Audubon’s peer-reviewed research predicts that climate change will threaten nearly half of all bird species in North America over the next century. With the climate report’s publication, Audubon continued its long tradition of buttressing advocacy with hard science. The report drew on climate models and observations from thousands of citizen-scientists to conclude that 314 species, including well known species like the bald eagle, brown pelican and common loon, were severely at risk due to global warming. Ten states were at risk of losing their official state bird.
Despite the project’s success, Audubon’s scientists knew they needed to harness the power of big data to take their analysis to the next level. “We needed a way to scale our quantitative analyses to incorporate the growing volume of citizen science data into our models and to make an even bigger impact with our conservation initiatives,” said Chad Wilsey, director of conservation science at Audubon.
In early 2017, Audubon started working with Domino Data Lab to amplify its data science work. Starting with the 35 North American grassland bird species, Audubon’s researchers are analyzing whether climate change will expand or contract the birds’ habitats. Domino’s platform has enabled the team to quickly and simply build complex climate change models spanning the entire North American land mass at high spatial resolution.
“Domino has allowed our quantitative researchers to scale-up our analysis faster and with less effort. We can generate hundreds of models, run diagnostics and select the best-performing model in relatively little time. This is critical, as the models they’re building incorporate huge areas of land, at high spatial resolution and need lots of processing time,” Wilsey said.
Whereas Wilsey’s team could previously only use one computer to test large-scale climate change scenarios, they can now run experiments on up to 15 compute environments in parallel. “We’ve been able to crunch years of work into months,” he said.
Thanks to the boost in productivity, Domino has enabled Audubon to rely on its existing in-house researchers instead of hiring new data scientists. The platform, which is language-agnostic, has also allowed Audubon to integrate its existing quantitative research tools, which primarily rely on the R programming language.
A major benefit of the platform is version control across an entire project. This allows the user to change inputs and track the impact on both intermediate models and their predictions. This efficiently provides insight that informs the team’s decisions. Domino’s sophisticated repository allows the team to review all steps leading to a given result. The team has worked closely with Domino’s customer support team to make sure they’re maximizing the platform’s tools.
The initial phase of research, which Audubon predicts will confirm that climate change is threatening grassland birds, will conclude in 2017. The peer-reviewed findings are slated to be published in 2018.
“Thanks to the horsepower of Domino, this will be the most in-depth research ever conducted on the effects of climate change on grassland birds,” Wilsey said. One of the project’s unique contributions will be its ability to display both broad-scale and local data.
Next, Audubon will use the Domino platform to build models for more than 500 North American bird species. The analysis will help Audubon identify where it should focus its most urgent conservation efforts. Ultimately, the goal is for the analysis to inspire vigorous, focused action around climate change. “We’re using this to direct on-the-ground conservation and to inform advocacy for birds and to reduce the impacts of climate change, ” Wilsey said.
Domino is helping Audubon save time and deploy its resources more effectively. For a non-profit taking on the threat of climate change, it’s not a moment too soon.