The potential for heathland restoration on formerly arable land at a site in Drenthe, The Netherlands

1. The suitability of formerly arable land for heathland restoration was studied on sandy soils in the Netherlands. For this purpose, nutrient (N, P) availability and the composition of the seed bank were investigated in three situations: a Calluna-dominated heathland (CH), an adjacent fallow-lying formerly arable field that had been Calluna-dominated heathland until about 1950 (FA), and a fallow-lying formerly arable field from which the topsoil had been removed (FR). 2. Total soil organic matter content and total soil C, N and P were highest in the FA soil and lowest in the FR soil. The soil from the Calluna heathland showed intermediate values. 3. There were only small differences in the amount of extractable N between the CH soil and the FA soil. Removal of topsoil resulted in a significant reduction of extractable N. 4. In the FA soil the amount of extractable P was 50-180 times as high as the amount of extractable N, with a maximum content of about 280 kg ha(-1) for the upper 10 cm of the soil. In the heathland soil, the amount of extractable N was equal to or exceeded the values for the FA soil, but the values for extractable P were about 100 times lower. After removing the topsoil, the amount of extractable P in the formerly arable field was reduced to very low levels. 5. Nitrogen mineralization was not significantly different between the heathland soil and the FA soil in two out of three sampling periods. The removal of the topsoil of the formerly arable field substantially reduced N mineralization. 6. There were few significant differences in P release from the different soil types. The calculated amounts of P released over a 6-week period were in all cases lower than the amount of extractable P. Exactly the opposite pattern was observed for N. 7. Seeds of Calluna were detected only in the soil of the Calluna heathland. In the FA and the FR soil, seeds of common or ruderal plant species only were detected. 8. These results show that heathland restoration at this study site requires a substantial reduction of P availability in the soil together with the active introduction of heather propagules. Topsoil removal appears to be an effective means of reducing nutrient availability.