Wyniki wyszukiwania - Biblioteka Nauki

Ten serwis zostanie wyłączony 2025-02-11.

Nowa wersja platformy, zawierająca wyłącznie zasoby pełnotekstowe, jest już dostępna.
Przejdź na https://bibliotekanauki.pl

Ograniczanie wyników

2 Ecological Engineering & Environmental Technology

2 2023

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Sortuj według:

Ogranicz wyniki do:

Deep Learning Approach for Runoff Prediction – Evaluating the Long-Short-Term Memory Neural Network Architectures for Capturing Extreme Discharge Events in the Ouergha Basin, Morocco

100%

Karmouda N. , Bouramtane T. , Tahiri M. , Kacimi I. , Leblanc M. , Kassou N.

Ecological Engineering & Environmental Technology

2023

tom Vol. 24, iss. 9

271--290

Rainfall-runoff modeling plays a crucial role in achieving efficient water resource management and flood forecasting, particularly in the context of increasing intensity and frequency of extreme meteorological events induced by climate change. Therefore, the aim of this research is to assess the accuracy of the Long-Short-Term Memory (LSTM) neural networks and the impact of its architecture in predicting runoff, with a particular focus on capturing extreme hydrological discharges in the Ouergha basin; a Moroccan Mediterranean basin with historical implications in many cases of flooding; using solely daily rainfall and runoff data for training. For this purpose, three LSTM models of different depths were constructed, namely LSTM 1 single-layer, LSTM 2 bi-layer, and LSTM 3 tri-layer, their window size and hyperparameters were first tuned, and on seven years of daily data they were trained, then validated and tested on two separate years to ensure the generalization on unseen data. The performance of the three models was compared using hydrogram-plots, Scatter-plots, Taylor diagrams, and several statistical metrics. The results indicate that the single-layer LSTM 1 outperforms the other models, it consistently achieves higher overall performance on the training, validation, and testing periods with a coefficient of determination R-squared of 0.92, 0.97, and 0.95 respectively; and with Nash-Sutcliffe efficiency metric of 0.91, 0.94 and 0.94 respectively, challenging the conventional beliefs about the direct link between complexity and effectiveness. Furthermore, all the models are capable of capturing the extreme discharges, although, with a moderate und erprediction trend for LSTM 1 and 2 as it does not exceed -25% during the test period. For LSTM 3, even if its underestimation is less pronounced, its increased error rate reduces the confidence in its performance. This study highlights the importance of aligning model complexity with data specifications and suggests the necessity of considering unaccounted factors like upstream dam releases to enhance the efficiency in capturing the peaks of extreme events.

Landslide Hazard Assessment in the Heterogeneous Geomorphological and Environmental Context of the Rif Region, Morocco – A Machine Learning Approach

84%

Hamidi M. , Bouramtane T. , Abraham S. , Kacimi I. , Barbiero L. , Kassou N. , Valles V. , Levy G.

Ecological Engineering & Environmental Technology

2023

tom Vol. 24, iss. 8

272--292

Landslides are considered to be one of the most significant and critical natural hazards in the heterogeneous geomor-phological setting of the Rif region of Morocco. Despite the high susceptibility to landslides, the region lacks detailed studies. Therefore, this research introduces four advanced machine learning methods, namely Support Vector Machine (SVM), Classification and Regression Trees (CART), Multivariate Discriminant Analysis (MDA), and Logistic Regression (LR), to perform landslide susceptibility mapping, as well as study of the connection between landslide occurrence and the complex regional geo-environmental context of Taounate province. Fifteen causative factors were extracted, and 255 landslide events were identified through fieldwork and satellite imagery analysis. All models performed very well (AUC > 0.954), while the CART model performed the best (AUC= 0.971). However, SVM demonstrated superior performance compared to other methods, achieving the highest accuracy (89.92%) and F1-measure (81.66%) scores on the training data, and the highest accuracy (83.01%), precision (81.74%), and specificity (79.46%) scores on the test data. The results do not necessarily indicate that LR and MDA have the lowest predictive ability, as they demonstrated high accuracy in terms of AUC and in some classification tasks. Moreover, they provide the significant advantage of easy interpretation of the geo-environmental processes that control landslides. Rainfall is the primary triggering factor of landslides in the study area. The majority of landslides occurred on slopes, particularly those located along rivers and faults, suggesting that landslides in the region are closely associated with active tectonics and precipitation. All four models predicted similar spatial distribution patterns in landslide susceptibility. The results showed that almost half of the area mainly in the north and northwest, has a very high susceptibility to landslides. The findings provide valuable references for land use management and the implementation of effective measures for landslide prevention.