Assessing the susceptibility of schools to flood events in Iran

Application of geographical information system-based analytical hierarchy process modeling for flood susceptibility mapping of Krishna District in Andhra Pradesh

Flooding is one of the most catastrophic natural disasters in terms of provoking socio-economic losses. The current study is to foster a flood susceptibility map of Krishna District in Andhra Pradesh (AP) through integrating remote sensing data, geographical information system (GIS), and the analytical hierarchy process (AHP). Eleven factors, including elevation, slope, aspect, land use/land cover (LULC), drainage density, topographic wetness index, stream power index, lithology, soil, precipitation, and distance from the streams, are considered for identifying and evaluating the spatial distribution of critical flood-susceptible regions. Thematic maps of different factors were derived in GIS using remote sensing data obtained from Sentinel-2A (satellite sensor), shuttle radar topography mission digital elevation model (SRTM DEM v3), and other scientific data products. An analytical hierarchy process is a mathematical approach for decision support, primarily based on the weight and rank of different causative factors. AHP technique is implemented for flood hazard modeling and ascertaining the Flood Hazard Index (FHI) to produce a flood susceptibility map. Different thematic maps weighed with the AHP framework are combined using overlay analysis to produce the flood susceptibility map of the study region. The outcomes of the study demonstrate the potential of GIS and AHP in providing a premise to recognize the vulnerable areas that are susceptible to flood. According to the findings, the Flood Hazard Index is 42% and the study region is classified into very high, high, moderate, low, and very low susceptible, respectively. Following that, historical flood data was used to validate the accuracy of the generated flood susceptibility map. This shows that a maximum of 90% of the data points are within floodplain.

Exploring urban growth-climate change-flood risk nexus in fast growing cities

This study looks at the nexus between urban growth, climate change, and flood risk in Doha, Qatar, a hot-spot, climate change region that has experienced unprecedented urban growth during the last four decades. To this end, this study overviews the main stages of Doha’s urban growth and influencing climatic factors during this period. A physically-based hydrological model was then built to simulate surface runoff and quantify flood risk. Finally, the Pearson correlation was used to verify the potential nexus between flood risk, climate change, and urban growth. Surveying showed that, between 1984 and 2020, urban areas grew by 777%, and bare lands decreased by 54.7%. In addition, Doha witnessed various climatic changes with a notable increase in air temperature (+ 8.7%), a decrease in surface wind speed (− 19.5%), and a decrease in potential evapotranspiration losses (− 33.5%). Growth in urban areas and the perturbation of climatic parameters caused runoff to increase by 422%, suggesting that urban growth contributed more than climatic parameters. Pearson correlation coefficient between flood risk and urban growth was strong (0.83) and significant at p < 0.05. Flood risk has a strong significant positive (negative) correlation with air temperature (wind speed) and a moderate positive (negative) correlation with precipitation (potential evapotranspiration). These results pave the way to integrate flood risk reduction measures in local urban development and climate change adaptation plans.

Image Classification and Land Cover Mapping Using Sentinel-2 Imagery: Optimization of SVM Parameters

Land use/cover (LU/LC) classification provides proxies of the natural and social processes related to urban development, providing stakeholders with crucial information. Remotely sensed images combined with supervised classification are common to define land use, but high-performance classifiers remain difficult to achieve, due to the presence of model hyperparameters. Conventional approaches rely on manual adjustment, which is time consuming and often unsatisfying. Therefore, the goal of this study has been to optimize the parameters of the support vector machine (SVM) algorithm for the generation of land use/cover maps from Sentinel-2 satellite imagery in selected humid and arid (three study sites each) climatic regions of Iran. For supervised SVM classification, we optimized two important parameters (gamma in kernel function and penalty parameter) of the LU/LC classification. Using the radial basis function (RBF) of the SVM classification method, we examined seven values for both parameters ranging from 0.001 to 1000. For both climate types, the penalty parameters (PP) showed a direct relationship with overall accuracy (OA). Statistical results confirmed that in humid study regions, LU/LC maps produced with a penalty parameter >100 were more accurate. However, for regions with arid climates, LU/LC maps with a penalty parameter >0.1 were more accurate. Mapping accuracy for both climate types was sensitive to the penalty parameter. In contrast, variations of the gamma values in the kernel function had no effect on the accuracy of the LU/LC maps in either of the climate zones. These new findings on SVM image classification are directly applicable to LU/LC for planning and environmental and natural resource management.

Suggestion for a new deterministic model coupled with machine learning techniques for landslide susceptibility mapping

Deterministic models have been widely applied in landslide risk assessment (LRA), but they have limitations in obtaining various geotechnical and hydraulic properties. The objective of this study is to suggest a new deterministic method based on machine learning (ML) algorithms. Eight crucial variables of LRA are selected with reference to expert opinions, and the output value is set to the safety factor derived by Mohr-Coulomb failure theory in infinite slope. Linear regression and a neural network based on ML are applied to find the best model between independent and dependent variables. To increase the reliability of linear regression and the neural network, the results of back propagation, including gradient descent, Levenberg-Marquardt (LM), and Bayesian regularization (BR) methods, are compared. An 1800-item dataset is constructed through measured data and artificial data by using a geostatistical technique, which can provide the information of an unknown area based on measured data. The results of linear regression and the neural network show that the special LM and BR back propagation methods demonstrate a high determination of coefficient. The important variables are also investigated though random forest (RF) to overcome the number of various input variables. Only four variables-shear strength, soil thickness, elastic modulus, and fine content-demonstrate a high reliability for LRA. The results show that it is possible to perform LRA with ML, and four variables are enough when it is difficult to obtain various variables.