The Broadbalk Wheat Experiment, Rothamsted, UK Crop yields and soil changes during the last 50 years

Long-term experiments (LTEs) are a vital source of information contributing to understanding and assessing the sustainability of agricultural systems. The Broadbalk Wheat Experiment at Rothamsted Research, UK, started in 1843, is the oldest continuing LTE in the world. In 1968, and later, major modifications were made to the experiment. These included the introduction of short-straw winter wheat varieties, growing wheat in rotation with other crops, for comparison with continuous (monoculture) wheat and testing higher rates of N fertilizer (up to 288 kg N ha−1). Other changes included the use of herbicides and other agrochemicals, except on selected sections. Regular liming continued to maintain soil pH. This paper reports crop yields and soil changes since 1968 and summarizes other studies including the incidence of weeds, pests and crop diseases, effects of treatments on nitrate and phosphate movement to drainage water, greenhouse gas fluxes, measurements of N and S inputs from the atmosphere, and many other factors.

The change to short-straw varieties led to a doubling of grain yields. In recent years with cv. Crusoe yields of first wheat after a 2-year break in a rotation were >3 t ha−1 greater than continuous wheat, mainly because of decreased incidence of soil-borne pests and diseases. On average, yields of continuous wheat tended to increase at N rates up to 240 or 288 kg N ha−1 while the 1st wheat after a 2-year break often needed less N to reach maximum yield. Best yields now exceed 12 t ha−1 in some years.