How Drone Mapping Improves Crop Yields in Costa Rica

By the time a stressed section of a field looks yellow, wilted or thin from the ground, the plant has already been struggling for days or weeks — and the yield in that zone is already reduced. That gap between when a problem starts and when a person can see it is the single most expensive thing about managing a farm at scale, because it turns manageable stress into lost harvest. NDVI and multispectral drone mapping close that gap. By capturing light wavelengths the human eye can't see, a drone flight reveals exactly which zones of a field are struggling — often one to three weeks before the same problem would be visible on a walk-through — giving a farm manager time to irrigate, treat or fertilize before the damage is done.

This guide explains what NDVI actually measures, how the data becomes a decision a farm manager can act on, which Costa Rican crops benefit most, and what a mapping programme costs. We already cover the mapping and boundary side of farm surveying in our agriculture drone survey guide and our precision farming & crop mapping overview — this post goes deeper on the specific technology that turns aerial imagery into a yield-protection tool. If you already know your farm and want a number, skip straight to a free same-day quote.

What NDVI actually measures

NDVI stands for Normalized Difference Vegetation Index, and it works because of how a leaf's cell structure interacts with light. Healthy, actively photosynthesizing plant tissue absorbs most red light for photosynthesis while strongly reflecting near-infrared light, which is invisible to the human eye but easy for a camera sensor to record. As a plant comes under stress — from drought, root damage, disease or a nutrient shortfall — that internal cell structure begins to break down before any visible symptom appears, and the ratio between reflected red and near-infrared light shifts measurably.

A drone carrying a multispectral camera captures both bands on every pass over the field. A processing pipeline compares them pixel by pixel and outputs a colour-coded map — typically red or orange for stressed vegetation shading through yellow to deep green for vigorous growth — at a resolution fine enough to separate individual plant rows. The result is not a photograph of the field; it's a measurement of plant health across every square metre, something no amount of staring at RGB imagery or walking the rows can replicate.

From map to action: the problems NDVI catches early

Irrigation stress. Uneven water delivery is one of the most common — and most fixable — sources of lost yield, and it's usually invisible until a dry patch is already scorched. An NDVI map flown mid-cycle typically shows irrigation gaps as clearly defined low-vigor zones that trace back directly to a clogged emitter, an uneven pivot arc or a low spot that drains too fast, letting a farm manager fix the mechanical problem instead of just watching the symptom.

Pest and disease outbreaks. Coffee leaf rust, banana Panama disease and pineapple root rot all reduce a plant's photosynthetic efficiency before the visible symptoms — lesions, wilting, discoloration — spread widely. Because NDVI is sensitive to that early physiological change, a recurring flight schedule can flag an outbreak while it is still a handful of plants rather than a spreading section of the field, when treatment or removal is still cheap and effective.

Nutrient deficiency and variable-rate fertilizer application. Nitrogen deficiency in particular shows up clearly in NDVI well before the classic pale-green leaf colour is obvious on the ground. Mapped against the field, that data lets an agronomist target fertilizer to the zones that actually need it — a variable-rate application — instead of a single blanket rate across the whole farm, which both protects yield in the deficient zones and avoids over-fertilizing the healthy ones.

Replanting and gap analysis. For row crops and orchards, an NDVI map combined with a standard orthomosaic (the kind of imagery we describe in our deliverables guide) quickly highlights missing or failed plants across a large area, turning what would be a slow manual count into a map a crew can work from directly.

Drone NDVI vs. satellite imagery: why resolution wins

Free NDVI layers from satellites like Sentinel-2 are useful for a quick regional glance, but they are a poor tool for actually managing a farm. Sentinel-2's public bands resolve to roughly 10 metres per pixel — a single pixel can cover dozens of individual coffee plants or a wide stretch of a pineapple bed — and revisit intervals of five days or more are frequently defeated entirely by cloud cover, which is a serious limitation in a country where the green season delivers heavy cloud for months at a time. A drone flight, by contrast, resolves to a few centimetres per pixel, flies below the cloud deck on demand rather than waiting for a satellite pass, and can be scheduled around irrigation cycles, spray windows or a specific disease scare rather than an orbit. The trade-off is coverage per flight — satellites win on sheer area — but for the field-level, plant-row decisions that actually move yield, the extra resolution and the ability to fly on your own schedule are what make the data usable at all.

What's behind the map: sensors, bands and consistency

Not all "NDVI drones" measure the same thing. A modified consumer camera that swaps a filter to catch a rough near-infrared signal can produce a usable stress map on a budget, but a proper multispectral payload — capturing red, green, red-edge and near-infrared bands separately and simultaneously — gives a materially cleaner signal and unlocks additional indices beyond basic NDVI, such as NDRE (Normalized Difference Red Edge), which is more sensitive to nitrogen status in dense, mature canopies like coffee and banana where standard NDVI tends to saturate. We select the sensor to match the crop and the question being asked, rather than defaulting to one tool for every job.

Consistency between flights matters as much as the sensor itself. Light conditions change hour to hour and day to day, so every flight is calibrated against a reference panel of known reflectance before and after the mission, which keeps NDVI values comparable across visits — without that step, a shift in sunlight can look like a shift in plant health and send an agronomist chasing a problem that doesn't exist. That same discipline around repeatable, ground-controlled data collection is what we apply across all of our survey work, detailed in our accuracy guide.

Costa Rica's crops and climate: where this matters most

Costa Rica's growing conditions make crop health mapping especially valuable. The country's coffee farms sit on steep, terraced hillsides where walking every row to check for leaf rust is slow and where microclimates can vary sharply within a single finca — exactly the kind of variability an aerial NDVI pass reveals at a glance. Pineapple and banana operations in the Caribbean and northern zones run at a scale where root disease can spread through a block before a scouting crew reaches it on foot; we cover the regional side of that work in our Caribbean agriculture drone survey guide. In Guanacaste, where dry-season irrigation scheduling is the difference between a strong and a weak harvest, NDVI is most often used to catch irrigation-system problems early — see our Guanacaste agriculture drone survey guide for the regional detail. And because Costa Rica's green season delivers intense, localized rainfall, drainage-related stress zones tend to appear and disappear quickly — another reason a single flight is far less useful than a recurring monitoring cadence.

How a monitoring programme works, and what it costs

A single NDVI flight is a snapshot; a farm's real value comes from comparing maps over time. Most commercial operations we work with fly every two to four weeks through the active growing season, tightening to weekly during flowering, fruit set or a known disease-risk window, and adding an extra flight after any major storm. That cadence lets a farm manager see not just where a problem exists today, but whether a flagged zone from the last flight is recovering, holding steady or spreading — the trend line is often more useful than any single map.

Every flight is captured with ground control for consistent, comparable geolocation between visits, following the same accuracy standards we set out in our survey accuracy guide. Deliverables typically include the raw NDVI raster, a classified stress-zone map ready to hand to an agronomist or field crew, and a standard high-resolution orthomosaic for visual reference — all in formats (GeoTIFF, shapefile) that drop straight into whatever farm-management or GIS software you already use.

Cost scales with farm area, flight frequency and whether you're running a single assessment or a recurring monitoring programme; as with our other survey work, a recurring schedule is priced more cheaply per flight than a one-off visit, and we break down the general cost drivers behind all our survey work in the 2026 cost guide. The fastest way to a firm number for your farm is the online quote calculator, which returns a free, same-day estimate based on your location and area — or send us a WhatsApp message describing your crop and acreage.