November 24, 2020
Web Feature

Bringing Soil Respiration Data into the Open

Open-access database standardizes Earth system research data

Instrument for measuring soil-atmosphere gas flux on leaf-strewn ground

Accessing data acquired in different field campaigns can be a challenge that makes large-scale comparative studies seemingly impossible.

(Photo courtesy of Ben Bond-Lamberty | Pacific Northwest National Laboratory)

Collecting datasets from laborious field campaigns seems like it should be more difficult than finding and comparing data from different campaigns. Unfortunately, that often is not the case. According to Pacific Northwest National Laboratory (PNNL) Earth scientist Jinshi Jian, “Assembling datasets for comparison is difficult and time-consuming work. Units, conventions, nomenclatures—a lot of things can vary in confusing ways.” Data can also be inaccessible behind paywalls or impossible to download.

Instruments for measuring soil-atmosphere flux data by trees in a wooded area
Creating a usable database took collaboration between researchers at over 70 institutions across the globe. (Photo courtesy of Ben Bond-Lamberty | Pacific Northwest National Laboratory)

These problems exist across most fields, but a collaboration between researchers working with measurements of gas exchange between the soil and atmosphere made by automated systems, also known as soil-atmosphere flux measurements or soil respiration, is moving their field toward a solution.

Understanding how levels of gases entering and leaving the soil change in response to changes in the atmosphere is crucial to get a full picture of the evolving global carbon cycle and resulting climate changes. Datasets measured in the field often focus on a highly specific geographic area and present broad formatting compatibility issues, but stitching them together to create a more global dataset could provide broad biogeochemical insights.

To move science forward, a global group of researchers from more than 70 institutions contributed to the COSORE (short for COntinuous SOil Respiration) database with the goal of creating a broad resource containing standardized soil flux datasets from around the world that have been converted into a useful form for researchers. In an effort led by PNNL’s Ben Bond-Lamberty, COSORE launched last month with over 8 million data points available. According to Bond-Lamberty, members of the research community are already excited about the project.

Creating an accessible package for students and researchers

A decade ago, Bond-Lamberty worked on a similar project and created a database (the Soil Respiration Database or SRDB) of annual flux measurements that has been widely used since its inception. The creation of SRDB provided valuable lessons on database design and the usefulness of a large data repository, which shaped his philosophy while building COSORE.

COSORE’s creation process was quite rapid. Bond-Lamberty took the idea from conception to publication in approximately eighteen months, a quick turnaround for such a large project.

“A guiding philosophy building COSORE was to make it robust and high quality, but also not to let the perfect be the enemy of the good. I wanted to make sure the package was usable and accessible to scientists with a range of statistical and programming skills,” said Bond-Lamberty.

COSORE’s usability was put to the test this summer. Bond-Lamberty worked with a high school intern over the summer who used the package to examine the temperature sensitivity of soil-atmosphere gas fluxes, paying particular attention to differences based on location and the day versus night cycle.

Graph showing the distribution of datasets in COSORE with an inset displaying the type of enviroment
COSORE has datasets from different areas of the globe, including locations in North America, western Europe, and China, that allow users to compare trends across these regions. (Figure courtesy of Ben Bond-Lamberty | PNNL)

A community repository for new research directions

A particular advantage of COSORE is that it creates new research possibilities. Many of the datasets it holds are so specialized that, on their own, only a small group of researchers would be interested in them. However, by combining these datasets, COSORE allows users to compare trends across different areas of the globe, potentially providing unique new insights.

Broadening the number of available datasets is just one benefit of COSORE. Having a central repository helps protect the scientific field from losing the flux data, even if the underlying data vanishes through broken storage and email chains. Finally, COSORE will hopefully accelerate science by saving researchers the time and effort of assembling disparate datasets.

Bond-Lamberty hopes COSORE will continue to grow as researchers submit more templated data to the database. Datasets from locations in North America, western Europe, and China dominate the database, leaving gaps in important tropical regions. As new datasets from important regions get collected, COSORE can make them available to researchers around the world who can use the resource to push their science further forward.

Published: November 24, 2020