This study presents two machine learning models, namely, the light gradient boosting machine (LightGBM) and categorical boosting (CatBoost), for the first time for predicting flash flood susceptibility (FFS) in the Wadi System (Hurghada, Egypt). A flood inventory map with 445 flash flood sites was produced and randomly divided into two groups for training (70%) and testing (30%). Fourteen flood controlling factors were selected and evaluated for their relative importance in flood occurrence prediction. The performance of the two models was assessed using various indexes in comparison to the common random forest (RF) method. The results show areas under the receiver operating characteristic curves (AUROC) of above 97% for all models and that LightGBM outperforms other models in terms of classification metrics and processing time. The developed FFS maps demonstrate that highly populated areas are the most susceptible to flash floods. The present study proves that the employed algorithms (LightGBM and CatBoost) can be efficiently used for FFS mapping.
The availability of rainfall data is fundamental for most hydrological analyses and for the design and management of water resources systems. There are many methods of spatial interpolation of rainfall despite their effectiveness and ease of use variables, the choice of one or the other depends on the aim of the study, but also on the territorial context of the area (surface, network density, etc.). The study area is the coastal city of Annaba located in eastern Algeria, and which contains a low rainfall network density. The purpose of this study is to determine the most adequate rainfall interpolation technique. Three interpolation techniques, determinist and stochastic, have been tested by visual observation and cross-validation, the best results were achieved by inverse distance weighting to characterize rainfall distribution
1 June 2021 in Environmental Science and Pollution Research
In arid and semi-arid areas where desertification is progressing, exploitation of water resources is focused on surface water and groundwater who pose sustainability problem. The exploitation of surface water uses “surface dam” heavily exposed to evaporation and cannot perform the function of “dam-reservoir” in the dry season. We envision the use of groundwater by the underground dams, designed to contain groundwater and accumulate water.Compared with conventional dam, underground dams have the advantage of not overwhelm the land, offer particularly low evaporation losses and preserve the environment. A DEM and the river network of the study area were obtained by ArcGIS software The results obtained were injected into Modflow to extract the discharge. The dam body stability has been verified and validated using a model of the fluid-structure interaction under the platform “COMSOL Multiphysics”.A significant gain in water volume is found by storing away the long intense sunshine and evaporation. COMSOL allows us to have an optimal design of the dam body.The development of the region requires necessary a control of water resources
Floods are frequent hazard in Algeria. They cause severe casualties, destroy infrastructures, and impair economies. In the past decades, Algeria experienced devastating floods. The dominant type of occurring floods are flash floods, which tend to be not well documented and studied in Algeria. This chapter presents a brief introduction to the flood phenomena within the Algerian climatic and management context, based on databases, scientific publications, and local technical reports. Existing studies about floods are reviewed. It also provides an analysis of the most disastrous floods that occurred in the past decades. Of the most noteworthy flash floods, a highlight of the Bab El Oued flash flood occurring in a heavily urbanized setting and the M’zab Valley flash flood, which took place in a UNESCO World Heritage Site. The monitoring network in Algeria is presented and data availability is discussed. The implementation of the first forecasting and early warning system are also presented. Different aspects of flash floods were presented including the effect of the increase of urbanization, the influence of climate change and the adopted strategies of flood risk management. Heavy and increasing urbanization and population growth increased the flood vulnerability and this trend must be mitigated.