In Morocco, cereal production is below the expected potential. The adoption of best agricultural practices that reduce vulnerability to climate is a major requirement. No-tillage (NT) agriculture is a system that could improve cereal production by enhancing soil fertility. Some factors, in combination with no-tillage, can further improve cereal yields, especially the choice of variety the adequate fertilization. The objective of our study is to investigate the effect of no-tillage agriculture and nitrogen fertilization on soil fertility and the yield of five durum wheat varieties developed in Morocco in a long-term (18 years) NT trial at the INRA Merchouch experimental station, Morocco. The results show that tillage type had a significant effect on soil organic carbon and CEC (measured before the start of the experiment), as well as on ammonium and nitrates (measured at the end of the experiment), whereas nitrogen dose had a significant effect on total nitrogen and nitrates (measured at the end of the experiment). Regarding wheat yield, as measured at the end of the experiment during the 2020–2021 cropping season, the results show that, under NT, the varieties Nachit, Faraj, and Louiza had grain yields of 4.5, 4.3, and 3.4 t ha⁻¹ and straw yields of 9.8, 7.8, and 6.8 t ha⁻¹, respectively, whereas the I.C and M.G germoplasms had grain yields of 4.05 and 3.72 t ha⁻¹ and straw yields of 8.25 and 8.39 t ha⁻¹, respectively. These values are low for a favorable area and correspond to a semi-arid area. In addition, no effects of nitrogen dose were observed due to water stress, which reduced nitrogen use efficiency. Nachit is the most adapted variety under NT, with the highest yield (5.1 t ha⁻¹) under a low dose of nitrogen (20 kg N ha⁻¹), followed by Faraj (4.7 t ha⁻¹), still under NT but with a higher nitrogen dose (40 kg N ha⁻¹) and Nachit (4.5 t ha⁻¹) with the minimum nitrogen dose but under both NT and CT. In conclusion, Nachit and Faraj wheat varieties performed the best under no-tillage conditions with the minimal nitrogen dose. However, grain yield values were reduced, owing to water stress, which reduced nitrogen use efficiency.

Conservation agriculture (CA), which could contribute to sustainable agriculture, maintains or improves soil nitrogen fertility by eliminating tillage (no-tillage). Quantitative assessment of soil constituents is enhanced by stable isotope techniques such as ¹⁵N, which are used to better understand nitrogen dynamics. This study was therefore carried out to assess the impact of tillage type and fertilizer application on soil and plant nitrogen fractionation. The trial consisted of two tillage types: no-tillage (NT) and conventional tillage (CT). Three nitrogen doses (82, 115, and 149 kg ha⁻¹) were applied. The experimental design was a randomized complete block with three replications. The Louiza variety of durum wheat was used in this study. Soil nitrogen sequestration was assessed using the stable nitrogen isotope (¹⁵N) method. The statistical analysis (ANOVA) showed that, overall, there was no significant difference between tillage types and nitrogen doses for grain and straw yields and grain total nitrogen. In contrast, the effect of both factors and their interaction were significant for straw total nitrogen. There was no difference between tillage types for grain nitrogen use efficiency (NUE), even though NT was superior to CT by 3.5%, but nitrogen doses had a significant effect and a significant interaction with tillage type. When comparing nitrogen doses for each tillage type separately, results showed that the average NUE for grain was 20.5, 8.4, and 16.5%, respectively, for the three nitrogen doses for CT compared with 26.8, 19.0, and 30.6% for NT, indicating clearly the better performance of NT compared to CT. Regarding straw, the NUE is 3.2, 3.5, and 5.4% for CT compared with 3.4, 4.9, and 9.2% for NT. NUE in grain and straw under no-tillage was higher than under conventional tillage in all three nitrogen doses. These results show that soil conservation techniques such as no-tillage and the integrated application of nitrogen fertilizer can be good strategies for reducing soil nitrogen losses.

