Irrigation
A. Kazemi Choolanak; F. Modaresi; A. Mosaedi
Abstract
IntroductionPredicting river flow is one of the most crucial aspects in water resources management. Improving forecasting methods can lead to a reduction in damages caused by hydrological phenomena. Studies indicate that artificial neural network models provide better predictions for river flow ...
Read More
IntroductionPredicting river flow is one of the most crucial aspects in water resources management. Improving forecasting methods can lead to a reduction in damages caused by hydrological phenomena. Studies indicate that artificial neural network models provide better predictions for river flow compared to physical and conceptual models. However, since these models may not offer reliable performance in estimating unstable data, using preprocessing techniques is necessary to enhance the accuracy and performance of artificial neural networks in estimating hydrological time series with nonlinear relationships. One of these methods is wavelet transformation, which utilizes signal processing techniques. Materials and MethodsIn this study, to evaluate the efficiency of discrete and continuous wavelet types in the Wavelet-Artificial Neural Network (WANN) hybrid model for monthly flow prediction, a case study was conducted on the Kardeh Dam watershed in the northeast of Iran, serving as a water source for part of Mashhad city and irrigation downstream agricultural lands. Monthly streamflow estimates for the upstream sub-basin of the Kardeh Dam were obtained from the meteorological and hydrometric stations' monthly statistics over a 30-year period (1991-2020). The WANN model is a hybrid time series model where the output of the wavelet transform serves as a data preprocessing method entering an artificial neural network as the predictive model. The combination of wavelet analysis and artificial neural network implies using wavelet capabilities for feature extraction, followed by the neural network to learn patterns and predict data, potentially enhancing the models' performance by leveraging both methods. The 4-fold cross-validation method was employed for the artificial neural network model validation, where the model underwent validation and accuracy assessment four times, each time using 75% of the data for training and the remaining 25% for model validation. The final results were presented by averaging the validation and accuracy results obtained from each of the four model runs. To evaluate and compare the performance of the models used in this study, three evaluation indices, Nash-Sutcliffe Efficiency (NSE), Root Mean Square Error (RMSE), and Pearson correlation coefficient (R), were employed. Results and DiscussionThe analysis of meteorological and hydrometric data in this study revealed that monthly streamflow in two time steps, T-1 and T-2, were the most effective predictive variables. Each of the two runoff variables of the previous month (Qt-1) and the previous two months (Qt-2) were analyzed by each of the Haar and Fejer-Korovkin2 discrete wavelet transforms and the two continuous Symlet3 and Daubechies2 wavelets at three levels. The results of each level of decomposition was given as input to the ANN model. The presented results at each decomposition level indicated that hybrid models could accurately predict lower flows compared to the single ANN model, and the estimation of maximum values also significantly improved in the hybrid models. Among the wavelets used, Haar wavelets exhibited the weakest performance, and the less commonly employed Kf2 wavelet showed a moderate performance. Since the Haar and Fk2 wavelets, with their discrete structure, did not perform well in decomposing continuous monthly streamflow data, continuous wavelet models outperformed discrete wavelet models. The hybrid models, combining wavelet analysis and artificial neural networks, demonstrated up to an 11% improvement over the performance of the single neural network model. ConclusionStreamflow is a crucial element in the hydrological cycle, and predicting it is vital for purposes such as flood prediction and providing water for consumption. The objective of this research was to evaluate the performance of different types of discrete and continuous wavelet models at various decomposition levels in enhancing the efficiency of artificial neural network (ANN) models for streamflow prediction. Since climate and watershed characteristics can influence the nature of data fluctuations and, consequently, the results of the wavelet model decomposition, choosing an appropriate wavelet model is essential for obtaining the best results. Considering the existing variations in the results of different studies regarding the selection of the best wavelet type, it is suggested to use both continuous and discrete wavelet types in modeling to achieve the best predictions and select the optimal results. Given that a lower number of input variables in neural network models lead to higher accuracy in modeling results, it is recommended to perform decomposition at a two-level depth to reduce input components to the neural network model, thereby reducing the model execution time.
