New tool offers local climate change information

June 4, 2018

Lindsey Heaney
907-474-1939

A new web-based tool will allow communities in Alaska and western Canada to see how global climate change could affect their regions.

A team in the Alaska Climate Adaptation Science Center at the University of Alaska Fairbanks created the tool, which transforms predictions from global climate models into more detailed information about local conditions.

“We recognized that these global climate models by themselves cannot be used at face value,” said UAF’s John Walsh, the project’s lead researcher. “The raw output is not suitable for what many users or decision-makers want. That recognition was the motivator.”

So the team created a community charts tool to help people see and interpret the data. The tool displays temperature and precipitation projections through 2100 under three scenarios for more than 4,000 communities in Alaska and western Canada. The scenarios are based on three different levels of greenhouse gas emissions linked to the burning of fossil fuels.

The tool can be customized. For instance, it allows users to ask questions like which winter months may transition to above-freezing temperatures. Season lengths can affect transportation, subsistence hunting and fishing, and other activities.

The tool is hosted at the UAF Scenarios Network for Alaska and Arctic Planning group’s website.

“You can get an output at the SNAP website for your community based on the work of thousands of people in the climate modeling enterprise around the world,” Walsh explained. “We’ve distilled all of these efforts down to tools that will get you the information you really want for your location.”

The tool uses data from a process called statistical downscaling, which takes global climate models and transforms them into models with finer resolution.

With the help of the SNAP group, the climate research center’s team looked at 30 global models and identified five that worked best for Alaska and the Arctic. They compared what the models said should have happened from 1958-2004 to actual historical data from that period.

“Since climate in Alaska varies considerably from year to year and decade to decade, we want to use the longest record possible," explained co-author Jeremy Littell, a climate center researcher with the U.S. Geological Survey in Anchorage. "Averages for 50 or 100 years are more reliable than 10 or 20 years."

The SNAP team also developed a tool for modelers and climate experts to better identify useful models. Of the roughly 30 global climate models to select from, each has strengths and weaknesses when considering different regions and variables.

“The community charts tool is something very useful that came out of this, but so is the companion tool that was created to help others in the research and modeling community to better identify accurate models to help their future projections,” said SNAP operations lead Tom Kurkowski.

The researchers described their work in a recently published paper in the journal Environmental Modeling and Software.

ADDITIONAL CONTACTS: John Walsh, 907-474-2677, jewalsh@alaska.edu

ON THE WEB:
Community data: https://www.snap.uaf.edu/sites/all/modules/snap_community_charts/charts.php
Tool for modelers: https://uasnap.shinyapps.io/ar5eval/