In the context of increasing food fraud risks and the need for sustainable supply chain management, this study explores a novel multi-criteria approach combining spectroscopic (ATR-FTIR) and isotopic (δ13C, δ15N, δ18O) analyses with advanced chemometric models for the authentication and traceability of sesame oils. By applying principal component analysis (PCA), hierarchical clustering, and classification algorithms (PLS-DA, Random Forest), this integrative methodology enables the robust identification of adulteration with other oils (e.g., soybean, sunflower), and discrimination of oils based on their geographical origin. The incorporation of these analytical features into multi-criteria decision-making (MCDM) frameworks allows for a more reliable and transparent system to support food quality control, consumer trust, and regulatory compliance. This chapter emphasizes the potential of coupling analytical chemistry with decision support tools to promote integrity also in the digital era where traceability and data-driven monitoring are essential

The adulteration of sesame oil poses significant challenges for public health, food safety, and environmental sustainability, requiring advanced analytical and decision-making tools for detection. This study presents an integrated framework combining Fourier Transform Infrared (FTIR) spectroscopy, δ13C isotope ratio mass spectrometry, and artificial intelligence within an environmental engineering perspective. FTIR spectra were processed using an unsupervised deep autoencoder, producing a Spectral Deviation Index (SDI) with excellent discriminatory power (AUC = 1.00). In parallel, a Box-Behnken design quantified δ13C isotopic responses, while machine learning models, including Random Forest (R2 = 0.93) and symbolic regression, accurately predicted adulteration patterns even at low levels. Symbolic regression provided interpretable equations supporting traceability and quantitative decision-making. The hybrid methodology is rapid, non-destructive, and automatable, offering a scalable pathway for sustainable food quality control and environmental monitoring.

Background: This study investigated the use of stable isotope ratio analysis to trace the geographical origins of Moroccan dairy products, as well as possible applications to other Moroccan agricultural products. Ten agro-ecological zones with different climates were selected, including the Garb, Middle Atlas, Meknes, Lukos, Agadir, Taroudant, Settat, andCasablanca regions. Milk samples were collected directly from cows on modern andtraditional farms, as well as the water they consumed. Methods: Analytical methods such as isotope ratio mass spectrometry (IRMS) andthe Piccaro system were used to determine stable isotope ratios in milk andwater samples. Chemometric methods including principal component analysis (PCA) andhierarchical cluster analysis (HCA) showed that stable isotope ratios (H, C, N, O) in milk were highly correlated with environmental factors specific to the region. Three distinct groups emerged, especially for oxygen isotopes, depending on altitude anddistance from the ocean. The δ18O composition was particularly distinct, indicating regional differences: northern, low-altitude coastal areas andhigh-altitude inl andareas. Result: The study also showed that the isotopic composition of milk is influenced by the isotopic composition of drinking water. Nitrogen isotopes, especially δ15N, distinguished two groups of milk: one from an area with fertilizer use (low δ15N values, intensive agriculture) andanother from an area with moderate agricultural practices (higher δ15N values). These results highlight the potential of stable isotope analysis as a tool to verify the geographical origin of dairy products in Morocco.

This chapter presents an explainable artificial intelligence framework for detecting trace-level adulteration in sesame oil using Fourier Transform Infrared spectroscopy. Due to its high economic value, sesame oil is often adulterated with cheaper oils or paraffin, creating significant challenges for reliable detection. The proposed approach integrates multivariate spectral analysis with machine learning techniques to improve both predictive performance and model interpretability. Principal Component Analysis and Partial Least Squares Discriminant Analysis are used for spectral exploration, while XGBoost identifies key discriminatory infrared bands. Symbolic regression generates explicit mathematical equations describing adulteration patterns, and a one-dimensional convolutional neural network with self-attention extracts complex spectral features for classification tasks. The results demonstrate high sensitivity in fraud detection and reliable identification of adulterant types, highlighting the potential of combining spectroscopy and explainable AI for food authenticity assessment.