Accurate seasonal forecasting of cereal yields is an important decision support tool for countries, such as Morocco, that are not self-sufficient in order to predict, as early as possible, importation needs. This study aims to develop an early forecasting model of cereal yields (soft wheat, barley and durum wheat) at the scale of the agricultural province considering the 15 most productive over 2000–2017 (i.e., 15 × 18 = 270 yields values). To this objective, we built on previous works that showed a tight linkage between cereal yields and various datasets including weather data (rainfall and air temperature), regional climate indices (North Atlantic Oscillation in particular), and drought indices derived from satellite observations in different wavelengths. The combination of the latter three data sets is assessed to predict cereal yields using linear (Multiple Linear Regression, MLR) and non-linear (Support Vector Machine, SVM; Random Forest, RF, and eXtreme Gradient Boost, XGBoost) machine learning algorithms. The calibration of the algorithmic parameters of the different approaches are carried out using a 5-fold cross validation technique and a leave-one-out method is implemented for model validation. The statistical metrics of the models are first analyzed as a function of the input datasets that are used, and as a function of the lead times, from 4 months to 2 months before harvest. The results show that combining data from multiple sources outperformed models based on one dataset only. In addition, the satellite drought indices are a major source of information for cereal prediction when the forecasting is carried out close to harvest (2 months before), while weather data and, to a lesser extent, climate indices, are key variables for earlier predictions. The best models can accurately predict yield in January (4 months before harvest) with an R² = 0.88 and RMSE around 0.22 t. ha⁻¹. The XGBoost method exhibited the best metrics. Finally, training a specific model separately for each group of provinces, instead of one global model, improved the prediction performance by reducing the RMSE by 10% to 35% depending on the provinces. In conclusion, the results of this study pointed out that combining remote sensing drought indices with climate and weather variables using a machine learning technique is a promising approach for cereal yield forecasting.

Water stress is one of the factors controlling agricultural land salinization and is also a major problem worldwide. According to FAO and the most recent estimates, it already affects more than 400 million hectares. The Tafilalet plain in Southeastern Morocco suffers from soil salinization. In this regard, the GIS tools and remote sensing were used in the processing of 19 satellite images acquired from Landsat 4–5, (Landsat 7), (Landsat 8), and (Sentinel 2) sensors. The most used indices in the literature were (16 indices) tested and correlated with the results obtained from 25 samples taken from the first soil horizon at a constant depth of 0.20 m from the 2018 campaign. The linear model, at first, allows the selection of five better indices of the soil salinity discrimination (SI-Khan, VSSI, BI, S3, and SI-Dehni). These last indices were the subject of the application of a logarithmic model and polynomial models of degree two and four to increase the prediction of saline soil.. After studies and analysis, we concluded that the second-degree polynomial model of the salinity index (SI-KHAN) is the most efficient one for detecting and mapping soil salinity in the Tafilalet oasis, with a coefficient of determination (R²) and the Nash–Sutcliffe efficiency (NSE) equal to 0.93 and 0.86, respectively. Percent bias (PBIAS) calculated for this model equal was 1.868% < 10%, and the low value of the root mean square error (RMSE) confirms its very good performance. The drought cyclicity led to the intensification of the soil salinization process and accelerated soil degradation. The standardized precipitation anomaly index (SPAI) is strongly correlated to soil salinity. The hydroclimate condition is the factor that further controls this phenomenon. An increase in salinized surfaces is observed during the periods of 1984–1996 and 2000–2005, which cover a surface of 11.50 and 24.20 km², respectively, while a decrease of about 50% is observed during the periods of 1996–2000 and 2005–2018.