A field experiment was established in 1996 on a loamy Xerofluvent Typic in the Ebro river valley (NE Spain). The experiment compared three nitrogen (N) fertilization levels (zero, 0 kg N ha?, medium, 60 kg N ha?, and high, 120 kg N ha?), under three tillage systems (CT, conventional tillage; MT and NT), annually cropped to barley (Hordeum vulgare, L.) as is usual in the region. Ten years after the experiment establishment, during four consecutive growing seasons, 2005?006 to 2008?009, we evaluated the response of soil water content, soil nitrate, above-ground dry matter, grain yield and yield components to long-term (>10 years) tillage and N fertilization treatments.
The long-term sustainability of NT and MT was confirmed. Mean yield and WUE under long-term conservation tillage systems were 66 % and 57 % higher than under CT, respectively. This improvement was mainly attributed to improved soil water usage under conservation tillage, mainly due to reduced water use during the pre-anthesis period. However, in a wet year yield did not significantly differ among tillage systems. The improvement of WUE with N fertilization was confirmed under NT, which medium and high N fertilizer level increased 98 % mean grain yield and 77 % mean WUE compared to CT. The increased response of crop and yield to N fertilization under NT was due to improved soil water content. Soil N accumulation together with the lower water accumulation explained the lack of response to N fertilization under CT, even on a wet growing season (i.e., 2008?009).
Long-term NT adoption was a sustainable practice for barley monoculture in the region, allowing for reduced costs and yield increase with N fertilizer additions. N fertilizer rates on rainfed Mediterranean croplands should be adjusted depending on the reduction of tillage intensity and rainfall of the year. In our system and as an example for this agroecosystems, N fertilizer rates should be kept at or below 60 kg N ha?, and should be further reduced on intensively cultivated soils.