Co-limitation of nitrogen and water, and yield and resource-use efficiencies of wheat and barley C. Mariano Cossani A,C , Gustavo A. Slafer A,B , and Roxana Savin A A Department of Crop and Forest Sciences, University of Lleida, Centre UdL-IRTA, Av. Rovira Roure 191, Lleida 25198, Spain. B ICREA (Catalonian Institution for Research and Advanced Studies), Catalonia, Spain. C Corresponding author. Email: mariano.cossani@pvcf.udl.cat Abstract. In semiarid Mediterranean environments, low nitrogen (N) and water availabilities are key constraints to cereal productivity. Theoretically, for a given level of N or water stress, crops perform better when co-limitation occurs. Empirical evidence of this theoretical concept with field crops is rather scarce. Using data from field experiments we evaluated whether N-use efficiency (NUE) and water-use efficiency (WUE) in small grain cereals increases with the degree of co-limitation. Four field experiments were carried out during three growing seasons including factorial combinations of bread wheat, durum wheat and barley, grown under different N fertiliser rates and water regimes. Yield gap was calculated as the difference between maximum attainable yield and actual yield while stress indices for N (NSI) or water (WSI) were calculated as the ratios between actual N uptake or water use and those required to achieve maximum yields, respectively. Water and N co-limitation was calculated as C WN =1 – |NSI–WSI|. The relationships of yield gap, NUE and WUE with the different co-limitation indices were evaluated. Yield gap (range from –3.8 to –8.1 Mg ha –1 ) enlarged (was more negative) with the highest levels of stress and, as expected from theory, it was reduced with the degree of co-limitation. WUE ranged from 6.3 to 21.8 kg ha –1 mm –1 with the maximum values observed under conditions in which co-limitation increased. Reduction in yield gap with increased degree of co-limitation was mainly due to a positive effect of this variable on WUE. Additional keywords: Hordeum vulgare, Mediterranean, nitrogen-use efficiency, Triticum aestivum, Triticum durum, water-use efficiency, yield gap. Introduction Nitrogen (N) and water availabilities are the main factors affecting cereal production in semiarid or arid Mediterranean environments (Passioura 2002). In rain-fed agricultural systems, N availability for crops depends on the initial N soil content, and also on seasonal mineralisation, which could be increased or reduced through management practices. However, water availability depends on the total amount of rainfall during fallow and the growing cycle. Management practices as type of tillage, type of fallow, sowing rate and fertilisation regime can alter the proportion of water evaporated directly from the soil, through modifying the amount of water actually transpired by the crops (Hatfield et al. 2001). Increasing the efficiency in the use of water rained during the season and/or stored before sowing is imperative to reduce the gap between maximum attainable and actual water-use efficiency (WUE) under rain-fed systems (Sadras and Angus 2006). The maximum attainable yield of a cultivar is obtained when grown in environments to which it is adapted without limitations of biotic or abiotic stresses, and with pests, diseases, weeds, lodging, and other stresses effectively controlled (Evans and Fischer 1999). Yield does not normally reach its maximum level as the most frequent agronomic condition is under varying degrees of stress, determining a yield gap. As mentioned above, water and N are recognised as the main factors limiting yield in small cereals (and then responsible for the yield gap) in Mediterranean conditions. Savin et al.(2006) reported a maximum attainable grain yield of wheat for the Mediterranean region of 9 Mg ha –1 based on data reported in published papers. Bloom et al.(1985) used an analogy from economics to understand resource limitation. They proposed that plants modify resource allocation so that their limitation of growth is nearly equal for all resources. Consequently, plants would maximise their growth when different resources are similarly limiting rather than when growth is severely limited by a single factor. In other words, growth and yield of stressed crops would be positively related to the degree of co-limitation. Sadras (2004), using a simulation model, supported the hypothesis that the gap between attainable and actual wheat yield of water-N stressed crops was negatively related to the degree of water and N co-limitation (C WN ). Together with the abovementioned analysis, other papers (Sadras and Roget 2004; Sadras et al. 2004) and a review with joint analysis (Sadras 2005) were published relating yield gap, WUE and N-use efficiency (NUE) to the degree of co- limitation. Although in all three publications the conclusions provided support to the hypothesis, only one of the regular Ó CSIRO 2010 10.1071/CP10018 1836-0947/10/100844 CSIRO PUBLISHING www.publish.csiro.au/journals/cp Crop & Pasture Science, 2010, 61, 844–851