Powered by Satellite Data+ML+Fluid Dynamics Simulations

Access field-level Soil Moisture Profiles without the need for sensors.

Reduce water waste
Prevent crop stress
Plan irrigation scheduling precisely
Sign Up for Beta Built for progressive farms & agribusiness
Secondary Dashboard Mockup Dashboard Mockup
Early Access

Pilot partners wanted: Join early to shape the platform and get white-glove onboarding.

Join the pilot

Intelligence built for your scale

Actionable moisture data whether you manage a single field or a global portfolio.

For Farmers & Agronomists

Precise, actionable insights to optimize your daily decisions on the ground.

  • High-res field-level moisture maps
  • Crop stress early warnings
  • Historical soil performance data

For Agribusiness & Enterprise

Scalable APIs and dashboard tools to monitor thousands of hectares seamlessly.

  • Global portfolio monitoring
  • Automated compliance reporting
  • Role-based access permissions

Managing large operations?

Request an Enterprise Pilot

How we calculate the unseen

We combine a unique ML model trained on satellite data with Richard's equation simulations to provide up to date soil moisture predictions. These are used to predict future soil moisture profiles, allowing for better resource management.

1

Satellite Observations

Ingesting Sentinel 1, Modis, HLS, Era5 and SMAP data every few days.

2

Priors & Context

Overlaying local weather, precise topography, and known soil types.

3

AI + Physics Engine

Unique machine learning model informed by numerical simulations.

4

Actionable Outputs

Delivering high-res maps (50/100 meters), 7-day trends, and irrigation alerts.

Shape the future of irrigation.

We're currently in closed beta. Join the waitlist to secure early access and help direct our product roadmap.

We only use this to contact you about the beta.

Frequently Asked Questions

By blending machine learning with physics-based fluid simulations, our models achieve high correlation with in-situ probes (typically within 3-5% volumetric water content margin of error), depending on crop canopy density and soil type.

We primarily ingest Sentinel-1 radar readings, HLS optical imagery, Modis land surface temperature, Era5 Meteorological data and SMAP mission data via Google Earth Engine. Data updates roughly every 3 to 6 days depending on satellite overpasses, but our physics engine interpolates daily states.

Yes. Our models utilize global soil grids (sand/clay/silt fractions) as priors. The system supports most broadacre and orchard crops, though extremely dense canopies (like mature corn) rely heavier on the physics/weather interpolation than direct radar returns.

No hardware is required to get started. Our platform is 100% remote-sensed. However, if you do have existing IoT sensors, we plan to offer API integrations to calibrate the spatial maps to your exact point data.

Beta access is completely free. Post-beta, we will introduce a tiered pricing model based on hectares managed, designed to be significantly cheaper than deploying and maintaining physical sensor networks.

Our beta is currently open globally, though our core training data is heavily calibrated for North America and Europe. We are rapidly expanding our validation regions.