In wheat cultivation, it is necessary to improve fertilizer application to increase the yield and the grain quality, to limit nitrogen losses and to optimize economic efficiency. The aim of this paper is to present a dynamic crop growth model based on real time data acquired by wireless microsensors and to evaluate its performance.

The field experiments were carried out on a silty soil in Belgium. The crop responses (wheat, Triticum aestivum L.) were measured under seven different nitrogen applications rates, from 0 to 240 kg [N]/ha. The eKo pro series system (Crossbow) was used as wireless monitoring system. It has the ability to extend the network as big as necessary to cover the field spatial heterogeneity. The research made use of several sensors, such as soil moisture, water content and temperature (at 2 depths), canopy and air temperature and humidity, solar radiation. The plant characteristics (LAI) were also regularly measured.

On the other hand, the crop model STICS (INRA-France) was selected because it provides insight into the mechanisms of plant development by taking into account the cultivation techniques, the soil and atmosphere system. The model was adapted to the specific wheat cultivar. The daily microclimate data issued from the wireless network were introduced into the model and the continuous interactions between all the inputs, and their daily evolution were simulated.

At the end of this first year, the model results, in terms of biomass growth, N-uptake by the crop, water and nitrogen balance of the soil were in close agreement with the experimental data. As an example, the simulated normalised grain yield at 15% water content was 11.53 t/ha, while the measured value reached 11.89 ± 0.41 t/ha. The methodology has the potential to be used as a tool for managing the nitrogen applications (date of application and rate).