Chemistry
Nutrients, pH, organic carbon, and cation exchange resolved by strata rather than by field average.
Viridi Data is a measurement-first environmental intelligence platform. Real soil cores, laboratory analysis, and cloud modeling join into one defensible dataset — delivered as reports, dashboards, and APIs.
Models interpret the measurements. They never replace them.
Measurement-first and externally verifiable — every figure traces back to a physical sample.
A single field can change texture, drainage, and carbon-holding capacity several times between fence lines. The data used to manage that field rarely reflects it.
Soil surveys still in daily use were mapped decades ago at a resolution far coarser than the decisions made on top of them. Satellite and model-only products fill the gaps with inference, and that inference carries straight into nutrient rates, risk pricing, conservation spending, and carbon permanence. The inference may be reasonable. It is not measured, and it does not hold up in an audit.
Viridi joins the layers normally collected apart — or not collected at all — into a single, audit-ready dataset, tied to real cores with known locations, depths, dates, and laboratory methods.
Nutrients, pH, organic carbon, and cation exchange resolved by strata rather than by field average.
Bulk density measured per horizon, so carbon and water figures are stated per volume instead of assumed.
Sand, silt, and clay fractions — what governs how a soil holds water, nutrients, and structure.
Infiltration and water-holding behavior derived from measured texture and density, with climate context.
Emerging microbial and fungal indicators, reported as the evidence matures rather than ahead of it.
Field and program deliverables written for review — measured values, methods, and stated uncertainty.
Maps and distributions where the modeled surface and the measured points stay visible together.
Queryable access to the same dataset, so the values reaching a model are the values in the record.
The dataset is processed and served from a Snowflake-based cloud platform. The data is the product: Viridi is built to operate as a data utility, not a consulting engagement.
The order matters more than any single step. Fieldwork comes first, laboratory analysis confirms it, and only then does statistical processing extend the result across the map.
Cores are pulled on a defined sampling design, by strata, with location, depth, and date recorded at collection. Nothing downstream exists without this step.
Samples go to laboratory analysis under standardized protocols, with quality control on handling and chain of custody. Every value traces back to a sample and a method.
Proprietary statistical processing interpolates between measured points and states the uncertainty. Models extend the measurements; they do not stand in for them.
Results land as reports, dashboards, and APIs, with assumptions named and the measured points visible beneath the modeled surface.
Different questions, one requirement in common: a figure that can be traced to measured ground when someone asks where it came from.
Defensible soil and land baselines for conservation programs, watershed and erosion work, and reporting that has to survive review.
Field-level variability at the resolution decisions are actually made — where nutrients respond, where water sits, where ground differs from the map.
Outcome reporting anchored to measured ground rather than a modeled estimate, for teams whose disclosures get checked.
High-resolution, multi-layer soil data with methods and uncertainty documented, structured for reuse across studies.
Ground-verified inputs for risk, valuation, and carbon permanence models, with the measurement history attached.
Six things decide whether an environmental dataset is worth building a decision on. Viridi is organized around all six.
Physical samples come first. Models interpret them; they never replace them.
The largest cost in this work — field sampling — is funded through an affiliated carbon program, so dense measured coverage is produced at marginal cost.
Chemistry, density, texture, hydrology, and biology are rarely joined at scale. Here they are one dataset.
Every deliverable traces to a sample, a method, and a date — built for verifiers and regulators, not for a pitch.
No grain buyer, input seller, or supply chain sits behind the data. The measurement has no other job.
Viridi adds precision where current methods are coarse. Where a result is directionally correct, it says so and names the assumptions.
If you need a soil figure that holds up under review, the fastest way to find out whether Viridi fits is a short call. Tell us the decision you are trying to defend.