Can Satellite Imagery Prove a Field Was Damaged by Heat, or Is Resolution Too Coarse?

A practical guide to whether satellite imagery can prove a specific field was damaged by a heat event during the growing season, why thermal and optical resolution are two completely different stories, the crop-specific temperature thresholds any proof gets checked against, what a genuine evidence package actually requires, and the one kind of heat damage no satellite layer can see directly.

The real question isn't resolution, it's what kind of heat damage happened

Heat damages a crop two different ways, and only one of them leaves a picture a satellite can actually see.

Corn is the clearest example. Research on heat-stressed maize shows pollen damage can cause 3 to 8% yield loss per day of stress during silking and pollen shed, through tassel blast, pollen abortion, and poor pollination, mechanisms that happen inside the flower, not on the leaf surface a satellite is actually imaging. A field can carry this exact damage and still show a perfectly normal NDVI map the week it happened.

Thermal resolution, the part that actually used to be coarse

Measuring a field's actual canopy temperature, not just its color, needs a thermal infrared sensor, and thermal sensors have historically carried much larger pixels than optical ones for physical reasons, a thermal detector needs more energy per pixel to get a usable reading.

The practical takeaway, the "thermal is too coarse" problem is a 2024-and-earlier problem. A field large enough for commercial row-crop agriculture was already resolvable by Landsat or ECOSTRESS; a genuinely small or oddly shaped field now has a real shot with the new sub-30m commercial thermal constellations, and that gap keeps closing as more of these satellites reach orbit.

Optical and multispectral resolution was never the limiting factor. SuperView Neo-1 resolves to 30 cm panchromatic and 1.2 m multispectral, SuperView-2 to 40 cm panchromatic and 1.68 m multispectral with a Red Edge band included, and Beijing-3 (21AT) sits in the same tier, all sharp enough to isolate one specific field, down to individual rows, from its neighbors with no ambiguity at all. The bottleneck was always the thermal side, since no commercial VHR optical constellation has historically carried a usable thermal band at the same sharpness as its visible and NIR sensors.

The crop-specific thresholds any claim gets checked against

A satellite signal means nothing on its own without a known damage threshold for that crop and growth stage to compare it to.

Duration compounds the effect on top of the peak temperature. One published analysis found 12 or more days at least two standard deviations above the long-term monthly average cut yields by as much as 36%, which is why a single hot afternoon and a sustained two-week heatwave are not treated as the same event in a proof package, even if both cross the same peak-temperature threshold.

What a genuine proof package actually requires

This is the same logic parametric insurers use to manage basis risk, the gap between what a regional weather or satellite index says happened and what actually happened on one specific insured field. The European Space Agency has reported that combining satellite monitoring with ground weather data can improve detection of extreme weather crop impacts by up to 50% compared to relying on ground data alone, precisely because it closes that single-field gap a regional index can't.

The one kind of heat damage no satellite layer proves directly

Reproductive-stage damage is the honest limit, and it's worth stating plainly rather than overselling what the data can do.

• Pollen sterility and flower abortion happen inside the plant, invisible to any sensor looking at the canopy from above, whether optical or thermal.
• The crop can look completely healthy through the rest of the season, with a normal-looking NDVI curve, right up until a visibly light or empty harvest.
• What satellite data can still do is tie the timing together, confirm the field was at its documented flowering or pollination stage exactly when the weather data shows the threshold was crossed, which is strong circumstantial evidence even with zero visible canopy symptoms.
• The downstream signal still shows up eventually, a lower end-of-season biomass or yield-model estimate than the field's own history would predict, consistent with the documented event, even though the event itself left no visible mark at the time.

In practice, this is the difference between "satellite imagery shows this field was damaged" and "satellite and weather data together are consistent with this field being damaged, at a stage where the damage mechanism itself isn't directly observable." The second framing is the honest one, and it's still a genuinely useful evidence package, just not an unqualified single-image proof.

What we offer

• Multi-temporal archive search over your exact field. Pull before, during, and after imagery for the documented event window, not a single scene.
• Commercial optical down to 25 cm. Resolves one field from its neighbors with no ambiguity, for the visible-damage cases.
• Access to thermal and multispectral data layers for canopy temperature and water-stress indices like CWSI, where the question is heat specifically rather than general vegetation decline.
• New tasking when the archive has a gap. If no suitable scene exists for the exact dates that matter, we can task a fresh capture.
• Instant, transparent pricing on every archive search or tasking request, so building an evidence file doesn't become its own slow procurement process.
• A human confirms coverage and cloud conditions for your exact field and dates before anything is charged.

