Votre recherche
Résultats 2 ressources
-
This paper presents a new framework for floodplain inundation modeling in an ungauged basin using unmanned aerial vehicles (UAVs) imagery. This method is based on the integrated analysis of high-resolution ortho-images and elevation data produced by the structure from motion (SfM) technology. To this end, the Flood-Level Marks (FLMs) were created from high-resolution UAV ortho-images and compared to the flood inundated areas simulated using the HEC-RAS hydraulic model. The flood quantiles for 25, 50, 100, and 200 return periods were then estimated by synthetic hydrographs using the Natural Resources Conservation Service (NRCS). The proposed method was applied to UAV image data collected from the Khosban village, in Taleghan County, Iran, in the ungauged sub-basin of the Khosban River. The study area is located along one kilometre of the river in the middle of the village. The results showed that the flood inundation areas modeled by the HEC-RAS were 33%, 19%, and 8% less than those estimated from the UAV’s FLMs for 25, 50, and 100 years return periods, respectively. For return periods of 200 years, this difference was overestimated by more than 6%, compared to the UAV’s FLM. The maximum flood depth in our four proposed scenarios of hydraulic models varied between 2.33 to 2.83 meters. These analyses showed that this method, based on the UAV imagery, is well suited to improve the hydraulic modeling for seasonal inundation in ungauged rivers, thus providing reliable support to flood mitigation strategies
-
In cold regions, ice jams frequently result in severe flooding due to a rapid rise in water levels upstream of the jam. Sudden floods resulting from ice jams threaten human safety and cause damage to properties and infrastructure. Hence, ice-jam prediction tools can give an early warning to increase response time and minimize the possible damages. However, ice-jam prediction has always been a challenge as there is no analytical method available for this purpose. Nonetheless, ice jams form when some hydro-meteorological conditions happen, a few hours to a few days before the event. Ice-jam prediction can be addressed as a binary multivariate time-series classification. Deep learning techniques have been widely used for time-series classification in many fields such as finance, engineering, weather forecasting, and medicine. In this research, we successfully applied convolutional neural networks (CNN), long short-term memory (LSTM), and combined convolutional–long short-term memory (CNN-LSTM) networks to predict the formation of ice jams in 150 rivers in the province of Quebec (Canada). We also employed machine learning methods including support vector machine (SVM), k-nearest neighbors classifier (KNN), decision tree, and multilayer perceptron (MLP) for this purpose. The hydro-meteorological variables (e.g., temperature, precipitation, and snow depth) along with the corresponding jam or no-jam events are used as model inputs. Ten percent of the data were excluded from the model and set aside for testing, and 100 reshuffling and splitting iterations were applied to 80 % of the remaining data for training and 20 % for validation. The developed deep learning models achieved improvements in performance in comparison to the developed machine learning models. The results show that the CNN-LSTM model yields the best results in the validation and testing with F1 scores of 0.82 and 0.92, respectively. This demonstrates that CNN and LSTM models are complementary, and a combination of both further improves classification.