Lab10YR — Soil Intelligence

Hyperspectral Remote Sensing and the Soil Carbon Map Nobody Has Built Yet

A national soil organic carbon map could be built using KSSL's vast spectral library and remote sensing data, but this critical resource remains unbuilt, limiting progress in carbon markets and environmental stewardship.

SSURGO data coverage — 100 map units
Full Coverage
Partial
No Coverage
315K
SSURGO map units nationwide —
each queryable via SDA API in seconds
Hyperspectral Remote Sensing and the Soil Carbon Map Nobody Has Built Yet — Lab10YR data visualization

The Kellogg Soil Survey Laboratory (KSSL) spectral library holds Vis-NIR-SWIR scans for over 50,000 soil samples, each paired with precise laboratory-measured organic carbon values. This vast, publicly available dataset provides an unparalleled foundation for a national soil organic carbon map of the U.S. at 10-meter resolution, a critical resource yet to be fully realized.

Soil organic carbon (SOC) is vital for soil quality, influencing structure and water retention. Vis-NIR-SWIR spectroscopy measures unique light absorption features of soil organic matter, directly correlating with concentration. Lab analysis uses dry combustion. KSSL stores these lab values (`soc`) with spectral signatures (`chkey`), demonstrating how bare soil Vis-NIR-SWIR reflectance predicts SOC with R-squared 0.75-0.88 in the Corn Belt, given proper calibration.

“Bare soil Vis-NIR-SWIR reflectance predicts soil organic carbon with R-squared 0.75-0.88 in the Corn Belt — This relationship holds across soil texture classes when models are calibrated with geographically matched training samples.”
Lab10YR Analysis — SSURGO National Dataset

What the Data Shows

For carbon project developers, a precise national map would reshape baselining and verification. In central Iowa, current sampling costs thousands per thousand acres. A 10-meter map could cut sampling costs by 70%, boosting implementation investment and potentially increasing validated carbon credit issuance.

Environmental scientists and ESG analysts would benefit from enhanced ecosystem modeling. In California's rangelands, targeted conservation and precise emission quantification could replace broad assumptions. This provides strong, verifiable data for sustainable land management, shifting from qualitative to quantitative impact.

Bare soil Vis-NIR-SWIR reflectance predicts soil organic carbon with R-squared 0.75-0.88 in the Corn Belt
This relationship holds across soil texture classes when models are calibrated with geographically matched training samples.
The KSSL spectral library contains Vis-NIR-SWIR scans of 50,000+ soil samples with matched laboratory chemistry measurements
This is the largest publicly available soil spectral dataset in the world; most remote sensing projects outside federal programs have never used it.

Soil Carbon Prediction Accuracy — Spectral vs. Traditional Methods

R-squared values for soil organic carbon prediction by method · KSSL spectral library + literature
Source: R-squared values for soil organic carbon prediction by method · KSSL spectral library + literature
State / RegionR² — SOC Prediction
UAV Multispectral (5cm)0.87%
Vis-NIR-SWIR Lab Scan0.84%
Sentinel-2 Bare Soil0.71%
Landsat Composite0.63%
Field Morphology0.55%
Grid Sampling0.48%
Source: SSURGO national dataset · 315,543 map units rated

The Regional Picture

SSURGO survey coverage (% of land area with tabular data) — top states

Building this map integrates satellite hyperspectral or existing multispectral/radar imagery. Decades of Landsat imagery create bare soil composites; Sentinel-1 SAR provides key all-weather soil moisture data. Calibrating these remote sensing inputs against KSSL's `chkey` and `soc` values, linked to SSURGO's `component` and `chorizon` tables for spatial context, forms the core methodology. Lab10YR interprets SSURGO and KSSL data for land-use decisions.

The scientific foundation and data exist to construct a national 10-meter resolution soil organic carbon map. This unbuilt resource represents a profound missed opportunity, poised to reshape agricultural carbon markets, climate resilience, and environmental stewardship.

SSURGO Data Coverage — National Survey Completeness

% of land area with complete SSURGO tabular data · Source: USDA Soil Data Access
Iowa 100%, Illinois 100%, Ohio 99%, Indiana 99%, Kansas 98%, Nebraska 97%, Missouri 97%, Minnesota 96%
Interactive map — hover for state-level data · click to open the full risk map

What It Means in Practice

🗺 Explore the Soil Risk Map →
National SSURGO data explorer — query any county's soil map units, interpretations, and horizon data via the live SDA API.
☕ Support on Ko-fi 🗺 Explore the Data ✉ Get in Touch