Simulation of wheat ontogenesis.

The main purpose of this study is to understand wheat ontogenesis and to formulate a simulation model of the development of wheat for application under field conditions. The basic idea that is behind this work is that plant development can be accurately simulated on the basis of leaf appearance provided that the final number of leaves is known. Apex and leaf development are in fact coordinated and in plants with terminal flowers, flowering occur when all the leaves are appeared on the stem. Wheat ontogenesis, apical and phenological development of wheat are described at first in the introductory chapter. A brief review of current knowledge about factors affecting wheat development and some information about existing ontogenetic models are also given.Subsequently, data of some experiments in the literature are compared and statistically analysed to confirm the assumption of independence of wheat leaf initiation from daylength, and to evaluate the effect of temperature. Independence for any daylength treatement is demonstrated by the strict proportionality between the total number of initiated leaf primordia and the time to double ridges. The temperature response of leaf initiation rate is evaluated as the slope of a regression line between these two variables for different thermal treatments. On such a basis a linear model is constructed where daily leaf primordium initiation rates are calculated as a function of temperature. Data from a specific field experiment are then used to test the model. In the experiment, two varieties (Maris Huntsman and Creso) and two sowing dates (late November and early February) are compared. Predicted and observed dates of double ridge appearance and the predicted number of initiated primordia match rather accurately for both sowing dates and varieties, confirming the previous hypothesis. It is concluded that vmost of the genotypic variabIlity in wheat ontogenesis is accounted for by the effect of photoperiod on earliness of floral induction and, therefore, on the final main stem leaf number.However, rate of leaf appearance differs among wheat crops sown in the field at different dates. This can be interpreted either as a direct effect of the date of sowing on the rate of leaf appearance or as an indirect effect of an ontogenetic decline in the rate of leaf appearance as the plant ages. This decline is attributed to the increasing distance that has to be traversed the youngest leaf in the apparent stem that is formed by the leaf sheets of the older leaves. Analysis of data from both laboratory and field experiments leads to the conclusion that the second hypothesis is more likely. A simulation model of leaf appearance is formulated on the basis of this hypothesis, and tested using experimental results from the literature and 2 original experiments carried out in Italy. A good agreement between experimental data and simulation results is found. The model simplifies the problem of simulation of leaf appearance considerably compared with existing models that are based on a spurious relation between the rate of change in daylength at crop emergence and the rate of leaf appearance, which does not have a physiological base.The problem of the prediction of the final number of leaves is then approached. The role of photoperiod in the regulation of wheat development and on final main stem leaf number is analysed by means of literature data from a number of laboratory experiments. A procedure is developed to calculate the final number and the date of heading of plants that do not require vernalisation. Subsequently, differences between calculations and field observation are ascribed to an effect of vernalisation. This allows the formulation of a model to predict the final number of leaves and date of heading of any given wheat variety. The model involves the following assumptions: (1) wheat varieties have different sensitivities to the daylength; (2) wheat varieties that require vernalisation are vernalised at a very early stage of growth if sown at the beginning of the coldest period of the year; (3) wheat varieties that are vernalised at a very early stage of growth immediately respond to external photoperiodic conditions; (4) wheat crops sown within a range of sowing dates tend to synchronize time of flowering. It is concluded that if the final main stem leaf number of a crop sown at a given date is known, the date of heading and the corresponding final leaf number of main stem leaves of every other sowing can be found. The model is validated using field data from 58 experimental trials performed in the USA and Europe.In the last part of the work, the effect of nitrogen fertilization, water shortage and of the genotype on both the rate of leaf appearance and the final leaf number are considered. Results of field experiments conducted in USA and of original experiments performed in Italy are used for such a purpose. Data analysis shows that leaf appearance, final leaf number and phasic development of wheat are independent of nitrogen fertilization and water shortage and that genotypic effect is restricted to the control of the final number of leaves but not of the rate of leaf appearance.Finally, the formulated models are extensively validated on basis of data recorded in field experiments and it is concluded that these appear to be well suited for monitoring wheat ontogenesis under field conditions. The models can be used in agronomy, for analyses of the influence of weather on the development of wheat and of the adaptation of varieties to different environments and for studying the impact of climatic changes on agriculture