A Simple Drought-Sensitive Model for Leaf:Stem Partitioning of Wheat

We investigated the leaf:stem partitioning of winter wheat (Triticum aestivum L. varieties Dekan' and Batis') with and without drought influence. Irrigated and drought-stressed winter wheat, grown in a rainout shelter in 2009/10 and 2013/14, were sampled during shoot elongation phase at the experimental Farm Hohenschulen located in Northern Germany. The data set contains leaf (DML) and stem dry masses (DMS), as well as measured water contents for several soil layers. A reduced relative dry matter allocation to leaves was observed under drought stress. Our results clearly show that, if simulated leaf:stem partitioning is not sensitive to drought, this will cause a positive bias in simulated leaf and a negative bias in simulated stem dry matter under water-limited conditions. This is in accordance with previous studies which revealed that crop simulators often overestimate the impact of drought on light-use efficiency, whereas the consequences on leaf area development are underestimated. However, the drought stress-induced shift in leaf:stem partitioning is yet not considered by most common wheat crop simulators. Our aim was to fill the gap in simulation of drought stress implications on leaf area development. A simple allometric model for leaf:stem partitioning (In(DMS)=gIn(DML)+h) was parameterized. Starting from the allometric leaf:stem relationship observed under optimum water supply, a correction term was introduced, which allows to adapt the partitioning to drought stress conditions. The lg-transformed root-weighted soil water potential in the rooting zone (lg(root), lg(hPa)), calculated as a function of measured water contents and simulated root distribution, was used as a drought stress indicator. The linear correction term assumes an increase of the stem fraction, proportional to the difference between lg(root) and a drought stress threshold (pF(crit), lg(hPa)). The analysis revealed that the shift in allometric partitioning towards stem fraction starts with lg(root) greater than 1.92 [lg(hPa)]. The slope of the relative increase of dry matter allocated to the stem fraction was determined with 0.26 [lg(hPa)(-1)]. Both parameters of the correction term were found to be highly significant. Implications for crop modelling are discussed.