Wyniki wyszukiwania - BazTech

1

Investigation of Transition Possibilities between Drought Classifications Using Standardized Precipitation Index for Wet and Dry Periods – Lower Seyhan Plain, Türkiye Case

Şimşek Serin Değerli, Çapar Ömer Faruk, Turhan Evren

Journal of Ecological Engineering

|

2023

|

Vol. 24, nr 5

201--209

EN

In this study, the Karaisalı region of Türkiye, which has a semi-arid climate and is known to contain the extensive plains and rich water resources of the Seyhan Basin, was preferred as a study area for investigating wet and drought periods for a long timescale. Forty-one years of total precipitation data, between 1980 and 2020, belonging to the closest precipitation observation station located in the Karaisalı region were used. By using the Standardized Precipitation Index (SPI), which is one of the frequently used meteorological drought indices, drought classification probabilities, expected first transition period and residence time in each drought severity class values were calculated for the 12-month time scale. As a result of the study, it was determined that the most drought period took place in 2012 according to the examined time duration. In addition, the most wet period was observed in 2001. When various time scales were considered, SPI-3 and SPI-6 have Near Normal Wet periods, while SPI-9 and SPI-12 have Near Normal Drought periods. Extremely Wet periods were more numerous, while Extremely Drought periods lasted longer. In addition, 3 months after the end of the drought categories, it can be seen that the Wet and Drought periods change into Near Normal Wet and Near Normal Drought periods.

2

Monthly streamflow prediction and performance comparison of machine learning and deep learning methods

Ayana Ömer, Kanbak Deniz Furkan, Keleş Mümine Kaya, Turhan Evren

Acta Geophysica

|

2023

|

Vol. 71, no. 6

2905--2922

EN

Streamflow prediction is an important matter for the water resources management and the design of hydraulic structures that can be built on rivers. Recently, it has become a widely studied research field where data obtained from stream gauge stations can be utilized for creating estimating models by resorting to different methods such as machine and deep learning techniques. In this study, we performed monthly streamflow predictions by using the following data-driven methods of machine learning: linear regression, support vector regression, random forest and deep learning (DL) models to compare the performances of ML's and DL's techniques. A general workflow that can be applied to similar regions is presented. An estimating model containing six-input combinations and time-lagged streamflow data is improved by means of the autocorrelation function (ACF) and partial autocorrelation function (PACF). Furthermore, moving average is used as a smoothing technique to make the dataset more stable and reduce the effects of noise data. A comparative evaluation has been conducted to determine the performances of the above-mentioned methods. In this study, we proposed four different DL models and compared them with existing techniques. For the comparison of the results, we used evaluation criteria such as Nash–Sutcliffe efficiency (NSE), mean square error (MSE) and percent bias (PBIAS). The experimental results indicate that our bidirectional gated recurrent units (BiGRU) model outperforms both ML algorithms and existing solutions with 0.971 NSE, 0.001 MSE and - 1.536 PBIAS scores.

3

A Comparative Evaluation of the Use of Artificial Neural Networks for Modeling the Rainfall-Runoff Relationship in Water Resources Management

Turhan Evren

Journal of Ecological Engineering

|

2021

|

Vol. 22, nr 5

166--178

EN

Recently, Artificial Neural Network (ANN) methods, which have been successfully applied in many fields, have been considered for a large number of reliable streamflow estimation and modeling studies for the design and project planning of hydraulic structures. The present study aimed to model the rainfall-runoff relationship using different ANN methods. The Nergizlik Dam, located in the Seyhan sub-basin and one of the important basins in Turkey, was chosen as the study area. Analyses were carried out based on streamflow estimation with the help of observed precipitation and runoff data at certain time intervals. Feed Forward Backpropagation Neural Network (FFBPNN) and Generalized Regression Neural Network (GRNN) methods were adopted, and obtained results were compared with Multiple Linear Regression (MLR) method, which is accepted as the traditional method. Also, the models were performed using three different transfer functions to create optimum ANN modeling. As a result of the study, it was seen that ANN methods showed statistically good results in rainfall-runoff modeling, and the developed models can be successfully applied in the estimation of average monthly flows.

4

The Investigation of the Applicability of Data-Driven Techniques in Hydrological Modeling: The Case of Seyhan Basin

Turhan Evren, Keleş Mümine Kaya, Tantekin Atakan, Keleş Abdullah Emre

Rocznik Ochrona Środowiska

|

2019

|

Tom 21, cz. 1

29--51

EN

Proper water resources planning and management is based on reliable hydrological data. Missing rainfall and runoff observation data, in particular, can cause serious risks in the planning of hydraulics structures. Hydrological modeling process is quitely complex. Therefore, using alternative estimation techniques to forecast missing data is reasonable. In this study, two data-driven techniques such as Artificial Neural Networks (ANN) and Data Mining were investigated in terms of availability in hydrology works. Feed Forward Back Propagation (FFBPNN) and Generalized Regression Neural Networks (GRNN) methods were performed on rainfall-runoff modeling for ANN. Besides, Hydrological drought analysis were examined using data mining technique. The Seyhan Basin was preferred to carry out these techniques. It is thought that the application of different techniques in the same basin could make a great contribute to the present work. Consequently, it is seen that FFBPNN is the best model for ANN in terms of giving the highest R2 and lowest MSE values. Multilayer Perceptron (MLP) algorithm was used to predict the drought type according to limit values. This system has been applied to show the relationship between hydrological data and measure the prediction accuracy of the drought analysis. According to the obtained data mining results, MLP algorithm gives the best accuracy results as flow observation stations using SRI-3 month data.