You are here

How valuable is an early season yield forecast?

Publication Date: 
24 June 2016

Amber Cottingham, Analyst,, 02476 478 698

The monthly MARS (Monitoring Agricultural ResourceS) report published by the European Commission’s science and knowledge service uses remote sensing to provide independent information on crop areas and yields. Regular crop yield predictions are given throughout the year based on the remote sensing information and modelling. Although interesting to look at, their forecasts at this point in the year tend to be more accurate in years when yields are close to average – in other words, the ability to foresee extremes by June is limited.

The June MARS report has forecast a 3.7% increase year on year in potato crop yields for the UK, and 4.5% above the five year average. However, this yield forecast has not changed from previous months. In previous years, the yield forecast has often been revised in June, but this revision could have been delayed this season due to late planting.

Figure 1 below shows the MARS yield forecasts for the NEPG countries and the EU-28 from the latest report published on Monday 20 June. Predictably, all of the NEPG countries (except the UK) and the EU-28 as a whole have had their yield forecasts revised downward since May. For many of the EU countries this will be in response to the damaging weather conditions and severe flooding experienced over the last month.

Note that MARS uses Defra yield figures for the UK as a baseline, which can differ considerably from AHDB’s figures. As such, we have only considered the % change and applied this to AHDB figures.

By looking at the MARS forecast percentage change in UK yield compared to the previous season and the five-year average, we can see how close the resulting yields have been to actual yields in previous years. Figure 2 below looks at this.

*Based on MARS June yield forecast

In 2012, we can see that both yield forecasts differed a huge amount from the final yield but this will be somewhat expected due to the freak nature of the weather that season. 2013 is a closer match, especially looking at the comparison to the five-year average. Both the 2014 forecasts were reasonably accurate but then in 2015 the forecast didn’t predict the record high yields achieved, which in itself was a surprise to many people. Looking at MARS’ yield changes for this season, the two forecasts give quite different figures.

Concluding comments

The MARS yield forecasts tend to be more accurate in years where there are less variations from the norm. In years 2012 and 2015 where yields were either significantly down or unexpectedly high the forecasting model did not provide a June yield forecast close to actual yields. However, in less dramatic years the earlier June predictions have been closer to the actual yield. The current forecasts for this season already vary depending on whether you look at changes vs a year ago or the five-year average. This, in addition to the unusually late planting progress, may indicate that 2016 yields are less easy to forecast at this point. However, this may change as MARS yield forecasts evolve throughout the season and new information becomes available.



*For the purpose of assessing a final yield prediction we have used the MARS yield prediction % change vs previous year and vs previous 5 year average. These % changes are applied to figures gathered by AHDB, with Northern Ireland’s area planted and production figures incorporated to give a total UK yield.

How useful did you find this information?
Only logged in users can vote. Click on a star rating to show your choice, please note you can only vote once.
No votes yet