Search

Numerous studies have dealt with the relationship between leaf nitrogen content and leaf irradiance. However, most of them refer to dense stands presenting reduced horizontal heterogeneity of foliage distribution. Both gradients of leaf nitrogen and leaf irradiance related to canopy depth are significant under such conditions, and modelling radiative exchange using a turbid-medium analogy and dividing the canopy into vegetation layers is sufficient. Conversely, row crops such as maize are characterized by strong horizontal heterogeneity of foliage distribution and the one-dimensional (1D) approach may be unsuitable. We thus modelled the three-dimensional (3D) geometry of maize canopies with varying densities and at different developmental stages using plant digitizing under field conditions. The nitrogen content per unit area of each leaf part was obtained subsequently by nitrogen analysis. We next calculated radiative exchange using a 3D volume-based approach within the canopies in order to estimate local leaf irradiance on a daily integration scale. Vertical gradients in leaf nitrogen content per unit area observed in dense stands during the vegetative phase corresponded largely to those reported in the literature. We also identified significant gradients in nitrogen content along the leaves, which had not before been clearly demonstrated. Our study shows that local light climate during plant development plays a major role in leaf nitrogen distribution and remobilization. Moreover, brutal plant thinning involves rapid changes in leaf nitrogen partitioning. It is concluded that taking account of the 3D heterogeneity of nitrogen and irradiance distribution may have implications for modelling crop photosynthesis and production.

The NitroEurope Integrated Project (EC FP6, 2006-2011) aimed to analyse the nitrogen cycle in order to quantify the effect of reactive nitrogen (Nr) on the greenhouse gas (GHG) balance in Europe and propose nitrogen abatement strategies (Sutton et al., 2007). The landscape represents the scale at which nitrogen management decisions are taken by farmers according to local meteorological conditions, regional contexts and agricultural policies. The landscape is also the scale at which analysing the spatial interactions (i.e., vertical and lateral transfer) between nitrogen sources (e.g., emission from individual farm buildings or from agricultural fields) and nitrogen sinks (e.g., semi-natural ecosystems) are highly relevant for assessing environmental impacts of nitrogen management (Cellier et al., 2011). The challenge was to develop a framework integrating pathways of Nr transfer and transformation to predict Nr and GHG budgets at the landscape scale. Recent studies attempted to assess the effect of anthropogenic activities on terrestrial ecosystems, but they focused on one or two compartments of the landscape (atmosphere, hydrosphere, vegetation or farm buildings), the others being described in less detail (Beaujouan et al., 2001; Hutchings et al., 2004; Theobald et al., 2004; Ducharne et al., 2007). To continue with those approaches, the NitroScape model has been developed by integrating Nr flows in the four compartments mentioned above. This paper presents the process of applying NitroScape to European landscapes and preliminary results on flows and GHG budgets.

Row spacing effects on light interception and extinction coefficient have been inconsistent for maize ( Zea mays L.) when calculated with field measurements. To avoid inconsistencies due to variable light conditions and variable leaf canopies, we used a model to describe three-dimensional (3D) shoot structures combined with a model of 3D light transfer. The MODICA model mimics 3D shoot structures of maize plants from digitizations in the field and makes it possible to simulate associated hypothetical canopies by re-arranging plants into different row spacings. All row spacings examined with the model had 10 plants m −2 . By using the light model RIRI, simulations showed the relative importance of development stage and time integration on fraction of light transmitted and on the extinction coefficient. Narrow row spacings consistently had less transmitted light and greater values of extinction coefficient. This modelling tool shows promise to effectively evaluate row spacing to optimize light interception.

BACKGROUND AND AIMS Nitrogen partitioning within stands has been described fairly comprehensively, especially for C(3) plants in dense stands where the horizontal heterogeneity of foliage distribution is relatively small. Nitrogen has been shown to be distributed vertically and in parallel to light, maximizing carbon assimilation and stand productivity. Conversely, row crops such as maize (C(4) plants) are characterized by strong horizontal heterogeneity of foliage distribution, and a three-dimensional (3D) approach is required to investigate the combined effect of spatial distribution of nitrogen and light on canopy photosynthesis. MODEL The 3D geometry of maize canopies was modelled with varying densities and at different developmental stages using plant digitizing under field conditions. For lamina parts, photosynthesis was measured and nitrogen content per unit area (N(a)) was described from analysis of nitrogen content per unit mass (N(m)) and dry mass per unit area (M(a)). Hyperbolic relationships between photosynthesis at irradiance saturation (P(max)) and N(a) were established as well as a linear relationship between dark respiration (R(d)) and N(a), whereas quantum efficiency (alpha) was found to be independent of N(a). KEY RESULTS AND CONCLUSIONS N(m), M(a) and N(a) were shown to change over time vertically (i.e. between laminae), which has been largely reported previously, and horizontally (i.e. within laminae), which has scarcely been described previously. Even if M(a) played a major role in N(a), a strong relationship between N(a) and M(a) could not be demonstrated, whereas several previous studies have found that N(a) was essentially related to M(a) rather than N(m). From simulations of radiative exchange using a 3D volume-based approach and lamina photosynthesis using a hyperbola, it was shown that real patterns of N(a) partitioning could increase daily crop photosynthesis by up to 8 % compared with uniform patterns of N(a), especially for the earliest stages of stand development.

Modelling crop architecture and light interception by maize usually assumes a uniform distribution of leaf azimuth. A developmental study of the spatial display of the shoot structure was carried out during the vegetative growth through a field experiment with four treatments: two initial plant orientations (one treatment with the first leaves of all plants initially perpendicular to the row direction and another treatment with the first leaves of all plants initially parallel to the row direction) and two densities (about 10 plants/m 2 and 20 plants/m 2 ). The position of each leaf was characterized over time using three parameters: leaf azimuth, leaf base inclination, and leaf insertion height. A circular data analysis was necessary to study leaf angle (azimuth, inclination). The first leaves grow in the same vertical plane until leaf 8 or 9. For these leaves, there was a low dispersion of leaf azimuths around the initial plant azimuth. For upper leaves however, the dispersion of leaf azimuths increased, tending to a uniform distribution for leaves 12 to 14. We observed no effect of initial plant orientation on final leaf mean azimuths. The only important effect of initial plant azimuthal clumping was to increase the trend toward azimuthal dispersion. For an individual leaf, azimuthal movements occurred mainly during the growing period, but some movements could occur between ligulation and silking too. These movements were caused at least in part by a twisting of the internodes, and may also result from the twisting of the sheaths and the nodes. These results, especially for the upper leaves, are in contrast with those of previous workers with comparable experimental design, but in growth chamber conditions, in which leaves tended to be orientated more and more perpendicular to the row, due to internode torsion. The effects of initial plant orientation on leaf inclination and height were less spectacular. Initial plant orientation and density had a small effect on leaf base inclinations. An effect of density was observed only on leaf insertion heights, but no intercalation of leaves between adjacent plants was noticed.