Skip main navigation

Plan, monitor, analyse

An article explaining how planning, monitoring crops and analysing data are all key to ensuring success with precision farming.
Aerial photograph of a tractor with sprayer in large precisely planted field
© EIT Food

Precision farming has been described as ‘a means of increasing the chance that the right crop management strategies are implemented in the right place at the right time’ [1]. But the techniques aren’t magic. To be successful, they need to be applied consistently and thoughtfully with constant planning, monitoring and analysis of the results to inform decision making.

Plan

Deciding what to monitor and how to monitor it are important aspects of precision farming. You might want to consider:

  • Do you already have or collect data on yields and soil status? Do you know what is or could be affecting crop growth or yields?

  • If you decide to monitor remotely using satellite data, which vegetation index will give you the most information?

What you monitor will depend on the crops you grow and what data you’ve collected in the past. It’s also important to consider changes you’ve made, or intend to make, to the management of fields and farm, and what you’ll need to monitor in order to see if the change has had an impact.

The plan will depend on whether or not you have the technology you need to make the measurements and collect data. We looked at sensor technologies in Step 2.8, but sensors can be costly. Are there low-tech/low-cost solutions available [2]? Or is there technology you already have that could be adapted, such as a smartphone [3]? You may find this article from CropLife useful: 20 Agriculture Apps You Should Know in 2020 and Beyond.

Monitor

Vegetation indices (covered in Step 2.5) provide a way of remotely monitoring crop growth and development and are useful for pinpointing areas with problems and supporting decisions on when to, for example, add fertiliser [4]. In some cases they can also enable identification of the problem, such as low moisture levels. However, they’re not yet effective at identifying crop diseases in the early stages of development, when treatment is more effective [5]. Vegetation indices can be used to optimise crop scouting and vegetation surveys reducing the risk of missing problem areas and leading to more precise and targeted interventions.

Crop scouting

Crop scouting is in-field monitoring of plant diseases and crop damage and is a valuable tool. The samples taken need to represent the entire cultivated area and there are several methods:

  • Convenience (non-random) sampling: samples are collected as and when ‘convenient’. So you might only sample where the field is easily accessible. This type of sampling is convenient and fast, but may not give adequately random (and therefore representative) samples. It can mean that the size of an infestation, for example, is over or underestimated.

birds-eye view of field with section marked out in red. 4 red dots represent sampling points which are clustered in one corner with just 2 spaced out in the middle of the section.

  • Simple random sampling: samples are taken from selected random points in the field. The randomness gives a balance of samples but it’s difficult to be sufficiently random without some kind of system. There are geographic support tools that help select sample locations, such as GPS detectors and sampling apps.

birds-eye view of field with section marked out in red. 8 red dots represent sampling points which are spaced out fairly randomly throughout the section

  • Systematic random sampling: samples are taken at regular spatial intervals. This is easier to apply than simple random sampling and produces spatially balanced samples. However, it’s highly sensitive to cyclical variables – variables that have a repeated pattern that could coincide with your sampling pattern.

birds-eye view of field with section marked out in red. 10 red dots are spaced out across the section in a grid pattern (2x5)

  • Stratified random sampling: before sampling the field is divided into homogeneous zones, these zones could be based on soil moisture, crop vigour, or other factors (stratification factors). Within each zone an equal number of samples are then chosen randomly. This type of sampling should only be used if you know the factor affecting the variation of what you’re monitoring. If applied correctly, this technique allows representative samples to be obtained with a limited number of measurements.

birds-eye view of field with section marked out in red. This section is divided into three different areas represented by different colours. 9 red dots representing sampling point are spaced randomly across the section, 3 in each different coloured area.

How many samples?

How many sampling points do I need to observe in the field to have a representative sample? There isn’t a simple answer to this question, but two things to take into account are:

  1. The higher the number of observations, the lower the chance of incorrect evaluation.

  2. The sample size should increase with increasing variability and not with the field size.

You’ll find links to resources on crop scouting methods in the See Also section at the end of this Step.

Analyse

Finally, the data must be analysed in order to inform changes in strategy such as changing fertiliser application rates, planting densities or even the crop. The analysis is specific to your situation but should include checking the data and calibrating to ensure the technology is working and measurements are accurate and consistent. For example, checking yield data for outlying values where harvester speed or distance travelled may have been measured incorrectly. Data should also be cleaned before the final output (a yield map for example) is produced [6].

The analysis might lead to:

  • Forecasting yields for a growing season [7] or on an annual basis [8].

  • Variable Rate Application: combining data such as yield maps, vegetation indices and soil maps and highlighting where higher or lower fertiliser input could increase yields or reduce waste [9].

  • Changing crops: analysis of yield maps, soils and vegetation indices could highlight marginal areas where it might be more sustainable (financially and environmentally) to grow a different crop or hybrid [10,11]. Or moving from arable to bioenergy crops on marginal land [12,13,14].

  • Set-aside zones: not planting in areas of the field where yields are consistently low or the cost of inputs (seeds, fertilisers, etc) is consistently more than the value of the yield [15,16].

Activity

Take a look at the action plan you started in Step 1.14. Choose one of the outcomes you want to prioritise. What do you need to monitor and how should you monitor it in order to provide the data that will allow you to measure improvement against your starting point?

© EIT Food
This article is from the free online

Innovation in Arable Farming: Technologies for Sustainable Farming Systems

Created by
FutureLearn - Learning For Life

Our purpose is to transform access to education.

We offer a diverse selection of courses from leading universities and cultural institutions from around the world. These are delivered one step at a time, and are accessible on mobile, tablet and desktop, so you can fit learning around your life.

We believe learning should be an enjoyable, social experience, so our courses offer the opportunity to discuss what you’re learning with others as you go, helping you make fresh discoveries and form new ideas.
You can unlock new opportunities with unlimited access to hundreds of online short courses for a year by subscribing to our Unlimited package. Build your knowledge with top universities and organisations.

Learn more about how FutureLearn is transforming access to education