Smart Farming
On-farm weather stations
The regional forecast describes a place that is not your field. For frost, spray windows and disease models, the difference is the whole decision.
The case for an on-farm weather station is not that it forecasts better than the meteorological service. It does not forecast at all. The case is that several important decisions depend on conditions at your site, and those conditions differ from the nearest official station by more than people assume.
The three decisions that justify it
Frost. Cold air drains downhill and pools. Across a single farm with modest topography, minimum temperature on a still radiative night can vary by several degrees, which is the difference between an untouched block and a lost crop. A regional forecast cannot resolve this. A station in the cold spot can, and it is the classic justification for on-farm measurement in orchards and vineyards.
Spray windows. Wind speed and direction at the time of application are both an agronomic requirement and, in most jurisdictions, a record-keeping one. A station that logs continuously produces that record automatically and settles the question of whether conditions were compliant.
Disease. Infection models for the major fungal pathogens run on temperature and leaf wetness or humidity duration. These are canopy-scale variables. A model fed regional data produces regional advice; fed local data it can genuinely reduce spray numbers while maintaining control, which is where the economic return sits.
Evapotranspiration is the quiet payoff
A station with temperature, humidity, wind and solar radiation can calculate reference evapotranspiration - an estimate of how much water the atmosphere demanded that day. Combined with a crop coefficient, that becomes an estimate of crop water use.
This is what turns soil moisture sensors from a rear-view mirror into something forward-looking. Moisture sensors tell you what the crop has already used. Evapotranspiration tells you what it is about to use, which is what an irrigation schedule actually needs.
When you do not need one
If the farm is flat, close to a public station, not irrigated, and the crop has no relevant disease model, the honest answer is that a station will produce interesting data and change no decisions. Public networks and free agricultural weather services have improved considerably, and for many arable situations they are sufficient.
The threshold is specific and worth applying: name the decision that will change, and check whether the nearest public station is representative of the place that decision affects. If it is, save the money.
Related
Weather data is one of the two inputs to sensible irrigation automation, the other being soil moisture. Under cover, the equivalent measurements drive greenhouse climate control, where the grower controls the weather rather than only observing it.
Sensor packages by purpose
Partner programmes for this category are not in place yet, so no product links are shown. The comparison is by configuration.
- 01
Basic package - temperature, humidity, rain
Configuration
Enough for frost alerts, growing degree day accumulation and rainfall records. The sensible starting point for most farms.
Strengths
- Lowest cost and least maintenance
- Covers the most common decisions
- Reliable, few moving parts
Limitations
- No wind data, so no spray drift assessment
- Cannot calculate reference evapotranspiration
- Limited disease model support
No partner link for this product yet - the comparison is editorial only.
- 02
Full agrometeorological package
Configuration
Adds wind speed and direction, solar radiation and often leaf wetness. Enables evapotranspiration calculation and most published disease models.
Strengths
- Supports irrigation scheduling by evapotranspiration
- Wind data for spray decisions and records
- Feeds disease models properly
Limitations
- Anemometer and radiation sensors need periodic service
- Higher cost, more to site correctly
- Leaf wetness sensors are notoriously hard to standardise
No partner link for this product yet - the comparison is editorial only.
- 03
In-canopy microclimate sensors
Configuration
Small sensors placed within the crop rather than at standard height, measuring the conditions the plant and pathogens actually experience.
Strengths
- Captures the humid canopy conditions that drive disease
- Cheap enough to deploy in several blocks
- Complements a standard station well
Limitations
- Readings are not comparable to standard meteorological data
- Placement strongly affects results
- Not a substitute for a properly sited station
No partner link for this product yet - the comparison is editorial only.
Frequently asked questions
Is an on-farm station better than a free forecast?
For forecasting, no - a station measures the present, it does not predict. For everything that depends on what actually happened at your site, it is decisively better. Frost risk varies by several degrees across a single farm with slope and cold air drainage, rainfall from a summer storm can differ by half its total across a few kilometres, and disease models need the humidity your canopy experienced, not the airport's.
Where should a weather station be sited?
Over short grass, away from buildings, trees and hard surfaces by at least four times their height, with temperature and humidity sensors in a ventilated radiation shield at roughly 1.5 to 2 metres, and the anemometer as high as practical. A station on a shed roof or beside a track reads the shed and the track. Bad siting is the most common reason on-farm weather data disagrees with reality.
How much maintenance do they need?
More than most buyers expect. Rain gauges block with debris and insects and should be checked monthly in season. Radiation shields fill with spiders. Anemometer bearings wear. Solar radiation sensors need cleaning. A station left unserviced for a season produces data that looks fine and is wrong.