Subscribe and save

Farmers Weekly from £127
Saving £36
In print AND tablet



We are in the process of making some changes to our website, which we are excited to be revealing soon. As part of these changes, the learning content provided by the Farmers Weekly Academy will soon be moved to the main site. If you have any queries please email

Fertiliser 2: Remote sensing

Course: Fertiliser management | Last Updates: 9th October 2015

Clive Blacker
Precision Decisions
Biography >>

Rising fertiliser prices and increasing environmental burden mean the pressure to apply nitrogen fertilisers correctly and keep accurate records will continue to grow.
Good farm management is practised on many farms and growing numbers are now adopting precision technology to map variability and apply inputs.
Variability can be due to a number of aspects including soil type, previous cropping, historic manure use, with nitrogen availability being the dominant feature. Soil variability inevitably reveals itself within the crop as variable biomass and colour, both of which can be measured using sensor and reflectance technology.
Real time measurement of the canopy during the crop's life has clear advantages and while growers have been encouraged to adopt canopy management techniques using Green Area Index; there have been few tools available to take actual measurements. Remote sensing offers this capability.

What is remote sensing and how does it work?

The principle of remote sensing is relatively simple. When light is emitted onto a surface, some light is reflected (as electromagnetic radiation) and some is absorbed.
The reflected light will be across the whole electromagnetic spectrum, which is far wider than we can see. The spectrum can be split into visible (blue, green and red) and invisible (near infrared, infrared and beyond) wavelengths. Each object (eg. crop) has its unique reflectance spectrum and as a result different objects are seen in different ways.
Cereal crops look green because they reflect the green light and absorb the other light wavelengths. They also reflect light more strongly from the infrared part of the spectrum.
The human eye can't detect this, but an infrared sensor can. These sensors have been developed specifically to measure crop reflectance at different levels, to help farmers target inputs more effectively. Broadly speaking, the near infrared wavelengths are a measure of the canopy size, while the visible bands are an indicator of colour. Bringing the two together gives a total green area measurement.

What sensors are available?

There are several sensors available. Satellites have been operational since the 1970s, and ground-based sensors have been in development since the early 1990s. So how do these sensors differ, and what should users know about the technology?
The main difference between sensors is in their ability to collect information. This can be judged in three ways:


This is the size of the parcel of information collected, and therefore the area of the field the sensor sees. Satellites can capture large amounts of information quickly, but the resolution can be poor.
Pixels determine the resolution. A pixel represents a fixed area or footprint on the ground. These vary, but the most common used for agriculture is one measuring 32m x 32m and therefore looks at a very large area.
Aeroplanes again offer the ability to capture large amounts of data. The amount and quality depends on the height of the plane. A clear limitation to aeroplane and satellite sensing is the inability to capture accurate data in cloudy conditions.
Tractor-based systems can capture information at a far higher resolution than satellites due to their proximity to the crop. Current application technology does not allow us to vary product rates across the bout width so the key element is that the image captured represents the working bout width, with the ability to sense and respond to a crop requirement before the applicator has passed it.


As the sensor is required to measure light reflectance from the crop, it is important to do this accurately. Passive sensors use the sun to light the crop and measure the subsequent reflectance. Sun angle and cloud cover affect these types of sensor.
Some passive systems have the ability to correct for climatic changes during image capture. There are also sensors that are unaffected by the ambient daylight as they emit their own light to illuminate the crop.
Sensors that can only capture information during daylight are classed as passive sensors, while those that emit their own light are termed active.

Wavelength measurement

Some sensors can only capture a limited number of wavelengths, while others have the ability to capture a wide number.
Generally active sensors can only capture a small number of wavelengths due to the difficulty of finding a bright enough light of the required colour for multispectral measurement. Typically active sensors measure between 2 and 4 wavelengths.
Passive, satellite or plane sensors usually measure 4-7 wavelengths while the passive tractor system used by Yara can capture up to 60 wavelengths.

Tractor spreader

How are these used in practise?

There are distinct differences as to how the various systems are used by farmers. The biggest difference is the timeliness of data collection and application. All planes, satellite and some tractor-mounted systems offer near real time application, which requires data to be captured correctly, positioned and then processed. The information collated is then run through a model to generate a map or recommendation. This data processing takes time and can be restrictive, and not relevant to the current situation.
One system – Yara's N Sensor – offers a real time solution where data collection, interpretation and application happen during the same pass through the crop.
All systems are used during the active growing season at each nitrogen application. The satellite/plane systems produce an actual application map with a nitrogen recommendation associated with the scanned areas, while the tractor-mounted alternatives require crop calibrations or office based and time consuming input, around which the nitrogen rate will be varied. New developments with Yara’s N Sensor (Absolute-N) allow the sensor to calculate the crop requirement and apply according to this demand, removing the need to calculate the optimum beforehand.
All systems will work on cereals, while some can be used on oilseeds and potatoes for nitrogen management. Precision application can be used for liquid and solid systems, and dont have to be limited to nitrogen, Growth regulators, desiccants and seed rates are targeted by Yara N sensor users .

What are the advantages of using remote sensing technology?

The advantages are associated with either over or under supply of nitrogen fertiliser. Undersupply will reduce yield and quality, while oversupply can cause lodging, increased disease, late maturity and poor combine performance (10-20% reduced output).
Very little replicated data is available other than that carried out by Yara and HGCA, which have both shown the financial benefits, the former typically quoting a 3-4% yield increase with no extra fertiliser used. The breakeven position for moving to variable application is in the range 140-200ha (350-500 arable acres), depending on farm variability.

Case study – Absolute-N in practice

For Essex farmer and contractor Tom Bradshaw the introduction of the Absolute-N mode has helped further his understanding how to efficiently manage the canopy of oilseed rape. “A canopy’s nitrogen content can be highly variable even in what looks like a consistent crop because nitrogen availability can vary widely, but the N Sensor recognises this and tailors applications to match plant needs.”
Tom admits the introduction of the N Sensor has led him to review conventional practice. “We’re almost rethinking our nitrogen policy. Human nature suggests it is better to err on the side of caution and apply too much rather than not enough, but the N Sensor removes that uncertainty. “It allows highly flexible timing and gives the grower confidence to delay applications until spring growth has commenced.”
Based on conventional assessment of plant requirements in 2011 he estimated that his 116.7ha of oilseed rape needed 300kg N/ha, but the N Sensor disagreed. At 146kg N/ha the rate applied was less than half the initial assessment yet the crop yielded 4.84t/ha after drying.
“We have a yield expectation of 4t/ha and a target of 5t/ha. We’re getting there steadily and the N Sensor has a vital role to play in the journey.”

Please login or register to take this test.