Digital agriculture has rapidly evolved into a cornerstone for tackling global challenges such as climate change, food security, and sustainable land management. Its convergence with artificial intelligence, agronomic modeling, and emerging crops like Cannabis sativa for medical purposes highlights new opportunities for both productivity and territorial development. This systematic review examines peer-reviewed studies published between 2000 and 2025, alongside major policy and institutional reports, to examine recent advances in digital agriculture and their implications for sustainability. The analysis highlights the growing role of artificial intelligence in agriculture, particularly through machine learning and deep learning techniques, which have significantly improved performance in yield prediction, stress detection, and pest monitoring. At the same time, crop modeling systems such as DSSAT, APSIM, STICS, and AquaCrop remain essential tools for simulating crop–soil–climate interactions and supporting decision-making processes. Recent progress in data integration, combining remote sensing products with in situ observations, has strengthened model robustness and enabled the first applications of agricultural digital twins for real-time scenario analysis. Emerging crops, and especially medical cannabis, illustrate both opportunities and constraints: they offer diversification and high economic value, but are also associated with high energy consumption in indoor facilities and regulatory uncertainty. In response, innovative digital solutions based on IoT technologies and blockchain systems have been proposed to improve monitoring, traceability, and transparency along the supply chain. Overall, the literature confirms that the combined use of artificial intelligence, crop models, and emerging crops can serve as a powerful driver of sustainable agricultural and territorial development. However, persistent gaps include methodological heterogeneity, limited interoperability between platforms, and uneven access to digital tools across rural areas. Future research should focus on creating open interoperable datasets, scaling hybrid AI-modeling frameworks into operational digital twin systems, and designing inclusive policies that reduce the digital divide. Only through a systemic integration of technological innovation, environmental sustainability, and social equity can digital agriculture fully realize its potential as an engine of resilient and inclusive territorial transformation.
Artificial Intelligence, Agronomic Modeling, and Emerging Crops: A Systematic Review on Pathways for Sustainable Agriculture
Nicola Favia
2025-01-01
Abstract
Digital agriculture has rapidly evolved into a cornerstone for tackling global challenges such as climate change, food security, and sustainable land management. Its convergence with artificial intelligence, agronomic modeling, and emerging crops like Cannabis sativa for medical purposes highlights new opportunities for both productivity and territorial development. This systematic review examines peer-reviewed studies published between 2000 and 2025, alongside major policy and institutional reports, to examine recent advances in digital agriculture and their implications for sustainability. The analysis highlights the growing role of artificial intelligence in agriculture, particularly through machine learning and deep learning techniques, which have significantly improved performance in yield prediction, stress detection, and pest monitoring. At the same time, crop modeling systems such as DSSAT, APSIM, STICS, and AquaCrop remain essential tools for simulating crop–soil–climate interactions and supporting decision-making processes. Recent progress in data integration, combining remote sensing products with in situ observations, has strengthened model robustness and enabled the first applications of agricultural digital twins for real-time scenario analysis. Emerging crops, and especially medical cannabis, illustrate both opportunities and constraints: they offer diversification and high economic value, but are also associated with high energy consumption in indoor facilities and regulatory uncertainty. In response, innovative digital solutions based on IoT technologies and blockchain systems have been proposed to improve monitoring, traceability, and transparency along the supply chain. Overall, the literature confirms that the combined use of artificial intelligence, crop models, and emerging crops can serve as a powerful driver of sustainable agricultural and territorial development. However, persistent gaps include methodological heterogeneity, limited interoperability between platforms, and uneven access to digital tools across rural areas. Future research should focus on creating open interoperable datasets, scaling hybrid AI-modeling frameworks into operational digital twin systems, and designing inclusive policies that reduce the digital divide. Only through a systemic integration of technological innovation, environmental sustainability, and social equity can digital agriculture fully realize its potential as an engine of resilient and inclusive territorial transformation.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
