I expected harvest to be completed in September, but we are now into October, with no progress for the past eight days.

All looked to be fine last Friday morning with a forecast of a few sunny days over the weekend giving the opportunity to cut the last of the wheat and finish off wheat drilling in good conditions.

However, by mid-morning the forecast had changed, the sunshine replaced by drizzle and the dry weather moving a few days further into the future.

See also: 3 pesticide approval changes to watch this autumn

Within 24 hours we had had 30mm of rain, and the dry spell is still to arrive.

It feels as if the accuracy of weather forecasting has decreased recently.

We have easy access to so many forecasts now that our expectations are maybe too high, but when 30mm of rain for the whole region falls within 24 hours of a forecast of a dry sunny day, I do feel somewhat cheated!

Many times this season the forecast for the following week has been promising, but with the exception of a couple of times when we had a few decent days in a row, dry days have been elusive, with an average of only one dry day in three since harvest began.

Low germination

The last of the spring malting barley has now almost gone off the farm and in general has been very good quality, with low nitrogen, screenings and grain skinning combined with above-average yields.

Only an odd load with unexplained low germination levels has returned to farm.

This raises a question in my mind about the accuracy and statistical significance of the current sampling and testing protocol, but as yet there does not seem to be a better way of determining grain germination for a large tonnage.

In the absence of flood or drought this season, the wheat varieties have mirrored the Septoria ratings for disease levels at the end of the season and this has translated to final grain size and yield.

It is certainly justification of the decision to drop varieties with a low score, as additional chemical cost has not been able to protect the yield potential.