TERRA COVER
A satellite-plus-machine-learning platform for irrigation-need and soil-moisture analysis
TERRA COVER uses satellite imagery plus machine-learning models to estimate farmland's soil-moisture status and crop irrigation needs. Its logic: rather than sticking sensors all over the field (expensive, easily broken, and only representing one point), infer the whole field's moisture status from the satellite's multispectral and thermal-infrared bands, then combine weather data to compute "how much water to apply these days."
Features and use scenarios
Core capabilities include soil-moisture inversion, crop water-demand estimation, irrigation-scheduling suggestions, and intra-field variability maps. The variability maps are especially practical — different areas of the same field often have very different water-retention capacity, and uniform irrigation means simultaneously wasting water and leaving some areas short; zone-based irrigation is where precision agriculture really saves money.
It suits medium-to-large farms, irrigation-equipment vendors, and agricultural consultants. Taiwan's water-shortage frequency has risen in recent years and agricultural water use is a high share, so this "using remote sensing instead of dense sensors" approach has reference value for zone-based irrigation management of rice paddies and orchards.
Main features
- Satellite soil-moisture inversion
- Crop water-demand estimation
- Irrigation-scheduling suggestions
- Intra-field variability maps
- Weather-data integration
Common uses
- Large-area farm irrigation management
- Water allocation during droughts
- Precision-agriculture zone-based irrigation
- Agricultural water-rights assessment
Key Features
- Satellite soil-moisture inversion
- Crop water-demand estimation
- Irrigation-scheduling suggestions
- Intra-field variability maps
- Weather-data integration
Pros
- No need to deploy large numbers of sensors in the field
- Can cover large areas at controllable cost
- Zone-based irrigation suggestions actually save water
Cons
- Cloud cover affects optical-image quality
- Inversion results are model estimates, needing local calibration
- Resolution may be insufficient for small plots
Use Cases
- Large-area farm irrigation management
- Water allocation during droughts
- Precision-agriculture zone-based irrigation
- Agricultural water-rights assessment
Editor's Note
Precision agriculture has been talked about for over a decade, and the best real-world landing has always been irrigation — because water and electricity bills are expenses farmers see every month. The money saved is very concrete, which is more convincing than talking about yield increases.
FAQ
Taiwan's fields are so small — is the resolution enough?
This is a real issue. The pixel size of public satellite imagery may be too coarse for Taiwan's common small plots, better suited to large contiguous farming; before adopting, be sure to confirm the minimum recognizable unit for your target fields.
What about clouds on rainy days?
Optical imagery is indeed blocked by clouds; usually radar imagery or model interpolation fills the gap, but data quality drops during continuous rainy spells — a common limitation of all satellite agriculture services.