Building the tools to make environmental data more accessible and forecasts more accurate

Pause for 1 minute before starting your next Google search. You may not know it, but whether you’re looking for the latest soccer hawkey scores or cheap airline tickets, a powerful search engine made possible by agreed-upon rules for defining information. You’re unlocking the process of data discovery, search, and organization. Run your search engine.

Stop again and imagine if every website used a different set of rules and search engines weren’t available. Given the daunting amount of information on the Internet and at your fingertips, how can you find what you need to make decisions and plan your life? For scientists navigating large amounts of environmental data scattered across the world, raising that query a few notches will help you understand the implications of a new research project at Virginia Tech.

Quinn Thomas, who holds dual positions at the University of Natural Resources and the Environment and the School of Science, is part lead at Virginia Tech in a $3.2 million research project funded by the Office of Advanced Cyber ​​Infrastructure at the National Science Foundation (NSF). I am a researcher.

Democratized Cyberinfrastructure for Open Discovery to Enable Research (DeCODER) aims to standardize and promote the way environmental data and model predictions are described and shared. This will ultimately make these resources available to more individual researchers and the scientific community.

Data is a key driver of this project and of the ecological forecasting work of Thomas, an associate professor in the Department of Forest Resources and Conservation, an associate faculty member at the Center for Global Change, and a Data Science Faculty Fellow at the university. Chemistry. He is a researcher with the ambitious goal of predicting the natural environment in the same way he predicts the weather, using shared data tools and computational infrastructure.

As project leader at Virginia Tech, Thomas will use the university’s share of a $535,000 NSF grant to support researchers interested in predicting environmental change. “My part of the project is to increase the discoverability of ecological predictions through the development of protocols and software for archiving and documenting model predictions of ecological dynamics,” he said. I’m here. “In the same way that we use Internet search engines (such as Google) to search for information, our job is to help researchers ask questions and find current projections to guide their decisions. to help initiate searches such as “find algae forecasts for lakes across the United States.” We support environmental management. ”

This grant advances work already being done on the EarthCube GeoCODES platform. EarthCube is an NSF-funded environment that improves data access, sharing, and visualization. GeoCODES is a special program for researchers working in the field of earth sciences, offering advanced ways to organize and easily access data, as well as a framework of new computational tools, registries, and user communities. It also provides best practices for

The new DeCODER platform builds on and leverages work already done as part of the EarthCube effort. Thomas takes the next step, helping researchers, especially those working on ecological forecasting, to more easily access data and create better models.

Again, given the example of a researcher who needs to predict algal growth across the United States, the DeCODER platform allows researchers to not only collect data and predictions, but also “translate these predictions into We can also quantify the strengths and weaknesses of the algae compared to the actual measurements, the predictions generated so far.”

Further, Thomas said: .”

This new platform is of particular value to researchers like Thomas. Researchers like Thomas need access to data and modeling. This makes it easier to discover what has already been done in the field to improve the model over time.

“Think about the weather forecast,” Thomas said. “It has improved over time. The 10-day forecast is as good as the 8- or 9-day forecast 10 years ago. We want to do this now, because we also need to evaluate other environmental predictions, and we can’t do that unless we can find all the historical data.”

He also said that individual scientists generate an incredible amount of data about the environment, but unfortunately not all are aggregated in one place. This new technology will allow data to be discovered anywhere, allowing researchers to determine if they are improving their predictions of environmental change.

Thomas plans to work closely with Associate Professor Carl Boettiger at the University of California, Berkeley (UCB) to apply the DeCODER platform to ecological forecasting. His main focus is developing software and protocols that enable people to discover the data they need. “The DeCODER project will help democratize research pipelines such as ecological forecasting and evaluation, bridge the scientific community, and better inform decision makers,” said Boettiger.

To achieve this ambitious goal, the project involves collaborative research by multiple teams with specific specialties. In addition to Virginia Tech and UCB’s focus on environmental forecasting data, the University of Illinois at Urbana-Champaign (lead institution) and the University of California, San Diego (UCSD) are helping all teams connect their work. I am developing a cyber infrastructure to use. Syracuse University and Texas A&M University are working on low-temperature geochemical data, and the Scripps Institute of Oceanography, with his UCSD, is focused on deep-sea science data.

According to Thomas, this newly funded NSF-funded project complements and advances the ongoing ecological forecasting research agenda at Virginia Tech. The DeCODER platform will ultimately accelerate the work of current forecasting projects related to water quality, forest carbon stocks, autumn colors and environmental dynamics in a changing environment.

“By focusing on a democratized approach to data and prediction discovery, progress is designed to last beyond the life of the project.

Related article:

Courtesy of Virginia Tech