Emrah Doğan

Modelling daily suspended sediment of rivers in Turkey using several data-driven techniques

Abstract The transport of sediment load in rivers is important with respect to pollution, channel navigability, reservoir filling, longevity of hydroelectric equipment, fish habitat, river aesthetics and scientific interest. However, conventional sediment rating curves cannot estimate sediment load accurately. An adaptive neuro-fuzzy technique is investigated for its ability to improve the accuracy of the streamflow—suspended sediment rating curve for daily suspended sediment estimation. The daily streamflow and suspended sediment data for four stations in the Black Sea region of Turkey are used as case studies. A comparison is made between the estimates provided by the neuro-fuzzy model and those of the following models: radial basis neural network (RBNN), feed-forward neural network (FFNN), generalized regression neural network (GRNN), multi-linear regression (MLR) and sediment rating curve (SRC). Comparison of results reveals that the neuro-fuzzy model, in general, gives better estimates than the other techniques. Among the neural network techniques, the RBNN is found to perform better than the FFNN and GRNN.

Discrimination of quarry blasts and earthquakes in the vicinity of Istanbul using soft computing techniques

Abstract The purpose of this article is to demonstrate the use of feedforward neural networks (FFNNs), adaptive neural fuzzy inference systems (ANFIS), and probabilistic neural networks (PNNs) to discriminate between earthquakes and quarry blasts in Istanbul and vicinity (the Marmara region). The tectonically active Marmara region is affected by the Thrace-Eskisehir fault zone and especially the North Anatolian fault zone (NAFZ). Local MARNET stations, which were established in 1976 and are operated by the Kandilli Observatory and Earthquake Research Institute (KOERI), record not only earthquakes that occur in the region, but also quarry blasts. There are a few quarry-blasting areas in the Gaziosmanpasa, Catalca, Omerli, and Hereke regions. Analytical methods were applied to a set of 175 seismic events (2001–2004) recorded by the stations of the local seismic network (ISK, HRT, and CTT stations) operated by the KOERI National Earthquake Monitoring Center (NEMC). Out of a total of 175 records, 148 are related to quarry blasts and 27 to earthquakes. The data sets were divided into training and testing sets for each region. In all the models developed, the input vectors consist of the peak amplitude ratio (S/P ratio) and the complexity value, and the output is a determination of either earthquake or quarry blast. The success of the developed models on regional test data varies between 97.67% and 100%.

Modelling of evaporation from the reservoir of Yuvacik dam using adaptive neuro-fuzzy inference systems

Adaptive neuro-fuzzy inference system (ANFIS) models are proposed as an alternative approach of evaporation estimation for Yuvacik Dam. This study has three objectives: (1) to develop ANFIS models to estimate daily pan evaporation from measured meteorological data; (2) to compare the ANFIS model to the multiple linear regression (MLR) model; and (3) to evaluate the potential of ANFIS model. Various combinations of daily meteorological data, namely air temperature, relative humidity, solar radiation and wind speed, are used as inputs to the ANFIS so as to evaluate the degree of effect of each of these variables on daily pan evaporation. The results of the ANFIS model are compared with MLR model. Mean square error, average absolute relative error and coefficient of determination statistics are used as comparison criteria for the evaluation of the model performances. The ANFIS technique whose inputs are solar radiation, air temperature, relative humidity and wind speed, gives mean square errors of 0.181mm, average absolute relative errors of 9.590%mm, and determination coefficient of 0.958 for Yuvacik Dam station, respectively. Based on the comparisons, it was found that the ANFIS technique could be employed successfully in modelling evaporation process from the available climatic data.

Prediction of swelling pressures of expansive soils using soft computing methods

Lateral and vertical swelling pressures associated with expansive soils cause damages on structures. These pressures must be predicted before the structures are constructed in order to prevent the damages. The magnitude of the stresses can decrease rapidly when volume changes are partly allowed. Therefore, a material, which has a high compressibility, must be placed between expansive soils and the structures in both horizontal and vertical directions in order to decrease transmitted swelling pressure on structures. There are numerous techniques recommended for estimating the swelling pressures. However, these techniques are very complex and time-consuming. In this study, a new estimation model to predict the pressures is developed using experimental data. The data were collected in the laboratory using a newly developed device and experimental setup also. In the experimental setup, a rigid steel box was designed to measure transmitted swelling pressures in lateral and vertical directions. In the estimation model, approaches of artificial neural network (ANN) and adaptive neuro-fuzzy inference systems (ANFIS) are employed. In the first stage of the study, the lateral and vertical swelling pressures were measured with different thicknesses of expanded polystyrene geofoam placed between one of the vertical walls of the steel box and the expansive soil in the laboratory. Then, ANN and ANFIS approaches were trained using these results of the tests measured in the laboratory as input for the prediction of transmitted lateral and vertical swelling pressures. Results obtained showed that ANN-based prediction and ANFIS approaches could satisfactorily be used to estimate the transmitted lateral and vertical swelling pressures of expansive soils.

