Every 30 meters.
A 30 m water grid. The weather context is one cell of about 11 km, copied from the field centre, so a difference between neighbouring cells is not a measured difference in millimetres.
Valthgard estimates, from space, the water and energy that land exchanges with the air, on a 30 m grid, across crops, forests and grasslands. Carbon comes next.
One science, many landscapes.
HydroFlux learns from direct measurements taken at research stations in croplands, forests, grasslands and wetlands. Each kind of land earns its status through tests set before the results are seen, and we show that status as it stands.
- Annual cropsCertifiedNorth American annual crops and forage only. Other agriculture is under validation. A field whose crop is not documented is not certified.
- ForestsIn development
- GrasslandsIn development
- WetlandsResearch
Where a landscape isn't supported yet, HydroFlux returns no number rather than a guess.
Beyond water, carbon is in research. See the carbon research →
Irrigation runs on averages. Averages are expensive.
In 2023, US farms spent $3.3 billion on energy just to pump irrigation water, and another $1.1 billion buying it. Uniform irrigation spreads that water as if every part of a field were the same. No field is.
- $3.3B
- on energy to pump irrigation water in 2023
- $20,700
- average pumping energy cost per farm
- 81M
- acre-feet of water applied
Source: USDA NASS, 2023 Irrigation and Water Management Survey
Irrigating by zone saves water.
In a two-year USDA field study, irrigating soybeans by zone instead of uniformly used 25% less irrigation water, with no loss of yield. Seeing those differences on a particular field is a separate question. A HydroFlux map is an estimate with coverage and abstentions; it is not that saving.
A field in eastern Nebraska, as HydroFlux estimated it.
Water and energy, cell by cell, over a 125.3 ha field for 16–31 August 2026.
Inference run on 29 September 2026.
What HydroFlux gives you
Every day.
A daily estimate from optical, radar and thermal satellites. Where there's no usable observation, it says so instead of guessing.
With its support status.
Where a number is served, its support status is served with it. An uncertainty range, when the product provides one, is that product's range. Whether a difference should change an action is a separate decision.
Built on field science.
Trained on direct measurements of evapotranspiration taken at research stations.
How a pilot works
Tell us your field.
Share a boundary and the season you care about.
We map the season.
HydroFlux estimates daily water use on the 30 m grid where the observations support it, and leaves a cell blank when they do not.
You see where it differs.
Find the zones worth checking first. The irrigation decision stays yours.
A flux tower sees a few hundred meters. Carbon projects cover whole landscapes.
Eddy-covariance towers are the most direct way to measure how land exchanges carbon and water. They are also costly and local. Each one measures only the few hundred meters upwind of it, and a conventional tower is expensive enough that researchers now design versions at about a quarter of the cost. Beyond that footprint, projects extrapolate.
The water map is a 30 m grid, with the limits above. Carbon was evaluated on features aggregated around towers, not as a map of cells. It stays in research. This site does not offer a carbon number, a field coverage guarantee, or a monthly interval.
Fewer towers, wider coverage.
The aim: a handful of towers anchoring estimates across a whole project area.
Anchored in direct measurements.
Built on eddy-covariance data, not on generic emission factors or regional averages.
Uncertainty from the start.
Every estimate would carry its limits, as HydroFlux's water estimates do today.
Carbon is in research: we don't offer carbon estimates yet.
Talk to us about carbon researchFootprint: Lloyd, FLUXNET field deployment handbook, 2023 · Tower cost: van Ramshorst et al., Atmospheric Measurement Techniques, 2024
Why now
Water is the input agriculture can least afford to waste, and the limits on it keep tightening. Open satellite data and modern models now make field-level water intelligence practical. Valthgard turns that into a water estimate growers and irrigation companies can pilot. Forests, grasslands and carbon stay at the status shown above.
Start with a real question about your land.
Request a pilotAccuracy depends on landscape and region. See where HydroFlux is validated