Authority guide · IoT sensor guide

IoT Sensors for Agriculture: Types, Placement & Farm Use

IoT sensors for agriculture are connected field devices that measure conditions such as soil moisture, weather, leaf wetness or plant status and send repeated observations to a logger, gateway or software platform. Their value comes from linking each measurement to a real farm decision, installing sensors in representative locations, maintaining the equipment and checking readings against crop and field conditions before acting.

Sources reviewed: 2026-09-30

What makes an agricultural sensor an IoT sensor?

An agricultural sensor becomes part of an IoT system when measurements are collected repeatedly and moved through a data path that can include a logger, low-power radio, gateway, cellular or other network connection, cloud or local software, and a dashboard or decision tool. USDA-NIFA describes modern agriculture as using sensors, devices, machines and information technology together, while USDA-ARS has demonstrated wireless field networks that combine soil-moisture and weather observations for irrigation management.

The important distinction is not the marketing label “IoT.” A useful system closes the loop from measurement to interpretation to a specific farm action. High-frequency data that are poorly placed, uncalibrated or never reviewed can create more noise rather than better decisions.

Sensor: measures a physical or biological condition at a known place and time.
Logger or node: records readings and may perform basic processing before transmission.
Gateway or network: moves data from field devices to a local or remote platform.
Software layer: stores, charts, compares or combines observations with other farm information.
Decision process: connects the measurement to irrigation, scouting, disease-risk, input or other management decisions.

Start with the farm decision, not the device

Precision agriculture works best when data collection is tied to a location- or time-specific management question. Before buying hardware, define what decision the measurement should improve: when to irrigate, whether the root zone is drying, whether rainfall actually reached the field, whether disease-favorable wetness is persisting, or whether an unusual reading deserves scouting.

This prevents a common failure mode in smart-farming projects: installing many sensors because they are available, then discovering that the farm has no threshold, workflow or responsible person for turning the data into action.

Write the decision first: “What will we do differently if this value changes?”
Choose a measurement that is directly relevant to that decision.
Decide where and how often the value must be measured to represent the management zone.
Define who checks the data and what field observation will confirm an unusual reading.

Soil-moisture sensors are among the most practical farm IoT tools

Soil-moisture systems can measure volumetric water content or estimate soil-water tension. University of Minnesota Extension recommends representative locations and multiple depths, commonly around one-third and two-thirds of the active root zone for irrigation monitoring, because a single shallow sensor can miss what is happening deeper where roots are extracting water.

The reading still needs soil and crop context. Capacitance and other electromagnetic sensors can be affected by factors such as salinity, clay content, temperature, bulk density and installation quality, and site-specific calibration may improve accuracy. Data loggers and remote retrieval make monitoring easier, but connectivity does not remove the need for good installation and interpretation.

Use more than one representative location when field variability is meaningful.
Separate substantially different soil textures or management zones when their water behavior differs.
Avoid wheel tracks, leaks, depressions or unusual spots unless those areas are the specific target of investigation.
Compare multiple depths to see whether water is staying shallow, reaching the root zone or moving below it.

Weather stations provide local context that field sensors alone cannot

On-farm weather stations can add local air temperature, relative humidity, rainfall, wind and other observations to crop and irrigation decisions. A local station can be more representative than a distant weather source, but only when it is sited and maintained correctly and its readings remain plausible.

Penn State Extension recommends checking weather-station components before the disease-management season and comparing questionable values with another nearby station. This is an important IoT principle: automated data should be quality-checked rather than trusted solely because it arrives continuously.

Inspect moving parts such as wind sensors for dirt or mechanical restriction.
Check rain and other station components for blockage or damage according to the manufacturer.
Compare suspicious readings with a nearby trusted station or manual observation.
Document maintenance so sudden data changes are not mistaken for real field changes.

Leaf-wetness sensors can support disease-risk monitoring

Leaf-wetness sensors estimate whether moisture is present on a sensing surface intended to represent crop-canopy wetness. They can be useful as one input in disease-risk or infection-period models because many plant pathogens are influenced by how long foliage remains wet together with temperature and humidity conditions.

These sensors require maintenance and context. Penn State Extension specifically advises inspecting leaf-wetness sensors for damage or corrosion. Placement should represent the canopy condition of interest rather than a location that is systematically wetter or drier than the crop.

Use leaf-wetness duration as a risk input, not proof that a disease is present.
Pair wetness observations with crop susceptibility, temperature, scouting and local disease guidance.
Inspect the sensing surface and replace damaged or corroded sensors.

Plant sensors are promising, but distinguish research tools from routine farm equipment

USDA-NIFA highlighted 2026 research on low-cost sensors attached to leaves, inserted into stalks and placed in soil. The research devices measure signals including humidity, temperature, bioelectric activity, nitrate and soil-water tension and use low-power wireless communication to send observations to a gateway.

