摘要

Water availability is a key biophysical factor determining agricultural production potential. The FAO crop water response model AquaCrop was developed to estimate crop production under water limiting conditions. This model uses the normalized water productivity, WP* (g m(-2) d(-1)), to estimate the attainable rate of crop growth under water limitation. In this study we assessed the value and the consistency of WP* under the semi-arid growing conditions of Inner Mongolia. Field experiments on productivity and water use of oats (Avena sativa), sunflower (Helianthus annuus), vetch (Vicia sativa), faba bean (Vicia faba) and potato (Solanum tuberosum) were conducted in 2009, 2010 and 2011. Yield, biomass, evaporation and transpiration varied substantially and significantly between crop species and years. Likewise, water use efficiency (WUE) was affected by these factors, including a significant interaction. WUE (units of yield/unit water) varied between years: from 1.1 to 1.3 kg m(-3) in oats, from 1.5 to 2.6 kg m(-3) in sunflower, from 0.5 to 1.1 kg m(-3) in the two legumes and from 3.1 to 4.4 kg m(-3) in potato (DM yields). Normalized water productivity did not vary significantly between crop species and years or their interaction, showing an overall value of 14.0 +/- 0.32 g m(-2) d(-1) in an overarching regression pooling all the data. We conclude that WP* differs from the conventional parameter for water productivity, but is a useful parameter for assessing attainable rate of crop growth and yield in the agro-pastoral ecotone of Inner Mongolia.