Satellite Farm Advisory Nigeria: 6 Fields, 4 Crops

The arc in one frame. The percentages are the client’s own reported estimates from our in-app feedback form; both are reconciled against the advisory data later in this article.

Smallholder farming is the hardest case for any advisory system, and this account is a good example of why. AgroGreen Dynamics – FarmTrack Hub is a farm-management service that tracks plots for growers across several Nigerian states. In the week we examined it was running six fields totalling 23.6 acres, carrying four crops at six different growth stages, split between the humid tropical south and the dry northern savanna. The median field is 4.2 acres. The smallest is half an acre.

In July 2026 the hub completed the feedback form attached to its satellite farm advisory and reported that the reports had helped it optimise irrigation by 30–50%, optimise fertiliser application by 30–50%, and take action on pest, disease and weed pressure. Both of those percentages turn out to be reproducible from the advisory data — and, unusually, through two entirely separate mechanisms. This article shows what the advisory said and where each saving actually sits.

Illustrative. The kind of ground this account monitors: smallholder maize on red laterite soil in the southern wet season, oil palm at the field edge. Representative imagery, not a photograph of the client’s own fields.

One hub, three states, four crops

The portfolio spans two agro-ecological zones that would normally require two different advisers. The Akwa Ibom fields sit in the humid tropical south, among oil palm, on red laterite soils in a long wet season. The Jigawa and Kaduna fields are in the northern savanna belt, on pale sandy soils with a short, sharp rainy season. The hub monitors both from one account, and each field is advised on its own.

Because the plots sit at different longitudes and latitudes across the country, they are not all imaged on the same satellite pass. For a hub that is an operational advantage rather than a complication: fresh guidance arrives for some part of the portfolio on most days of the week, so field visits can be planned against whichever plot was most recently updated. By the date of the feedback the account had received 59 field advisories across 27 satellite acquisition dates. The yield estimates alone show why a single averaged figure would be useless to this hub.

Four crops, six growth stages, two climate zones — and six advisories that share almost nothing with each other.

What the satellite farm advisory delivers per field

Each field gets its own satellite farm advisory per satellite pass: a day-by-day irrigation schedule, per-nutrient fertiliser rates with both organic and chemical sources, pest and disease risk with control options, weed pressure, soil chemistry, and a yield estimate. On this portfolio the two maize fields at V6 and the one at V9 are treated as three different problems, because at those stages they are.

Pest and disease: the right pest, and the right aphid

Across the six fields the advisory raised 15 alerts at 80% probability or above, and each one belonged to the crop it was raised on. The maize fields led with Fall Armyworm (Spodoptera frugiperda) at 90% — the correct headline threat for maize in West Africa — alongside Maize Streak Virus and the stem borer Busseola fusca.

The detail that shows the model is reading the crop rather than the region is the aphids. The sesame field was flagged for Aphis gossypii. The groundnut field, two states away, was flagged for Aphis craccivora — the legume-feeding aphid, and the vector of Groundnut Rosette Disease, which the same report flagged directly beneath it. Two different aphid species, correctly assigned by host, with the disease one of them transmits named alongside.

The maize card: Fall Armyworm (Spodoptera frugiperda), Southern Corn Leaf Blight (Bipolaris maydis), Stem Borer (Busseola fusca) and Maize Rust (Puccinia polysora) — the tropical rust species, not the temperate one. Organic options include neem oil and releasing Trichogramma wasps.

Illustrative. The ragged window-pane feeding and frass in the whorl that a 90% Fall Armyworm alert is telling a grower to go and look for. Representative imagery.

Every alert arrived with both an organic and a chemical option, and the organic ones are locally obtainable rather than aspirational: neem oil for armyworm, Trichogramma wasp release for stem borer, Trichoderma viride for leaf blight. In a smallholder context that matters more than the chemical name, because the chemical is often the option that is out of reach.

Irrigation: every other day, not every day

The irrigation mechanism here rewards a close reading. Every field in the portfolio is on a strict alternate-day pattern: a set depth on one day, zero the next, repeating across the week. Twenty-two of forty-two field-days carry a zero, and the okra plot at continuous harvest is told not to irrigate at all for the entire week.

Against a baseline of irrigating every day at the same depth — the common pattern when water is applied by opening a channel or running a hose, where the dose is however long a pass takes rather than a measured depth — following the schedule applies 42.9% less water. That sits inside the 30–50% the hub reported, and it is achieved without changing a single piece of equipment.

The maize field’s irrigation card. The summary line gives the irrigation-days count alongside the total water — on this portfolio that count, not the depth, is where the saving sits. The method selector shows the method the field actually uses.

The report also discloses headroom the hub could take later. On the fields where the methods differ, drip is listed at 17–25% below the flood figure — 5.0 mm against 6.0 mm on one maize field, 3 mm against 4 mm on the sesame. Flood irrigation is nonetheless what the advisory recommends, because that is the method these fields actually have. The cheaper option is shown as a future step rather than imposed as advice the grower cannot act on today.

Fertiliser: zero nitrogen for the groundnut

The fertiliser section reports five nutrients per field — nitrogen, phosphorus, potassium, sulphur and zinc — each with a modelled current level, the ideal range for that crop, a rate, and both an organic and a chemical source. The groundnut zero is the headline, but the spread across the rest is just as consequential.

Illustrative. Groundnut at peak flowering on sandy northern soil — the crop that was correctly prescribed no nitrogen, because it makes its own. Representative imagery.

Potassium ran from 12.0 to 107.5 kg/acre — a 9.0-fold difference — and the okra plot was prescribed potassium and nitrogen but no phosphorus, sulphur or zinc at all. Summed across the portfolio the advisory prescribed about 2,567 kg of nutrient; applying one uniform average rate to every acre instead would have misallocated 33.6% of it, over-feeding some fields and starving others. That figure also lands inside the hub’s reported 30–50%.

