Wyniki wyszukiwania - BazTech

Ograniczanie wyników

2 Ecological Engineering & Environmental Technology

2 2025

Powiadomienia systemowe

Sesja wygasła!

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Wyszukiwano:
w słowach kluczowych: Fez-Meknes region

Sortuj według:

Ogranicz wyniki do:

Impact of rainfall variability on cereal yield in the Fez-Meknes region (Morocco)

Belmahi Mohamed, Hanchane Mohamed, Benbaqqal Hicham, Kabach Mohammed, Zouhri Driss, Kessabi Ridouane, Elkhazzan Bouchta, Zriouli Khalid

Ecological Engineering & Environmental Technology

2025

Vol. 26, iss. 2

109--121

The Fez-Meknes region is distinguished by its agricultural vocation and its emergence as a hub in the agro-food industry. This study aims to assess the main crop yield, production, and percentage of the agricultural area within each province of the Fez-Meknes region from 2000 to 2020, based on an analysis of descriptive statistics and cartography data. The objective is to determine the national ranking of autumn cereals within the region. Then, multiple linear regression between precipitation and cereal yield in the region’s provinces was established, and the trend in sown areas and cereal yield was analysed using the Man Kendall test. The results revealed that the area sown to autumn cereals accounted for 15% of the national cereal area. Despite that, regional cereal production is ranked second nationally after the Casablanca-Settat region, with a small difference that does not exceed 1.5%. Regarding regional provinces, Taounate and Taza account for almost half of the region’s cereal production. The correlation coefficient between monthly precipitation and cereal yield ranged from 0.51 in Boulmane province to 0.84 in Fez and Moulay Yaacoub province. The coefficient of determination ranged from 0.21 in Boulmane province to 0.70 in Fez province. On the other hand, precipitation in November, December, January, and March had the greatest impact on cereal yields. The differences between observed and estimated yields using multiple regression are acceptable in all region’s provinces, especially when only one predictor was retained. Finally, the Man-Kendall test indicates that the area sown to autumn cereals has a slight downward trend of 4965 ha/year, with a significance of α = 0.07. However, cereal yield also tends to increase by 0.34 q/year with a p- value α = 0.12.

Developing an explainable and interpretable machine learning model for flood susceptibility mapping

Khaldi Loubna, El Bilali Ali, El Abed Alae, Krakauer Nir, El Khanchoufi Abdessalam

Ecological Engineering & Environmental Technology

2025

Vol. 26, iss. 1

201--215

This study evaluates flood susceptibility in the Fez-Meknes region of Morocco by comparing the performance of five machine learning (ML) models using 14 environmental variables. The selected models, including Random Forest (RF), Support Vector Machine (SVM), K-Nearest Neighbor (KNN), Recursive Partitioning and Regression Trees (RPART), and Logistic Regression (LR), were assessed for prediction accuracy and enhanced with Partial Dependence Plots (PDP) and Local Interpretable Model-Agnostic Explanations (LIME) to increase interpretability. Results indicate that the RF model outperforms other models, achieving a high prediction accuracy with an AUC of 96%, low Mean Absolute Error (MAE) of 0.26, and Root Mean Squared Error (RMSE) of 0.31, along with strong Nash-Sutcliffe Efficiency (NSE) and correlation coefficient (R²). Through PDP and LIME, the primary factors influencing flood susceptibility were identified as proximity to rivers, drainage density, slope, NDVI (Normalized Difference Vegetation Index), TRI (Terrain Roughness Index), and LULC (Land Use and Land Cover). These findings highlight the potential of interpretable ML models to enhance flood risk assessment, providing valuable insights for urban planning and flood mitigation strategies in vulnerable regions.