THE INFLUENCE OF WATER AVAILABILITY ON WINTER WHEAT YIELDS

Abstract

Field experiments were conducted comparing yield and yield components of winter wheat (Triticum aestivum L.) grown under different soil water conditions. Soil water was controlled by excluding precipitation from a 150-m² plot area with an automatic rain shelter. Treatment regimes were described according to their relative preanthesis/postanthesis soil water content as high/high (H/H), high/low (H/L), and low/high (L/H) in 1978–1979; an additional treatment, low/low (L/L) was added in 1979–1980. A neutron probe was used to periodically monitor soil water to the 150-cm depth in each regime. Plot yields ranged from 559 g/m² in regime H/H (1978–1979) to 267 g/m² in L/L (1979–1980) and were positively correlated with head number per square metre (r = 0.70) and kernel number per head (r = 0.79). Low preanthesis soil water reduced head number per square metre in both years. Regimes L/H and L/L in 1979–1980, which averaged the lowest preanthesis soil water of all regimes both years, had reduced kernels per spikelet compared to regimes with high preanthesis soil water. Increased kernel weight. associated with postanthesis irrigations, generally was not enough to compensate fully for fewer kernels per square metre associated with low preanthesis soil water. The results indicate that, if drought develops before grain filling in the spring, improved tiller survival and/or floret fertility could increase yields, even if some stress continued through grain filling. Under nonstress conditions, yield appears limited most by the amount of assimilate required to fill a high number of kernels per square metre.