DOE OSTI · 3030515
Do we have globally representative data to understand soil processes?
Abstract
Understanding and modeling soils and soil organic matter (SOM) are central to a variety of human needs, from food production to ecosystem management. Soil data have been collected for over a century, but the global spatial and process representativeness of soil data remains unclear. We assessed the representativeness of currently available soil data that could be used to understand a variety of SOM processes. We used 16 open-source soil databases and data from over 281,000 unique locations globally, categorizing the databases into three main data types necessary to understand SOM processes: soil carbon stocks and fluxes, mechanistic drivers of these stocks and fluxes, and soil carbon gain or loss potential. We found that stock and driver data have extensive global coverage. However, data on soil carbon gain or loss potential, particularly data describing change in soils over time such as time series data, are severely limited in their global coverage. We conclude that while significant strides have been made in measuring soil carbon stocks and fluxes, and their drivers, we are limited in global data related to changes in soils over time. Our recommendations for soil data generators are to ensure precise metadata reporting and prioritizing sampling in underrepresented areas like tropical, arctic, mountainous, wetland and arid regions. We also encourage designing revisit schemes that explicitly support change detection and reporting multi-modal datasets that can aid in model development. Targeted measurement of low coverage soil data types and regions is necessary for a range of applications including current and future biogeochemical predictions, and their management and policy implications.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Malhotra, Avni [Pacific Northwest National Laboratory (PNNL), Richland, WA (United States)] (ORCID:0000000278506402), von Fromm, Sophie F. [Dartmouth College, Hanover, NH (United States)], Bond-Lamberty, Ben [Pacific Northwest National Laboratory (PNNL), Richland, WA (United States)], Doetterl, Sebastian [Eidgenoessische Technische Hochschule (ETH), Zurich (Switzerland)], Georgiou, Katerina [Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States); Oregon State Univ., Corvallis, OR (United States)], Graham, Emily B. [Pacific Northwest National Laboratory (PNNL), Richland, WA (United States); Washington State Univ., Pullman, WA (United States)], Heckman, Katherine A. [USDA Forest Service, Houghton, MI (United States). Northern Research Station], Jia, Ruofei [Massachusetts Inst. of Technology (MIT), Cambridge, MA (United States)], Patel, Kaizad F. [Pacific Northwest National Laboratory (PNNL), Richland, WA (United States)], Rod, Kenton A. [Pacific Northwest National Laboratory (PNNL), Richland, WA (United States)], Santos, Fernanda [Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)] (ORCID:0000000191555623), Terrer, César [Massachusetts Inst. of Technology (MIT), Cambridge, MA (United States)], Todd-Brown, Katherine [Univ. of Florida, Gainesville, FL (United States)], Zheng, Jianqiu [Pacific Northwest National Laboratory (PNNL), Richland, WA (United States)], Hofmockel, Kirsten [Pacific Northwest National Laboratory (PNNL), Richland, WA (United States)], Bailey, Vanessa [Pacific Northwest National Laboratory (PNNL), Richland, WA (United States)]. 2026-02-20. Do we have globally representative data to understand soil processes?. https://doi.org/10.1007/s10533-025-01301-z
Cite the original work for its findings. Save a collection to share your selection of sources.