جستجوی مقالات مرتبط با کلیدواژه « پیش بینی سیل » در نشریات گروه « مهندسی آب »

Floods are one of the most devastating types of natural disasters that every year causes the loss of human lives and properties around the world. The purpose of this study is to evaluate and compare the capability of three machine learning models namely Naïve Bayes Tree (NBTree), Alternating Decision Tree (ADTree), and Random Forest (RF) to predict flood risk in Maneh and Samalqan county. The novelty of the present study is the presentation of tree-based hybrid models that have been less used in previous research. To prepare a flood reference map in the study area, 300 flood-prone locations were identified and were divided into training and validation data sets through random selection with a ratio of 70 to 30. The spatial database of the flood was created using 15 hydrogeological and environmental criteria affecting the flood. Finally, three flood risk prediction maps were generated using NBTree, ADTree, and RF models. To validate the flood risk predicting models, the Area Under the Curve (AUC) factor and the statistical criteria of Positive predictive rate, negative predictive rate, sensitivity, specificity, and accuracy were used. The results showed that the RF model had higher accuracy than the NBTree and ADTree models in predicting flood risk in the study area. The results also showed that the risk of flooding in the central areas of the study area is higher than other areas due to lower altitude and slope.

Parameters uncertainty of rainfall-runoff models are the main sources of uncertainty in real time flood forecasting. In this paper, the Monte Carlo method is used to estimate the uncertainty of the forecasted flood hydrograph due to uncertainty in the calibration parameters of the rainfall-runoff model in Dez Basin in southwestern Iran. Precipitation and air temperature were predicted using Weather Research and Forecasting (WRF) model. The HEC-HMS hydrological model was used to forecast the flood hydrograph corresponding to the predicted precipitation and air temperature. The SCS-CN, Clark Unit Hydrograph, and Muskingum-Cung methods were used to model losses, transform and flood routing, respectively. The results show that the best scheme in WRF model is MYJLG to predict hourly precipitation and air temperature in Dez Basin. Therefore, the MYJ boundary layer scheme, Lin cloud microphysics scheme and GODDARD radiant scheme have the best performance in flood forecasting in Dez basin. In addition, the results of this study show that considering the simultaneous uncertainty in all parameters, the uncertainty in peak discharge of the forecasted flood hydrograph is higher than the uncertainty in the volume of the hydrograph. So that the uncertainty in peak discharge and the volume of forecasted flood hydrograph due to the uncertainty of all parameters are equal to 32.5 and 21.2%, respectively. Thus, with the lack of flood forecasting and warning based on risk, quantifying uncertainty has provided additional information about forecasts that will help decision makers make better decisions.

Nowadays, much effort has been made in improving meteorological forecasts. In this regard, ensemble forecasting systems have been developed to reduce forecast uncertainties. In this study, the performance of ensemble precipitation forecasts of seven numerical models in 2019 Gorganroud floods was studied. Initially, the precipitation forecasts of the seven numerical models were bias-corrected via gamma quantile mapping method. Then ensemble streamflow forecasts were obtained by ensemble precipitation of seven models using the GR4J conceptual rainfall-runoff model. Based on the optimized parameters, ensemble streamflow forecasts were performed with precipitation forecasts inputs while uncertainty of the models was analyzed based on inputs to the hydrological model. The results showed that the bias correction had a great impact on the improvement of flood forecast in the study basin such that the uncertainty bands of the ECMWF, NCMRWF and UKMO models well covered the observed flood values. P-factor and R-factor values of the ECMWF model was 0.5 and 0.96, respectively; however, the upper and lower bands of ECMWF model was symmetrical. The NCEP and CMA models had poorer performance in flood forecast compared with other models so that their P-factor values were 0.2 and 0.15, respectively. The JMA model overestimated the 2019 flood. Although the ECCC model bands covered 65% of the observed flood values, the gap between the upper and lower bands was quite high. Overall, the results of a number of NWP models in the study basin were satisfactory and their application is generally recommended for flood warning systems.

Flood is one of the most destructive natural disasters that have historically done considerable harm to humans. Unfortunately nowadays despite great technological improvements, there is still no acceptable prevention method for this natural phenomenon. This is mainly due to lack of knowledge or the limited facility in this field. The main aim of this article is to recognize the patterns of synoptic systems leading to the torrential rainfall in the south of Iran in order to predict the showers that lead to floods. These patterns also helps the better use of the torrential rainfall. Twenty major flood events were selected according to the high discharge, the daily rainfalls, and the related damages. The synoptic maps were investigated and were classified into four different groups regarding their arrangement, patterns, extensions, and the movement directions. With synoptic analysis of severe storms that lead to heavy rains and torrential rainfall, we could realize that these storms are mainly caused (influenced) by strengthening the center of Sudan heat low and Red Sea convergence zone. We also showed that sometimes these storms are caused by simultaneous merging of the Sudan heat low system and the Mediterranean frontal system changing into a dynamic or thermodynamic system.