23 January 1999

EYES IN THE SKY

Move over, crop walking? Aerial sensing is the modern way to keep a watchful eye on your fields, as Gilly Johnson reports.

BEFORE this summer, Keith Norman of farm manage-ment company Velcourt was convinced that the best way to judge crop health was the traditional, time-consuming meth-od. That meant walking the field, parting the plants and checking out growth by eye.

But now Mr Norman has had a chance to weigh up the potential for the space-age alternative – using images of the crop from the sky instead of field walking. And his views have changed.

"Some might think that remote sensing technologies are too sophisticated to be any practical use. Its true that its complex, but at the right price it could be the basis of a viable crop management tool for the future," he concludes.

Remote sensing is able to exploit near infra-red wavelengths which are not visible to the naked eye, and so can pick up crop differences that might escape even the keenest-sighted grower. Developments in sensor technologies make it possible for sensors to focus down to 4 sq metres.

Mr Norman has put remote sensing to the test on a Velcourt farm in Cambridgeshire this season, under the PAAGE project (the Pilot study for Agriculture and AGri-Environment), which is funded by the EU with one million ecu from its enviromental budget, and managed in the UK by the National Remote Sensing Centre.

The idea was to calibrate the different images of a crop generated either from sensors mounted on aircraft, or from space via satellites. Then by interpreting a series of these images, transformed into colour coded charts, a grower could work out the status of the crop for nutrition, health, growth and possibly pest factors.

The potential for this technology hinges on whether the remote sensing is able to pick up and quantify small differences in crop status that might, or might not be visible to the naked eye. This seasons results show that crop performance can be monitored accurately.

Mr Norman split a field into 30m long, 9m wide plots and varied the crop management for each section. Different seed rates, fertiliser doses, fungicides and growth regulator strategies were applied, and images across all the plots were taken at key timings: first node – GS31 (27 April), flagleaf – GS39 (15 May), ripening – GS59 (15 July) and pre-harvest (5 Aug).

Colours of health

The resulting colour codes are shown in the diagram. Different shadings can be found to tie in with variations in crop status that on occasions were not visible to the naked eye in the field. The near infra-red wavelengths were tuned to pick up crop characteristics: leaf dry mass, leaf brown pigments, leaf area index, chlorophyll index and soil brightness, which is related to soil type and moisture holding capacity. These need to be taken together to give a meaningful interpretation.

For example, a thin, but disease-free crop growing vigorously at a low seed rate might give a low leaf mass index and brown pigment rating, but a high chlorophyll content and high soil brightness. Chlorophyll content is related to nitrogen status, and brown pigment rating would increase with the growth of disease lesions on the leaves.

"The results cant be looked at in isolation," says Mr Norman. "Its important to see them together. A low chlorophyll content might indicate low nitrogen – but it might also be because the plant isnt taking up nitrogen because of take-all or some other soil related problem. To interpret them correctly, all factors must be studied."

This is the most difficult part of the project; working out what the management response ought to be, to a certain combination of colour codes.

But the potential is there. In the commercial crop surrounding the trial plots, nitrogen rates were about 45kg/ha (18 units N/acre) too high for cost-effective yield, according to Mr Normans calculations from the imagery. If this had been spotted earlier, and remedied, with the help of remote sensing not only would there have been a cost saving per hectare, but also optimum nitrogen would have been applied to that specific field so reducing any potential environmental problems resulting from applying too much.

For the purposes of this trial, the sensor was mounted on an aircraft. But in future the pictures could be taken by satellite, says Meryl Strang, agriculture specialist with the National Remote Sensing Centre. Although cloud cover does limit the opportunities for clear imaging, the satellites pass over so frequently that cloud is not a practical problem.

Last but not least, cost. If remote sensing is to take off, then it has to be affordable, says Mr Norman. What might he pay for a series of field images? "Figures being discussed are in the range of £10/ha for a seasons series of images. Given the potential of the technology, that could be worthwhile."

European trials

France and Spain are also conducting trials as part of the PAAGE project. In France the emphasis is on assessing correct nitrogen usage and minimising leaching; in Spain, water use is the priority.

EU officials now must decide whether to take the PAAGE project further and supply more funding. One factor which might sway a decision in favour is if PAAGE shows that it could help EU growers meet environmental objectives by tightening up integrated crop management.

PAAGE has attracted significant interest in Brussels, and the next stage is to evaluate its commercial applications. Further trials are expected to be commissioned shortly.

&#8226 PAAGE falls under the Matra Marconi Space XSTAR umbrella. XSTAR is an intiative to launch an agriculture dedicated satellite in 2002.