Developing and validating on-farm sampling protocols: sampling in store and during out-loading


Cereals & Oilseeds
Project code:
01 September 2002 - 31 August 2003
AHDB Cereals & Oilseeds.
AHDB sector cost:
£24,900 (HGCA Project No. 2748).
Project leader:
J D KNIGHT, D R WILKIN and J RIVETT Imperial College London, Silwood Park Campus, Ascot, Berkshire SL5 7PY



About this project


The aim of this project was to establish whether sampling of grain whilst in store provides a reliable method of determining the quality of a grain batch. The results of in-store sampling were compared with those obtained from sampling at intake and during out-loading.

Grain was sampled at a total of 9 farms starting from harvest 2002 through into 2003. The farms selected used a variety of drying methods (high and low temperature and ambient air) and storage methods (bin and on-floor). All sampling in the store was done using a 1.7m grain spear. Results for sampling and grain quality at harvest were from previous work investigating sampling at intake (HGCA Project Report 301).

The stored grain was sampled after approximately 1, 4 and 6 and, in one case, 10 months. In-store sampling was compared with sampling at out-loading by taking samples from the bulk to be loaded using a spear and then taking a litre of grain from each 2 tonne bucket as it was loaded into the lorry. The results of the quality analyses were then compared for the two sets of samples. The grain sampled at out-loading was also sampled at intake to the mill and the analyses done by the project were compared with the results obtained by the mill.

The quality parameters measured (protein, specific weight, moisture and hardness) did not vary markedly during the period of storage and sampling in store. Broadly similar values were obtained for the quality parameters compared with the samples taken at intake and at out-loading. However, because fewer samples are generally taken in store, it is unlikely that the whole range of variability was covered.

Results for mill samples were not statistically different from the results obtained on-farm but were sufficiently different to be commercially important. However, differences did not consistently favour or penalise either mill or supplier.

Despite its limitations, in-store sampling would appear to give a reasonable sample to determine the quality characteristics and obviously is essential