This illustrates where agricultural IoT is heading: more direct measurements of plant response rather than only the surrounding environment. It does not mean every experimental sensor is commercially mature or that one reading can prescribe water or nitrogen on every farm. Farmers should separate demonstrated research capability from products that have been validated for their crop, field and management use.

Connectivity, power and data logging are part of sensor reliability

A field IoT system can fail even when the sensing element is accurate. Weak radio coverage, gateway outages, depleted batteries, solar charging problems, damaged cables, clock drift or interrupted logging can create gaps that look like agronomic changes. Wireless systems therefore need basic operational monitoring in addition to agronomic interpretation.

USDA-ARS field work on wireless soil-moisture and weather networks and current NIFA research using low-power radios and gateways show the practical architecture: sensors collect observations locally and communication hardware moves those observations to a place where they can be reviewed. The system should make missing or stale data obvious rather than silently presenting the last value as current.

Record the timestamp of every observation and flag stale data.
Monitor battery or power status where hardware supports it.
Design around real field connectivity rather than assuming continuous internet access.
Keep a fallback field-check method for decisions that cannot wait for a network repair.

Calibration and data-quality checks matter more than dashboard polish

The most useful sensor system is one whose limitations are understood. University Extension guidance emphasizes installation quality, representative placement and calibration considerations for soil-moisture sensors, while Penn State recommends cross-checking weather observations when values appear wrong. These practices are more important than adding more charts to a dashboard.

Build a simple quality-control routine: know the plausible range for each sensor, compare abrupt changes with rainfall or field events, inspect equipment after maintenance or field operations, and confirm surprising values manually before changing irrigation, fertilizer or crop-protection plans.

Treat impossible or abrupt values as a data-quality question before treating them as a crop event.
Keep sensor model, installation depth, location and calibration notes with the data.
Use nearby sensors, manual checks or trusted external observations to investigate anomalies.
Do not hide missing data by automatically carrying old readings forward as if they were current.

Where AI fits into an agricultural IoT system

AI can help summarize time-series data, combine several sensors, identify unusual patterns and prioritize what a farmer should inspect. In precision agriculture, the larger goal is to relate high-resolution observations to actions that vary by place or time rather than simply collect more data.

AI should not conceal weak inputs. A model trained on clean data cannot correct a sensor installed in an unrepresentative location or a rain gauge that is blocked. A good decision-support system keeps the underlying observations visible, indicates uncertainty or missing data, and lets farmers compare automated interpretations with field reality.

Use AI to organize and prioritize observations, not to erase the underlying measurements.
Keep timestamps, locations and sensor identity available for traceability.
Combine sensor patterns with crop stage, weather, field history and direct scouting.
Escalate unusual or high-impact decisions to local agronomic expertise when the data do not agree.

Common IoT sensors used in crop production

The best sensor depends on the decision being supported. The table below summarizes common examples and the main interpretation limit that should be considered before acting on the data.

Sensor	What it measures	Typical placement	Decision it can support	Main limitation
Soil moisture / water-content sensor	Volumetric water content or a related estimate of soil water	Representative root-zone locations at one or more depths	Irrigation timing, root-zone depletion and post-irrigation verification	Readings depend on soil, installation, depth, calibration and the area represented by the probe
Soil-water tension sensor	How strongly water is held by the soil	Representative root-zone depths matched to crop and soil	Irrigation trigger decisions and tracking drying trends	Useful operating range and interpretation vary with soil and sensor technology
On-farm weather station	Local temperature, humidity, rainfall, wind and other weather variables depending on the station	Open representative site with correct exposure for the station components	Water-balance context, heat or weather-risk monitoring and local field records	Poor siting, blockage, mechanical problems or sensor drift can make continuous data misleading
Leaf-wetness sensor	Presence or duration of surface wetness on the sensing element	Representative crop-canopy position according to the disease-monitoring purpose	Disease-risk models and scouting prioritization when combined with temperature and crop context	Wetness indicates a favorable condition, not proof of infection, and the sensing surface requires maintenance
Plant wearable or in-plant research sensor	Plant-level signals such as temperature, humidity, bioelectric activity or nitrate depending on the device	Leaf or stalk on representative plants in the monitored area	Research and emerging direct monitoring of crop water or nutrient response	Many technologies are still research-specific and should not be treated as universal commercial prescriptions

Field checklist

The farm decision the sensor should support is written down before hardware is selected.
Sensor location represents the soil, crop, irrigation or microclimate zone being managed.
Installation depth and method are recorded and follow the sensor manufacturer and agronomic purpose.
Calibration or field baseline has been established when the sensor technology requires it.
Logger, battery, power, gateway and communications status can be checked separately from the agronomic value.
Missing or stale readings are visibly flagged instead of being presented as current data.
Unexpected values are compared with another observation before a costly management change is made.
A person is responsible for reviewing the data and connecting it to the farm decision.

Related practical guides

irrigation scheduling guide

Related AgroAdvisor resources

Sources and further reading

This authority guide uses a controlled source registry. Consult original sources when local design, field calibration or a high-impact farm decision requires more detail.