Clustering seismic activities using linear and nonlinear discriminant analysis

Identification and classification of different seismo-tectonic events with similar characteristics in a region of interest is one of the most important subjects in seismic hazard studies. In this study, linear and nonlinear discriminant analyses have been applied to classify seismic events in the vicinity of Istanbul. The vertical components of the digital velocity seismograms are used for seismic events with magnitude (Md) between 1.8 and 3.0 that occurred between 2001 and 2004. Two, time dependent parameters, complexity and S/P peak amplitude ratio are selected as predictands. Linear, quadratic, diaglinear and diagquadratic discriminant functions are investigated. Accuracy of methods with an additional adjusted quadratic models are 96.6%, 96.6%, 95.5%, 96.6%, and 97.6%, respectively with a various misclassified rate for each class. The performances of models are justified with cross validation and resubstitution error. Although all models remarkably well performed, adjusted quadratic function achieved the best success rate with just 4 misclassified events out of 179, even better compared to complex methods such as, self organizing method, k-means, Gaussion mixture models that applied to same dataset in literature.

Development of artificial neural network for prediction of salt recovery by nanofiltration from textile industry wastewaters

This paper presents the use of artificial neural network (ANN) to develop a model for predicting rejection rate (R o) of single salt (NaCl) by nanofiltration based on experimental data-sets. The re...

Investigation of Lake Sapanca water pollution, Adapazari, Turkey

Lake Sapanca has been the only source of drinking and recreational water for the city of Adapazari, Turkey. This paper reports a study of the variation of nutrient loading and trophic state of the lake, and also water quality parameters of Lake Sapanca compared to those of the neighbouring Lake Iznik. Through one year, samples were taken every three months from 15 different points on the streams feeding and draining off the lake. Nitrate, NO2‐N, NH3‐N, TKN, PO4‐P concentrations on the 12 streams fe and three draining off points were determined. Then, loading, discharge, and accumulation amounts of nitrogen and phosphorus causing eutrophication were calculated and the trophic state of the lake was determined. A simple model was used to analyse the response of Lake Sapanca when the phosphorus loading rate was changed. Through this model, the variation of different parameters (t, M, K, Q, V and A) with respect to phosphorus concentration (C) was studied to identify effects and results. The consequences of an eutrophic state and measures to protect the lake are also discussed.

Classification of River Yields in Turkey with Cluster Analysis

Clustering is necessary for lack of data in a basin based on hydrometeorological homogeneity. Even principal characteristics of river basins, such as; climate, geology, and topography affecting water yields are different, some of them yield similar hydrologic outcome. In this study, 1410 stations of Turkey Rivers were classified by the cluster analysis on the basis of hydrological homogeneity. Monthly average yields (m 3 /s/km 2 ) of 1410 river gauge stations on 26 river watersheds were used. It is aimed that the clusters to be homogeneous, the elements of the same cluster to be similar while they are not similar to those of a different cluster and the most meaningful groups to be made. The cluster number was found by using the agglomerative hierarchical cluster analysis method. Tests were conducted that stations from different geographic locations are considered in the same cluster independent of their geographic position. Turkey river basins were separated into 6 homogeneous regions and the yield distribution map of Turkey was obtained.

Aşağı Sakarya Nehrinde Taşkın Yayılım Haritalarının Elde Edilmesi

The Sakarya River Basin in Turkey frequently floods. The allure of riverside settlement and of nutrient-rich riverbank soil has led to extensive residential and agricultural development in flood plains. In this study, the 100 years return period possible flood carrying capacites of last 113 km of the Lower Sakarya Riverbed were investigated, also dam break and risk analyses were performed by applying different scenarios for the floods likely to occur. Flooding scenarios and water depth within the floodplain during these scenarios were calculated with the HEC-RAS software program and results were converted into a map in HEC-GeoRAS,ArcGIS 9x and ArcView 3.2 programs. As a result, it was observed that the Lower Sakarya River is susceptible to flooding. Recent observations of the study area confirm the study findings. This study tries to underscore the importance of taking into account the different scenarios regarding flood prevention and reduction studies.

Creating Flood Inundation Maps For Lower Sakarya River

The Sakarya River Basin in Turkey frequently floods. The allure of riverside settlement and of nutrient-rich riverbank soil has led to extensive residential and agricultural development in flood plains. In this study, the 100 years return period possible flood carrying capacites of last 113 km of the Lower Sakarya Riverbed were investigated, also dam break and risk analyses were performed by applying different scenarios for the floods likely to occur. Flooding scenarios and water depth within the floodplain during these scenarios were calculated with the HEC-RAS software program and results were converted into a map in HEC-GeoRAS,ArcGIS 9x and ArcView 3.2 programs. As a result, it was observed that the Lower Sakarya River is susceptible to flooding. Recent observations of the study area confirm the study findings. This study tries to underscore the importance of taking into account the different scenarios regarding flood prevention and reduction studies.