben Hickson, a geospatial specialist with the UA Libraries' Office of Digital Innovation & Stewardship, developed and applied to the field data.

"We provide the volunteers with the latitude and longitude coordinates along hiking trails to the prime butterfly habitats," Prudic says, "and then they will go out and actually see whether the butterflies we think that should be there at that time are indeed there."

Adventure Scientists specializes in getting scientists and conservationists the data they need from remote areas by recruiting, training and managing an international team of skilled volunteers from the outdoor community. They train the volunteers to collect research-grade data, and they communicate the results through their wide media reach. For the season, the organization seeks to enlist 100-150 volunteers who will survey 50 sites, including areas recently burned by wildfires. The data will be used by the U.S. Forest Service to inform land management plans. In the Tucson area, the project focuses on higher elevations, as those tend to have more butterfly diversity than the low desert.

In teams of two and equipped with butterfly net and smartphone, the volunteers hike to the same site three times over the course of three to four months and record the butterflies they encounter along their route. They take a picture, record the GPS coordinates where they found the animal, and take a note of any plants on which it landed and fed. They submit their photos and observations to iNaturalist, a free app, where they will be reviewed and identified by algorithms or the user community from around the world. Once confirmed, that data is forwarded to the Global Biodiversity Information Facility, which serves as an archive and clearinghouse for biodiversity information.

UA undergraduate students are involved in the project through an interdisciplinary collaboration among Prudic and Nirav Merchant, director of the UA's Data Science Institute, and Eric Lyons, associate professor in the UA's School of Plant Sciences. As part of a course on advanced cyberinfrastructure concepts, students used the data collected in butterfly pollinator biodiversity hotspots during the project's pilot phase last fall and developed workflows capable of extracting the information Prudic needs to expand the backcountry surveys across the continent. Using computer code developed by Jeff Oliver, a data science specialist with UA Libraries, the students were able to scale up analyses that were too computationally intensive to run on a normal desktop computer.

"The students were tasked with developing a cyberinfrastructure pipeline that would take tens of thousands of data points from iNaturalist, clean them up and put them into a form that is analyzable, so we can understand them," Prudic says. "It takes large data inputs from iNaturalist and filters and turns them into something that I can use to answer the questions I'm working on, and make data visualizations, mostly maps, with an interactive interface."

While not known to be important pollinators of crops, butterflies pollinate wild plants and serve as valuable indicators of an ecosystem's health and nutrient cycling.

"You could say butterflies, especially when still caterpillars, are nature's hot dogs," Prudic says. "They eat a tremendous amount of plant matter, so they serve as the major protein source for many animals, especially birds. People have noticed butterflies for hundreds, maybe thousands, of years, so there is a treasure trove of knowledge about them and their interactions."

Because they are so flashy and easy to recognize, people enjoy looking for them, according to the project's coordinator, Michelle Toshack of Adventure Scientists.

"Many of our volunteers tell us how excited they are about this opportunity to give back in ways they never could before," she says. "They may have been hiking in some of those areas many times, but only for their pure enjoyment. Now they get to explore in a whole new way, by collecting data for research that helps to preserve the places they love."