Irrigation
A. Noori; J. Omidvar; F. Modaresi; K. Davary; S. Nouri; A. Asadi
Abstract
IntroductionLimited fresh water resources and access to these resources as well as providing food security for the growing world population have led researchers to make extensive efforts in the field of optimal management of water consumption and determining the cultivation pattern in different regions. ...
Read More
IntroductionLimited fresh water resources and access to these resources as well as providing food security for the growing world population have led researchers to make extensive efforts in the field of optimal management of water consumption and determining the cultivation pattern in different regions. Therefore, identifying cultivated crops in a region and determining their area can be very effective in land management and water allocation in these regions. With the growth and advancement of technology in the field of satellite and remote sensing in recent decades, the use of satellite images in order to identify types of land use and types of cultivated products has expanded greatly. Sentinel-1 (radar) and Sentinel-2 (multi-spectral) satellites have been very popular in agriculture due to their improved spatial resolution (10 meters) and appropriate time resolution (5 days for Sentinel 2 and 12 days for Sentinel 1).Materials and MethodsThe studied area is located downstream of the Fariman dam in an area of 22.51 square kilometers (5122 hectares) and the central coordinates are 35 degrees 41 minutes and 59 seconds north latitude and 59 degrees 50 minutes and 49 seconds east longitude. In order to classify satellite images and produce crop maps, ground observation data is needed to train the classification model and also evaluate the accuracy of the results. For this purpose, sample points were taken from different land uses in the region, using GPS. Since it was not possible to take enough samples for all land uses and crops in the determined border, a larger sampling area was selected. Then, all collected data were sorted and for each class, 70% of the data was randomly used to train the classification model and 30% was used to validate the obtained classification results. In the present study, Sentinel 2 satellite images for the first 6 months (crop season) of 2021 and 2022 and digital elevation image (DEM) of the study area were considered. According to the surveys conducted and the reports of the agricultural jihad of Fariman city, the main crops cultivated in the region include maize, tomato, sugar beet, wheat and barley. Therefore, according to the phenological stages of these products in the region, the appropriate time series of images was selected. The accuracy of the classified map was evaluated using the Kappa coefficient and overall accuracy.Results and DiscussionIn order to identify and separate the land use in the study area according to the major cultivated crops, first the agricultural calendar of the crops was determined. Then, satellite images were selected based on crop cultivation period. Based on the evaluation indexes of commission error, omission error, overall accuracy as well as the Kappa coefficient, it was observed that the identification of classes and land use was done well and with high accuracy, so that the overall accuracy for the classification map of 2022 is equal to 0.97 and the kappa coefficient value was 0.94. In order to compare land use changes during the two years 2022 and 2021, classification was also done for the images of the crop year 2021. Since the training samples of agricultural crops were not available separately and in sufficient numbers in the crop year of 2021, the classification map of this year was produced only based on the type of land use, and all crops in one class entered the classification model training process. The values of overall accuracy and kappa coefficient in 2021 were obtained as 0.97 and 0.95 respectively. According to the obtained results, the area of the orchard class has increased since 2021 compared to 2022. After repeated field visits to the study area and investigation of some land uses that had been changed and turned into orchard use, it was found that in some areas in 2022 there was the growth of villa gardens and in some areas the farmers have converted cropland to orchard (construction of an orchard). Even in some cases, the old orchard in the region was destroyed by the farmers and the land was fallow for 2 to 3 years (2021, fallow). In 2022, the farmer built a new orchard. It is also necessary to mention that fallow lands are included in the soil class depending on whether they are newly plowed or have no vegetation, and if weeds have grown on these lands, they are included in the rangeland class. ConclusionThe effective management of water resources from dams for agricultural purposes necessitates the identification of land use downstream of the dams, along with determining the types of crops and their respective areas. In this study, Sentinel 2 satellite images were employed to classify and delineate land use associated with agricultural cultivation downstream of the Fariman dam in Razavi Khorasan Province, spanning the crop years of 2021 and 2022. The results indicate that the Sentinel 2 satellite demonstrates a high capacity to differentiate between various types of land use and crops. The generated map depicting changes in land use and crop cultivation areas can be instrumental in water use planning and the allocation of water resources.