The fertiliser card gives current-versus-target per nutrient and names both an organic and a chemical source. The organic column is locally grounded: composted poultry manure for nitrogen, bone meal for phosphorus, wood ash for potassium.

Reverse-engineering the reported 30–50%

Zero kilograms of nitrogen for the groundnut — because that crop makes its own.

Both reported figures reconcile against the advisory data, which is worth stating plainly because it is not automatic.

Neither figure was reverse-engineered to fit the answer. We computed both from the advisory data first and compared afterwards, and each landed inside the band the hub had already reported. For a buyer assessing this kind of product, that independence matters more than the size of either number: two unrelated parts of the same report were each doing measurable work on the same portfolio, in the same week.

Grower-reported outcomes, as submitted

The cards below reproduce the hub’s answers to our in-app feedback questionnaire for the advisory dated 17 July 2026, exactly as recorded. They are structured survey responses, not paraphrases.

The percentages are the hub’s own estimates from a structured form, not metered volumes or invoiced tonnages. What our analysis establishes is that the advisory they were reading contained instructions capable of producing savings of that size, through mechanisms we can point at directly.

Building soil, not just applying fertiliser

One finding runs across all six fields and shapes the whole nutrient strategy: soil organic carbon sits between 0.10% and 0.15%. On sandy northern soils in particular, organic matter is what holds both water and nutrients in the root zone, so a low reading changes what the right recommendation is. Soil pH came back at a uniform 6.5 and salinity low throughout — both comfortable for these crops.

That is why the advisory’s organic column is not decorative. Composted poultry manure, bone meal and wood ash are named because they are obtainable in these markets and because they build the organic fraction while supplying the nutrient — something urea and muriate of potash cannot do. Wood ash as a potassium source is long-standing West African practice; the advisory quantifies it rather than replacing it.

Illustrative. Sesame in the northern savanna — one of four crops on an account whose fields span roughly a thousand kilometres of Nigeria. Representative imagery.

Why per-field beats per-farm at smallholder scale

Underneath every section sits the same asset: a per-pixel read of each field from satellite, translated through crop coefficients and regional pest models into instructions for that specific plot. The advisory’s own references name the inputs — FAO crop guidelines, regional pest calendars, and Sentinel-1 radar and Sentinel-2 optical indices.

Zone-level crop health for one of the maize fields, read per pixel from satellite. Within-field variation like this is what makes a targeted intervention cheaper than a blanket pass.

For a hub advising growers rather than farming itself, that per-field resolution is the entire product. The question is never “how is the portfolio doing” — it is “which of these six plots needs attention this week, and what exactly do I tell that grower”. On the evidence here, the answer differs on every field: different water, different nutrients, different pest, different stage.

Illustrative. Okra at continuous harvest — the half-acre plot that was told to apply no irrigation at all that week, and only two of the five nutrients. Representative imagery.

Running this on your own farms

1. Map each field separately, and map several. The per-field differences are the product, and one field cannot show you a difference.
2. Set the crop correctly. The nitrogen zero on the groundnut and the species-level pest naming both follow from the crop field being right.
3. Read the irrigation-days count, not just the depth. On this portfolio the interval carried the entire water saving.
4. Compare nutrient mixes field by field. The totals hide the misallocation; the per-field rates expose it.
5. Use the organic column. Poultry manure, bone meal and wood ash are named because they are locally obtainable — and because low organic carbon needs organic matter, not only mineral fertiliser.

Explore the platform at sat.farmonaut.com, read more about our satellite and AI-based farm advisory in the JEEVN AI overview, or talk to our team about onboarding a multi-field operation. JEEVN AI is a paid product; plan detail is on the platform.

Frequently asked questions

Glossary

V6 / V9 (maize)

Vegetative growth stages counted by fully collared leaves. V6 and V9 differ materially in water and nitrogen demand, which is why two maize fields on the same account can get different schedules.

Nitrogen fixation

The process by which legumes such as groundnut convert atmospheric nitrogen into plant-available form via root-nodule bacteria — the reason the groundnut field was prescribed zero nitrogen.

Fall Armyworm (Spodoptera frugiperda)

An invasive caterpillar pest of maize, now established across sub-Saharan Africa. Feeds in the leaf whorl, leaving ragged holes and frass.

Groundnut Rosette Disease

A viral disease of groundnut transmitted by the aphid Aphis craccivora. The advisory flagged both the vector and the disease on the same field.

Soil organic carbon (SOC)

A proxy for soil organic matter, governing water-holding capacity and nutrient buffering. Measured at 0.10–0.15% across these six fields, which is what drives the organic-amendment guidance.

Case-study basis: JEEVN AI field advisories for a single Farmonaut account, taking each field’s most recent advisory as of the client’s feedback submission (portfolio state at 15–17 July 2026, six fields), plus the account’s in-app feedback submission for the advisory dated 17 July 2026. Client organisation name, crops, states, field count and area are reported from the account’s own records; individual grower names and precise field locations are withheld. The 30–50% figures are the client’s own estimates from a structured feedback form, not metered or invoiced measurements; the 42.9% water figure is our comparison against a stated baseline of daily application at the same depth, and the 33.6% figure is the share of the nutrient programme a uniform blanket rate would misallocate. Pest, disease and weed percentages are modelled probabilities, not confirmed infestations. Yield figures are estimates, not guarantees. Imagery: the four platform screenshots and the crop-health zone map are genuine captures of this account’s own advisory and satellite data; the five wider field photographs are AI-generated illustrative imagery, each labelled as such in its caption, and do not depict the client, its growers or their fields. JEEVN AI is a paid product.