Accurate nitrogen (N) management is crucial for the economic and environmental sustainability of cropping systems. Different methods have been developed to increase the efficiency of N fertilizations. However, their costs and/or low usability have often prevented their adoption in operational contexts. We developed a diagnostic system to support topdressing N fertilization based on the use of smart apps to derive a N nutritional index (NNI; actual/critical plant N content). The system was tested on paddy rice via dedicated field experiments, where the smart apps PocketLAI and PocketN were used to estimate, respectively, critical (from leaf area index) and actual plant N content. Results highlighted the system’s capability to correctly detect the conditions of N stress (NNI < 1) and N surplus (NNI > 1), thereby effectively supporting topdressing fertilizations. A resource-efficient methodology to derive PocketN calibration curves for different varieties—needed to extend the system to new contexts—was also developed and successfully evaluated on 43 widely grown European varieties. The widespread availability of smartphones and the possibility to integrate NNI and remote sensing technologies to derive variable rate fertilization maps generate new opportunities for supporting N management under real farming conditions.

Cultivar recommendation is a key factor in cropping system management. Classical approaches based on comparative multi-environmental trials can hardly explore the agro-climatic and management heterogeneity farmers may have to face. Moreover, they struggle to keep up with the number of genotypes commercially released each year. We propose a new approach based on the integration of in silico ideotyping and functional trait profiling, with the common bean (Phaseoulus vulgaris L.) in Northern Italy as a case study. Statistical distributions for six functional traits (light extinction coefficient, radiation use efficiency, thermal time to first pod and maturity, seed weight, plant height) were derived for 24 bean varieties. The analysis of soil, climate and management in the study area led us to define 21 homogeneous contexts, for which ideotypes were identified using the crop model STICS (Simulateur mulTIdisciplinaire pour les Cultures Standard), the E-FAST (Extended Fourier Amplitude Sensitivity Test) sensitivity analysis method, and the distributions of functional traits. For each context, the 24 cultivars were ranked according to the similarity (weighted Euclidean distance) with the ideotype. Context-specific ideotypes mainly differed for phenological adaptation to specific combinations of climate and management (sowing time) factors, and this reflected in the cultivar recommendation for the different contexts. Feedbacks from bean technicians in the study area confirmed the reliability of the results and, in turn, of the proposed methodology.