Resource use efficiency of field-grown sunflower, sorghum, wheat and chickpea. II. Water Use Efficiency and comparison with Radiation Use Efficiency.

Water use efficiency (w) is a crop parameter of outstanding importance in crop simulation models, derived here as the slope of the relationship linking crop carbon gain to cumulative transpiration. This study aims at: (i) evaluating w of field-grown sunflower, sorghum, wheat and chickpea, at three levels of aggregation—day-time net assimilation (A), daily net carbon gain, as difference between A and night-time dark respiration (R), and biomass; (ii) assessing the robustness of w parameter in terms of ability to discriminate between C3 and C4 species, pre- and post-anthesis, impact of nitrogen status; (iii) investigating the opportunity to normalize w by climate, using FAO-Penman-Monteith reference evapotranspiration (Eref) or atmospheric saturation vapour pressure deficit (D); (iv) comparing w with the corresponding radiation use efficiency (e) results presented in the companion paper. Field experiments were conducted in 1998 and 1999 in southern Italy. All crops were well watered. Sunflower and sorghum had two nitrogen application treatments: wheat had only one and chickpea had no added nitrogen. Closed-system canopy chambers, automated for continuous measurements, were used to monitor gas-exchanges at the canopy scale, in terms of both carbon exchange rate (CER) and evapotranspiration E, and the cuvette method to monitor gas-exchange at the leaf scale. w, expressed in terms of A, A  R, or above-ground biomass, was linear over the entire cycle of sorghum and wheat, and up to anthesis in sunflower and chickpea, independent of temperature, vapour pressure deficit and radiation regimes. In sunflower, deviation from linearity was observed after anthesis due to higher carbon cost in yielding oil seeds. No conclusions could be drawn for post-anthesis chickpea due to the destruction of the crop by a thunderstorm. The overall response patterns were common to w and e. Nevertheless, w proved to be more robust than e due to its high capacity to discriminate between species groups (C3 from C4) and its effectiveness in normalizing the values for climate, provided it is implemented through Eref rather than D. All the above confers larger extrapolative ability to w-based crop models