Winter barley is the major crop on semiarid drylands in central Aragon (NE Spain). In this study we compared, under both continuous cropping (BC) (5-6-month fallow) and a crop-fallow rotation (BF) (16-18-month fallow), the effects of three fallow management treatments (conventional tillage, CT; reduced tillage, RT; no-tillage, NT) on the growth, yield and water use efficiency (WUE) of winter barley during three consecutive growing seasons in the 1999-2002 period. Daily precipitation measurements and monthly measurements of soil water storage to a depth of 0.7 m were used to calculate crop water use (ET) and its components. The average growing season precipitation was 195 mm. Above-ground dry matter (DM) and corresponding WUE were high in years with high effective rainfalls (>10 mm day -1) either in autumn or spring. However, the highest values of WUE for grain yield were mainly produced by effective rainfalls during the time from stem elongation to harvest. Despite the similarity in ET for the three tillage treatments, NT provided the lowest DM production, corresponding to a higher soil water loss by evaporation and lower crop transpiration ( T), indicated by the lowest T/ET ratio values found under this treatment. No clear differences in crop yield were observed among the tillage treatments in the study period. On average, and regardless of the type of tillage, BF provided the highest values of DM and WUE and yielded 49% more grain than BC. These differences between cropping systems increased when water-limiting conditions occurred in the early stages of crop growth, probably due to the additional soil water storage under BF at sowing. Although no significant differences in precipitation use efficiency (PUE) were observed between BC and BF, PUE was higher under the BC system, which yielded 34% more grain than the BF rotation when yields were adjusted to an annual basis including the length of the fallow. The crop yield under BF was not dependent on the increase in soil water storage at the end of the long fallow. In conclusion, this study has shown that, although conventional tillage can be substituted by reduced or no-tillage systems for fallow management in semiarid dryland cereal production areas in central Aragon, the practice of long-fallowing to increase the cereal crop yields is not longer sustainable.
