Elham Mehrabi Gohari; Roghaye Shahriyaripour; Ahmad Tagabadipoor; Seyed Roohollah Mousavi
Abstract
This study aims to evaluate and compare the efficiency of Artificial Neural Network (ANN), Regression Tree (RT) and Neuro-Fuzzy (ANFIS) models using a digital soil mapping framework to predict soil texture in a part of Sirjan province. Sampling was carried out at 84 observation points with a regular ...
Read More
This study aims to evaluate and compare the efficiency of Artificial Neural Network (ANN), Regression Tree (RT) and Neuro-Fuzzy (ANFIS) models using a digital soil mapping framework to predict soil texture in a part of Sirjan province. Sampling was carried out at 84 observation points with a regular grid of 2x2 km, and soil texture components were determined from the soil surface depth of 0 to 30 cm. Auxiliary variables included primary and secondary derivatives of the digital elevation model (DEM), a geomorphological map and remote sensing (RS) spectral indices. The appropriate variables selected using the Principal Component Analysis (PCA) feature selection method. Based on PCA, eight topographic variables and six vegetation indices and spectra from RS selected to predict soil texture components (sand, silt and clay). The efficiency of the models was evaluated using coefficient of determination (R2), mean error (ME), root mean square error (RMSE) and normalised root mean square error (nRMSE). The RMSE values in the neuro-fuzzy model compared with the regression tree model. The results of the neuro-fuzzy model were 1.43% for clay, 1.98% for sand and 2.1% for silt, which were 4.32%, 5% and 4.54% lower respectively compared to the regression tree model. The results of this study showed that the ANFIS model was more accurate in predicting clay, silt and sand compared to ANN and RT. Also, the geomorphology map, topographic wetness index, multi-resolution valley bottumn flatness index and Landsat 8 bands 5 and 6 had the highest relative importance in predicting soil texture components.
Mojtaba Nassaji zavareh; Bagher Ghermezcheshmeh; Fatemeh Rahimzadeh
Abstract
Daily constant discharges are needed estimating daily discharge in the hydrological model. The different number of statistical years, statistical deficiencies, and measurement error leads to the formation of time series with an uncommon time base. Hence the reconstruction of daily discharge data is of ...
Read More
Daily constant discharges are needed estimating daily discharge in the hydrological model. The different number of statistical years, statistical deficiencies, and measurement error leads to the formation of time series with an uncommon time base. Hence the reconstruction of daily discharge data is of paramount importance. In this research, daily discharge was reconstructed in two stages in one of the upstream of Karoun River. In both stages of research, daily discharge data from two upstream stations were used to reconstruct daily discharge of the downstream station using artificial neural networks, neuro-fuzzy and two variables regression methods. In the second stage, the magnitudes of discharge, based on dry, normal and wet years was used to reconstruct the daily discharge. The results showed higher accuracy in the artificial neural network and neuro-fuzzy methods compared to two variable regression methods in the reconstruction of daily discharge. Multi-layer perceptron model has better potential among all different method of artificial neural network and neuro-fuzzy models. Classification of discharge into dry, normal, and wet years decreases error in the reconstruction of daily discharge. Based on the mean relative error (MRE), error in reconstruction of daily discharge is the least in normal, wet, and dry years, respectively
Sadegh Tali-Khoshk; Mohsen Mohseni Saravi; Mahadi Vatakhah; Shahram Khalighi-Sigarodi
Abstract
Because of insufficient factors including facilities, budget, human resources as well as time watershed operation is not applicable throughout the basin. As a result, watershed operation should be performed in the sub-basins in which is more affectionate and the risk frequency of some events such as ...
Read More
Because of insufficient factors including facilities, budget, human resources as well as time watershed operation is not applicable throughout the basin. As a result, watershed operation should be performed in the sub-basins in which is more affectionate and the risk frequency of some events such as destruction, degradation; physical and financial damage and also flooding are considerably high. In addition, due to hydrometric stations, defects or the lack of stations in some areas, some efforts have been made experts recently to assess and consequently introduce some novel and reliable methods for prioritizing on the basis of current data obtained from sub-basins features of different geographical regions. In current study, the utilization possibilities of neuro-fuzzy technique and SCS in HEC-HMS model that have different potential to examine a wide range of advantageous and disadvantageous in making various decisions were studied. To determine the prediction accuracy of these methods, the rate of flow and sediment output of Taleghan sub-basins were taken over one year. The results of these methods were then compared with the maximum two-year return period flow observations. Our results revealed that in making prioritization, neuro-fuzzy as compared with the SCS method can produce the best prediction as long as the coefficients of errors, efficiency compared to the observational data and predictions are taken into account.
