ASPECTS OF MAIZE MODELING IN EASTERN CANADA

Maize (Zea mays L.) is a crop of growing importance in Eastern Canada. Modelling the temperature effects on phenological development, crop a rchitecture and disease infection in maize contributes to the developm ent of well-adapted, early-maturing varieties. Details of modelling th ese three aspects of maize growth were presented. The first focussed o n quantifying the effect of air or soil temperature on maize phenologi cal development. Crop growth was divided into two periods: vegetative (planting to silking) and grain filling (silking to maturity). A third period (planting to emergence) was separated within the vegetative pe riod. Heat unit systems based on daily temperature response functions were developed to produce the most consistent heat unit sums for each period. The best fits of these functions were determined by minimizing standard deviations and coefficients of variation of these sums for e ach thermal period over locations and years. Calculated temperature re sponse functions estimated thermal periods more consistently than grow ing degree days (GDD) for all three periods. The largest improvement w as made in the silking to maturity period. The second aspect was a stu dy of crop architecture. Methods were developed to mathematically char acterize the structure of a canopy in terms of leaf area and leaf angl e distributions with crop height and across the row. These calculation s, in turn, were input to a soil-plant-atmosphere model to calculate i nterception of photosynthetically active radiation (PAR). Model calcul ations of PAR interception compared well with measurements for a range of plant types and plant population densities (R-2=0.76). The third a spect was quantifying growth of Fusarium in maize. Differential equati ons were used to relate Fusarium rates of growth in maize ears to air temperature, relative humidity and precipitation. Integration of these equations over time produced quantitative estimates of fungal infecti on. Model calculations were compared to visual ratings of fungal infec tion for two Fusarium species over three years (R-2=0.92). In each of the three aspects of this study, modelling tested our understanding of the processes involved and the dominant factors controlling these pro cesses. Thus, modelling was an integral part of the scientific approac h, synthesizing experimental data in a quantitative conceptual framewo rk and identifying dominant factors and parameters which required addi tional focussed experimental evaluation.