a b s t r a c tA stumbling block to the adoption of silvoarable agroforestry systems is the lack of quantitative knowledge on the performance of different crops when competing for resources with trees. In North-Western Europe, light is likely to be the principal limiting resource for understorey crops, and most agronomic studies show a systematic reduction of final yield as shade increases. However the intensity of the crop response depends on both the environmental conditions and the shade characteristics. This study addressed the issue by monitoring winter wheat (Triticum aestivum L.) growth, productivity and quality under artificial shade provided by military camouflage shade-netting, and using the Hi-sAFe model to relate the artificial shade conditions to those applying in agroforestry systems.The field experiment was carried out over two consecutive years (2013-14 and 2014-15) on the experimental farm of Gembloux Agro-Bio Tech, Belgium. The shade structures recreated two shade conditions: periodic shade (PS) and continuous shade (CS), with the former using overlapping military camouflage netting to provide discontinuous light through the day, and the latter using conventional shade cloth. The experiment simulated shading from a canopy of late-flushing hybrid walnut leaves above winter wheat. Shading was imposed 16 (2013-14) and 10 (2014-15) days before flowering and retained until harvest. The crop experienced full light conditions until the maximum leaf area index stage (LAI max ) had been reached. In both years, LAI followed the same dynamics between the different treatments, but in 2013-2014 an attack of the take-all disease (Gaeumannomyces graminis var. tritici) reduced yields overall and prevented significant treatment effects. In season 2014-15 the decrease in global radiation reaching the crop during a period of 66 days (CS: -61% and PS: -43%) significantly affected final yield (CS: -45% and PS: -25%), mainly through a reduction of the average grain weight and the number of grain per m 2 . Grain protein content increased by up to 45% under the CS treatment in 2015. Nevertheless, at the plot scale, protein yield (t/ha) did not compensate for the final grain yield decrease.The Hi-sAFe model was used to simulate an agroforestry plot with two lines of walnut trees running either north-south or east-west. The levels of artificial shade levels applied in this experiment were compared to those predicted beneath trees growing with similar climatic conditions in Belgium. The levels used in the CS treatment are only likely to occur real agroforestry conditions on 10% of the cropped area until the trees are 30 years old and only with east-west tree row orientation.

Agricultural productivity relies on a wide range of ecosystem services provided by the soil biota. Plowing is a fundamental component of conventional farming, but longterm detrimental effects such as soil erosion and loss of soil organic matter have been recognized. Moving towards more sustainable management practices such as reduced tillage or crop residue retention can reduce these detrimental effects, but will also influence structure and function of the soil microbiota with direct consequences for the associated ecosystem services. Although there is increasing evidence that different tillage regimes alter the soil microbiome, we have a limited understanding of the temporal dynamics of these effects. Here, we used high-throughput sequencing of bacterial and fungal ribosomal markers to explore changes in soil microbial community structure under two contrasting tillage regimes (conventional and reduced tillage) either with or without crop residue retention. Soil samples were collected over the growing season of two crops (Vicia faba and Triticum aestivum) below the seedbed (15-20 cm). Tillage, crop and growing stage were significant determinants of microbial community structure, but the impact of tillage showed only moderate temporal dependency. Whereas the tillage effect on soil bacteria showed some temporal dependency and became less strong at later growing stages, the tillage effect on soil fungi was more consistent over time. Crop residue retention had only a minor influence on the community. Six years after the conversion from conventional to reduced tillage, soil moisture contents and nutrient levels were significantly lower under reduced than under conventional tillage. These changes in edaphic properties were related to specific shifts in microbial community structure. Notably, bacterial groups featuring copiotrophic lifestyles or potentially carrying the ability to degrade more recalcitrant compounds were favored under conventional tillage, whereas taxa featuring more oligotrophic lifestyles were more abundant under reduced tillage. Our study found that, under the specific edaphic and climatic context of central Belgium, different tillage regimes created different ecological niches that select for different microbial lifestyles with potential consequences for the ecosystem services provided to the plants and their environment.

No abstract