Flash flood susceptibility mapping using a novel deep learning model based on deep belief network,back propagation and genetic algorithm

Flash floods are responsible for loss of life and considerable property damage in many countries.Flood susceptibility maps contribute to flood risk reduction in areas that are prone to this hazard if appropriately used by landuse planners and emergency managers.The main objective of this study is to prepare an accurate flood susceptibility map for the Haraz watershed in Iran using a novel modeling approach(DBPGA) based on Deep Belief Network(DBN) with Back Propagation(BP) algorithm optimized by the Genetic Algorithm(GA).For this task, a database comprising ten conditioning factors and 194 flood locations was created using the One-R Attribute Evaluation(ORAE) technique.Various well-known machine learning and optimization algorithms were used as benchmarks to compare the prediction accuracy of the proposed model.Statistical metrics include sensitivity,specificity accuracy, root mean square error(RMSE), and area under the receiver operatic characteristic curve(AUC) were used to assess the validity of the proposed model.The result shows that the proposed model has the highest goodness-of-fit(AUC = 0.989) and prediction accuracy(AUC = 0.985), and based on the validation dataset it outperforms benchmark models including LR(0.885), LMT(0.934), BLR(0.936), ADT(0.976), NBT(0.974), REPTree(0.811), ANFIS-BAT(0.944), ANFIS-CA(0.921), ANFIS-IWO(0.939), ANFIS-ICA(0.947), and ANFIS-FA(0.917).We conclude that the DBPGA model is an excellent alternative tool for predicting flash flood susceptibility for other regions prone to flash floods.     

Investigation of spatio-temporal patterns of seasonal streamflow droughts in a semi-arid region

Natural Hazards - Spatio-temporal patterns of seasonal streamflow droughts were investigated for the semi-arid Karkheh watershed, located in western Iran with an area of 41,470 km2,...     

Groundwater potential recharge estimation in bare soil using three soil moisture accounting models: field evaluation for a semi-arid foothill region

Performances of conventional and improved soil moisture balance as well as locally calibrated empirical models were evaluated in simulating potential recharge (R) and soil moisture content for a semi-arid foothill region. Models comparison with observed values using lysimeter data during [(2011–2012), (2012–2013)] reveal poor performance of conventional soil moisture balance model, underestimating annual R values. Improved soil moisture balance model provided acceptable estimation of annual R for 2011–2012 by considering the wetting of the near surface soil storage. However, it produced the worst simulation for daily soil moisture content once rainy season was over. Sensitivity analysis revealed that the precision degree of initial soil moisture deficit value would strongly influence estimation of R by improved soil moisture balance model, which can be viewed as a limiting factor. Additionally, locally calibrated model produced the best estimation of annual R and daily soil moisture content, which is suggested for the study region.     

Moment tensor inversion of nine events in Iran using INSN data

In this research, moment tensor (MT) inversion procedure in frequency domain is applied to nine events in order to evaluate the data efficiencies in computing source parameters of earthquakes in Iran. The events are recorded by Iranian National Seismological Network (INSN) stations, currently comprising 22 broadband stations. For the purpose of this study, first, the inversion procedure is applied to obtain source parameters. Then, the results obtained here are compared with those of Harvard global centroid moment tensors (GCMT) in order to estimate their reliabilities. The results show reasonable consistencies with those of Harvard GCMT within 0.075 uncertainties in logarithmic scalar moment or 0.05 in moment magnitude and 20° in Kagan angle for focal angles. Furthermore, according to the present research, the calculated source parameters are not significantly sensitive to the azimuthal gap.