Lessons learned from the farm scale evaluation of GMHT crops

When the farm scale evaluations (FSE) of GMHT crops were initiated in the UK there was very little information in the UK on the variability of many farmland species within fields to inform the design of such trials. The original assumptions used in the calculations of site replication were tested using data collected in year 1 and 2. At this time the possibility of increasing site numbers in year 3 remained, but site replication was found to be sufficient. Upon completion of the trials, analyses of the data collected have validated the site replication and the power of the study to detect biologically important differences between conventional and GMHT management systems. Sub-samples of data corresponding to reduced sampling protocols at each site were analysed to investigate ways to reduce in-field sampling time whilst retaining the same precision as in the original scheme. The relative costs of the various protocols employed, both in staff time and financially have been published, as this too is an important constraint when planning any project. Besides producing the data necessary to test the null hypothesis, the FSE has also provided a significant body of information on the UK experience with the variability of farmland plants and invertebrates with a change in management practices, which is a key resource for future researchers in designing other field-based studies.