Wageningen University was built over 100 years on the basis of traditional agricultural research, but the advent of sensors and big data could shake the traditional institutes. Will every farmer soon have his own experimental farm?
The university recently celebrated its 100th anniversary and thus 100 years of high-quality agricultural research. In the 50s to the 80s, CT de Wit laid the foundations for the relationship between photosynthesis and crop production, but also for the application of fertilizers and the mixing of different crops that positively influence each other.
This kind of scientific research was built up in the second half of the last century through an enormous amount of measurements, theoretical models and ultimately the development of advice that agriculture still uses today.
Sensor as data source
The data required for this research is often collected on experimental fields with different soil types, varieties, doses of fertilizers or crop protection products and often on experimental farms. However, in recent decades we have seen that digitization is taking over part of this data collection and knowledge development. Sensors make it possible to collect data on a large scale and to develop insights that we as humans cannot foresee.
Automatic control of planters and seeders, fertilizer spreaders and sprayers via task cards makes it possible to effortlessly set up field trials. The yield sensors can automatically record the results of these tests. For example, every grower is a researcher on his own farm and the costs of research are drastically reduced. In addition, the results become much more relevant, because everyone has research results from their own company, instead of those from an experimental farm. There are also more and more possibilities to use self-learning algorithms for data analysis.
American practice
An example comes from the American Climate Corporation, part of Bayer. This company recently launched a digital seed adviser launched. By using smart algorithms on a database of soil maps, data from seeders and yield maps from combines, they provide automated advice per soil type for the best maize variety. No agricultural experiment has been carried out here. This was developed on the basis of data collected from 200 growers on 40.000 hectares, with data that was often already available. Self-learning algorithms have determined a strategy that yields the highest profit for each type of soil.
At Droneworkers we are also continuously creating new insights based on data (in collaboration with chain partners and advisors). We take soil samples for the season based on soil and elevation maps. Subsequently, the emergence of each plant is mapped and we can follow the growth pattern during the season, supplemented with soil samples. We use thermal imaging cameras to visualize the moisture balance and predict the yield towards harvest. We use new insights from this big data to build up new knowledge and improve existing advice.
Colored advice
For you as a grower, this means that new insights become available at a faster pace about the best strategies. Current cultivation advice is often very general. Depending on the situation, a dose is too low or causes damage. Advice can also be colored by the seller or advisor, but also by the way of testing on which the advice is based. Tests often have limited relevance for practice, while data from practice overcomes this problem.
Farmer wisdom will always be necessary to run the business properly, but business operations will more often be supported by new forms of guidance based on data. Not only for cultivation, but also from the accountant, the government and together with colleagues.
Traditional research
If we can visualize so many new aspects of cultivation and develop new knowledge with the data, at much lower costs and with a shorter development time, is there still a role for traditional agricultural research? Or does the new generation of researchers know how to seize this as an opportunity to develop new knowledge together with the sector?
© DCA Market Intelligence. This market information is subject to copyright. It is not permitted to reproduce, distribute, disseminate or make the content available to third parties for compensation, in any form, without the express written permission of DCA Market Intelligence.
This is in response to it Boerenbusiness article:
[url=http://www.boerenbusiness.nl/column/10880970/is-landbouwkundig-onderzoek-nog-nieuwe]Is agricultural research still necessary?[/url]
Anyone involved in precision farming would do well to delve into the theory of the "flat payoff function". This theory was developed by Prof. David Pannell of the University of Western Australia. The core of his story is that many control variables (eg planting distances or fertilizer application) have hardly any influence on the financial yield over a wide range. The consequences of this for many forms of precision agriculture are major: they yield hardly anything.
Precision farming is actually a kind of icing on the cake, while the cake is still far from in order for most arable farmers. Better to spend your money on "the cake" than on "the icing". It simply works better there.
We need the independent agricultural research in my opinion more than ever to help us take this "pie" to the next level.
Someone who preaches homeopathy by definition does not understand what thorough and honest research is.