Soil Organic Indicators at Large Scale for Artificial Intelligence (SOILS-AI) Campaign
[Closed]
Soil Function proposals are now being accepted from members of the scientific community to join the ongoing Soil Organic Indicators at Large Scale for Artificial Intelligence (SOILS-AI) campaign for the Molecular Observation Network (MONet).
Participate in the MONet SOILS-AI Campaign
We invite the community to submit soil function proposals targeting soil orders critically needed in the MONet database—Histosols, Inceptisols, Entisols, Ultisols, Gellisols, Andisols, Vertisols, Aridisols, Spodosols, and Oxisols. Each proposal may include up to nine sample sets for small-scope proposals, representing one or a combination of these soil orders. Proposals for other soil orders will be declined. Review steps for determining soil taxonomy for your site.
One sample set consists of six cores. This includes
- one 30 cm deep core for biogeochemical measurements (i.e., DNA, FTICR)
- top and bottom 10 cm segments will be analyzed
- one 30 cm deep core for physical measurements (i.e., XCT)
- four 10 cm deep cores to account for soil heterogeneity.
SOILS-AI Campaign Proposal Submission and Timelines
- Proposal Submission: July 31, 2025, to August 29, 2025
- Decision Notice: Mid September 2025
- Sampling Period: September 30, 2025, to December 30, 2025
Proposal decision notices will be communicated from emsl@pnnl.gov.
About SOILS-AI Campaign
The SOILS-AI campaign is a sampling effort across the continental United States designed to improve the representation of high value soil taxa in the MONet database. AI models trained on MONet data have identified critical data gaps in existing predictive models for soil organic carbon (SOC), particularly in estimating carbon use efficiency (CUE) and soil respiration at regional scales. Addressing these gaps requires targeted sampling of critically needed soil taxa. Improving the representation of these key soil orders will strengthen predictions of soil organic indicators and biogeochemical processes that regulate organic carbon dynamics, improving Earth system models (ESMs) and the prediction of environmental impacts on energy infrastructure. The MONet platform supports the objectives of the DOE Biological and Environmental Research program while remaining focused on achieving outcomes at regional up to continental scale.
SOILS AI-Campaign Summary
Sakthi Kumaran, Campaign Lead
Central State University
A key barrier to scaling up SOC dynamics is the inherent spatial variability of soils, along with the lack of standardized molecular-level data for ESM prediction from local to continental scale. Accurately representing SOC processes at such scales requires integrating meter-scale variability into model parameterization. Pedotransfer functions (PTFs) provide a systematic framework for linking fine-scale observational data to model parameters, enabling the upscaling of biogeochemical processes and SOC dynamics.
PTFs integrated with AI modeling are being applied by the campaign lead Sakthi Kumaran of Central State University to MONet’s molecular SOC data integrated with the Soil Survey Geographical Database (SSURGO) open-data source. This approach enables the team to upscale local soil measurements to regional and continental scales. Simulations indicate significant limitations due to the lack of specific soil taxa at the soil order level in the MONet database, resulting in SOC molecular-scale data gaps and reduced accuracy in CUE and soil respiration predictions.
To address coverage gaps and improve the scalability of PTFs, the SOILS-AI Campaign will conduct distributed soil sampling across the United States, focusing on the following critically needed soil orders in MONet database: Histosols, Inceptisols, Entisols, Ultisols, Gellisols, Andisols, Vertisols, Aridisols, Spodosols, and Oxisols. SOILS-AI calls on broad community participation, with field collections planned over two years and seasonal sampling efforts designed to capture the temporal variability in soil respiration and microbial activity. This AI-enabled, integrative approach will advance the predictive understanding of soil organic indicators, improve the representation of soil processes in ESMs, and generate foundational knowledge to support resilient infrastructure and land-use strategies aligned with national energy goals.
Environmental Compliance
Because of PNNL’s contractual relationship with the U.S. Department of Energy, compliance with Federal, state, and local environmental regulations must be completed prior to approval of any application. EMSL will perform an environmental compliance review in collaboration with the PI. Proposals with a long compliance review may not fit the approval timeline of the specific requested proposals.
We ask proposal submitters to please provide complete and accurate documentation and respond to compliance team requests as quickly as possible to help ensure a timely process. Please note that an access agreement needs to be appended to the request form for each landowner. An access agreement is a formal letter or permit indicating explicit access granted to the PI to sample soil by a manager/supervisor of their field sites. Review examples of access agreements. Failure to do so will result in an automatic proposal rejection.
