Field pea is an important annual crop due to its contribution to soil fertility and other rotational benefits. However, weeds and ascochyta blight limit pea yield, particularly in organic systems. Leafed and semi-leafless pea types differ in lodging resistance, and may affect weeds and disease through differences in canopy light penetration and air flow. Mixtures of the two leaf types may improve weed and disease suppression and yield compared with monocultures of the same cultivars. To test this hypothesis, replicated field experiments were conducted under organic and conventional management in Saskatoon and Vonda, SK, in 2011 and 2012. Mixtures of a leafed and semi-leafless cultivar, CDC Sonata and CDC Dakota, were sown in ratios of 0:100, 25:75, 50:50, 75:25, and 100:0 leafed to semi-leafless pea, at target seeding rates of 88 and 132 plants m-2. Conventionally managed plots were inoculated with ascochyta blight-infested pea straw and received overhead irrigation to encourage disease. Mixtures of 50% or more semi-leafless pea adopted the greater lodging resistance and weed suppression of the semi-leafless cultivar. Mixtures comprised of 25% leafed and 75% semi-leafless pea increased both seed and biomass yield compared with either cultivar grown alone. Yield enhancement was attributed to the leafed cultivar, whose seed yield was 76% higher in mixture than expected based on monoculture yield. Ascochyta blight epidemics were of moderate severity, and leafed and semi-leafless monocultures reached 36 and 43% necrosis in 2011, and 33 and 38% necrosis in 2012, respectively. The disease reaction of mixtures fell between the two component cultivars. At disease onset in 2012, lower light interception and shorter moisture durations coincided with the lower ascochyta blight severity of leafed monocultures. In 2011 and the later phase of the 2012 epidemic, disease severity was negatively associated with vine length, and positively associated with number of nodes and tissue senescence. Despite the advantages of leafed and semi-leafless pea mixtures, the limited selection of leafed cultivars impedes adoption of this technique by growers. For pea breeders, developing mixtures of pea lines isogenic for leaf type may increase yield compared with single cultivars.

La culture d’associations de variétés sensibles et résistantes au sein d’une même parcelle permet de réduire la propagation des maladies fongiques foliaires aériennes. L’architecture des plantes a un impact sur la dispersion de spores et le microclimat, mais est rarement prise en compte dans la conception des associations. L’objectif de cette thèse était de comprendre le rôle joué par l’architecture dans des associations de variétés différant par leur niveau de résistance à une maladie dispersée par éclaboussement, la septoriose du blé, causée par Zymoseptoria tritici. Une expérimentation en conditions contrôlées a permis de quantifier des relations entre interception de spores et architecture des couverts. Des différences de densité entre couverts purs et associés ont donné lieu à une réduction du niveau de maladie sur les plantes sensibles cultivées en association par rapport à celles cultivées pures. Lors d’une expérimentation au champ, les associations de variétés de hauteurs de tiges contrastées étaient moins malades que celles ayant des hauteurs similaires. Ces résultats ont été mis en relation avec des effets de l’architecture sur la dispersion de spores et la durée d’humectation au sein des couverts. Une approche de modélisation spatialement explicite a permis d’identifier des mécanismes de dispersion par éclaboussement liés à l’architecture des couverts associés. Dans des associations de variétés de hauteurs différentes, la quantité d’inoculum éclaboussée dépendait de la surface foliaire présente au dessus des feuilles malades (effet parapluie). La quantité d’inoculum interceptée par un étage foliaire était liée à la différence de hauteur entre la source d’inoculum et l’étage foliaire (effet hauteur). Les différences de hauteur de plantes entre variétés d’une association ont modulé l’interception de spores par des feuilles résistantes (effet barrière). Nos résultats suggèrent qu’une prise en compte de l’architecture des variétés dans la conception des associations variétales permettrait de mieux maîtriser les maladies par éclaboussement. / Growing mixtures of susceptible and resistant cultivars in the same field makes it possible to reduce the propagation of airborne fungal plant diseases. Plant architecture has an impact on spore dispersal or microclimate, but is rarely taken into account in mixture design. The objective of this work was to understand the role of canopy architecture in mixtures of cultivar of different levels of resistance to a disease dispersed by rain-splash, septoria tritici blotch, caused by Zymoseptoria tritici. A controlled conditions experiment made it possible to quantify relationships between spore interception and canopy architecture. Differences of canopy density between pure stands and mixtures led to a reduction in disease on susceptible plants grown in mixture, compared to the susceptible pure stand. During a field experiment, mixtures of cultivars with contrasted stem height were less diseased than those with similar stem height. These results were related to the effect of canopy architecture on spore dispersal and leaf wetness duration. A spatially explicit modeling approach made it possible to identify splash dispersal mechanisms related to the architecture of mixed canopies. In mixtures of cultivar with diverse plant height, the amount of splashed inoculum depended on leaf area located above diseased leaves (umbrella effect). The amount of inoculum intercepted by a leaf layer was related to its difference of height between the inoculum sources (height effect). Differences of plant height between cultivars composinga mixture modulated the interception of spores by resistant leaves, providing an increased protection of susceptible leaves (barrier effect). Our results suggest that considering cultivar architecture in the design of cultivar mixtures could make it possible to improve the management of splash-dispersed diseases.