Frequently asked questions

Can satellite imagery actually prove that a specific field was damaged by a heat event during the growing season or is the resolution too coarse?

Partially. Optical and multispectral resolution has resolved individual fields down to 25cm for years, not coarse at all. Thermal resolution used to be the real limit, Landsat at 100m and ECOSTRESS at 70m, but new commercial thermal constellations now reach 25m and as fine as 3.5m. The bigger honest limit isn't resolution, it's that no single satellite layer proves causation alone, a credible case combines weather data, imagery of the specific field, and a before-and-after or neighboring-field comparison.

What resolution satellite imagery is needed to detect heat damage on one field?

For visible damage like scorch or biomass loss, commercial optical imagery at 25cm to a few meters easily isolates one field. For direct canopy-temperature measurement, thermal sensors need finer resolution than they used to have, Landsat and ECOSTRESS resolve to 70 to 100m, while newer commercial constellations like Hydrosat and SatVu now reach 25m and down to 3.5m.

Can satellites detect heat stress that doesn't show visible crop damage?

Not directly, if the damage is reproductive-stage, such as pollen sterility or flower abortion in corn during silking. These mechanisms happen inside the plant with no visible canopy change at the time, and can cause 3 to 8% yield loss per day of stress despite a normal-looking NDVI map. The eventual yield or biomass shortfall is the only satellite-visible signal, and it shows up after the fact.

What temperature causes heat damage in corn, soybean, and wheat?

Corn heat stress indemnities typically begin around 30°C and intensify sharply by 38°C, with pollination-specific damage from roughly 35°C. Soybean losses increase from around 33°C. Winter wheat can see yield-reducing heat from roughly 20°C and above during its sensitive flowering and grain-fill window. Duration matters as much as the peak, 12 or more days of sustained heat above normal can cut yield by up to 36%.

What evidence does a credible heat-damage claim need beyond satellite imagery?

Weather-station or reanalysis data confirming the regional event crossed the crop's known damage threshold during its most sensitive growth stage, satellite imagery showing the specific field's actual response, and a comparison against the field's own history or a neighboring unaffected field to rule out other causes like drought, disease, or pests.

What is the difference between thermal and optical satellite resolution for crop monitoring?

Optical and multispectral sensors measure reflected light and have reached 25 to 50cm resolution commercially for years. Thermal sensors measure emitted heat directly and have historically needed much larger pixels to collect a usable signal, 70 to 100m on Landsat and ECOSTRESS, though new commercial constellations are closing that gap to 25m and finer.

Does combining weather data with satellite imagery improve heat-damage detection?

Yes. The European Space Agency has reported that combining satellite monitoring with ground weather data can improve detection of extreme weather crop impacts by up to 50% compared to ground data alone, since it closes the gap between a regional event and what actually happened on one specific field, the same basis-risk problem parametric crop insurers manage.

Sources and further reading

• USGS Landsat 8/9 TIRS thermal band specifications, 100m native resolution
• NASA JPL, ECOSTRESS mission specifications, 70m resolution, ISS-based platform
• Hydrosat and SatVu (Satellite Vu) commercial thermal constellation specifications, 2026
• SuperView Neo-1, SuperView-2, and Gaofen-5B (GF-5B) VIMS payload sensor specifications
• Peer-reviewed research on maize pollen sterility and reproductive-stage heat stress, tassel blast, pollen abortion, and daily yield-loss rates during silking
• Crop insurance and agricultural economics research on corn, soybean, and wheat heat-damage temperature thresholds
• Research on sustained above-normal temperature duration and yield loss
• European Space Agency reporting on combined satellite and ground-data detection of extreme weather crop impacts
• Research on basis risk in parametric and index-based crop insurance

For how satellite data separates heat and drought stress from pests, disease, and nutrient issues more broadly, see our guide to diagnosing crop stress in satellite imagery, and for how this evidence feeds an actual insurance claim, see satellite imagery as insurance claims evidence.