The drive to increase yields
The challenge facing agriculture of producing more whilst impacting less is real and immediate, and requires ever closer engagement of researchers and farmers. These two communities often work separately, at different scales, on different aspects of crop performance. Also knowledge is often seen to flow one way, from the researcher’s lab to the farmer’s field. However, extrapolation from fine-scale scientific research to broad-scale field application often entails untestable ‘leaps of faith’. Meanwhile, farmer-led innovations are often highly effective, yet are rarely tested scientifically, so may not be widely trusted or disseminated amongst the research and farming communities.
Where does innovation come from?
‘Best practice’ is not a rigidly defined recipe, rather it continually evolves through recent innovations and experience. Any analysis of agricultural progress reveals that it is the farmers, or those close to farms, who make the most numerous and telling innovations1. Farmers are practical experimentalists who continually innovate, test and adapt agronomic practices, cultivations and technologies, but this is often unrecognized by formal science. Richard Holbrook’s experiments2 on his Northamptonshire farm in the 1980’s are a rare exception.
Most large arable farms now use farm management software to record cropping information, and an increasing proportion of arable farms utilize precision farming technologies to monitor and treat their crops3. Yield monitors are ubiquitous now on modern combine harvesters, giving farmers instantaneous measures of yield during harvest and yield estimates by field.
What is Agronōmics?
Agronōmics is the science of field-scale cropping: it studies and compares spatially-defined cropped areas by gathering, analysing and interpreting multiple spatially-defined data, often with farmers using ‘precision farming’ technologies.
Although not yet widely recognised, the most valuable attribute of precision farming technologies is their capability to assess the effects of management decisions. On-farm testing is often carried out in ad hoc tramline or split field comparisons; now this has been made easier with the availability of new technology. Such trials tend to be ignored by scientists as their lack of randomisation and statistical analysis makes them ‘unscientific’. However, the huge replication provided by yield monitors and crop sensors could provide a new way of making credible high-precision farm-scale comparisons (Figure 1).
Figure 1: A farmer’s yield map, showing the challenge of deciding whether low (L) and high (H) nitrogen levels really differed in yield from the standard (S) level.
What is needed are new statistical techniques and analytics that realise the benefits from the new ‘big’ datasets. Thus the immediate and vital challenge for Agronōmics is to integrate effectively the various data sources currently available (e.g. soil, weather, crop sensing, satellite sensing, historic yield maps and imagery), and to enable statistically robust comparisons so that meaningful results can be extracted that are of value both in practice and in science. We suggest that the essential components of effective Agronōmics systems are;
- Accepted explanatory concepts
- Precise on-farm machinery, capable of making variable applications, and spatial records
- Data management software
- New spatial statistical techniques
- Motivated and coordinated networks of farmers
ADAS is now developing these components using funding from Innovate UK, collaborating with the British Geological Survey, AgSpace, BASF, Trials Equipment Ltd. and VSNi. In particular we have been developing farmer networks, harvesting protocols and machinery, software and spatial statistics that will enable farmers and researchers to obtain robust results from tramline-scale treatments.
Engaging farmers with Agronōmics
The vision for Agronōmics is of facilities, techniques and infrastructure (virtual, web-enabled networking) whereby increasing numbers of interested farmers with yield mapping capability can easily club together to make coordinated tramline comparisons, to address their most compelling questions. For the existing research community initial farm-research networks are already augmenting results from conventional small plot experiments to provide additional broad-scale perspectives. Examples of current initiatives are:
- The AHDB LearN project, working with 18 farms using simple tramline comparisons to assess the main causes of variability in N requirements, hence to make their N use more efficient
- To support marketing of hybrid barley varieties, Syngenta organised a series of reference fields where their hybrids’ performance is compared with conventional varieties in split-field experiments.
- Within the Yield Enhancement Network (www.yen.adas.co.uk) groups of farmers are testing the value of various products in keeping crops alive for longer and increasing yields
- With Sainsbury’s and the Camgrain Wheat Development Group, satellite information, NIR sensing technologies and tramline trials were combined to find ways to improve grain protein and ultimate functionality.
The ADAS Digital approach to food production and environmental management
Agronōmics was presented at the 12th European International Farming Systems Association Symposium, held at Harper Adams in July 2016. The associated paper can be downloaded here, which includes further details on the agronōmics approach to farm-based research.
Agronōmics is part of a wider development of digital technologies across ADAS. The recently launched ‘ADAS Digital’ is a new initiative that aims to coordinate the delivery and communication of all of our projects that utilise data science or digital technology.
You can read more about the ADAS Digital initiative here, and full details of our digital services can be found on our new ADAS Digital webpage.
If you have ideas about topics for a new farm research network or would like to know more about Agronōmics, please contact Daniel Kindred; Daniel.Kindred@adas.co.uk.
If you would like to know more about ADAS Digital and our associated services, please contact Lucy Wilson; Lucy.Wilson@adas.co.uk.
- Sylvester-Bradley, R. (1991). Modelling and mechanisms for the development of agriculture. Aspects of Applied Biology 26, The Art and Craft of Modelling in Applied Biology, 55-67.
- Holbrook, J.R., Byrne, W.R. & Ridgman, W.J. (1983). Assessment of results from wheat trials testing varieties and application of nitrogenous fertilizer. The Journal of Agricultural Science 101, 447-452.
Defra. (2013). Farm Practices Survey 2012: https://www.gov.uk/government/statistics/farm-practices-survey-october-2012-current-farming-issues