EGE 12TH INTERNATIONAL CONFERENCE ON APPLIED SCIENCES
Stream Flow Prediction for Murad River Branches
Yazarlar:
Ibrahim A. HASAN
Mehmet Ishak YUCE
Yayıncı:
Academy Global Publishing House
Effective planning and management of water resources rely on addressing the uncertainties inherent in hydrological models. A key challenge in accurately assessing a basin's water resources is the limited availability of data for hydrologic modeling. This study employs a multiple linear regression (MLR)-based approach to predict streamflow (Q) by analyzing its relationships with temperature (T), potential evapotranspiration (PET), and precipitation (P), excluding topographical parameters of the study area. The research explores and compares the performance of four MLR models applied to four sub-basins that contribute to the Murad River Basin. The analysis utilizes recorded monthly data from meteorological stations spanning 40, 30, 29, and 33 years for Models 1, 2, 3, and 4, respectively, covering the dependent variable (Q) and independent variables (T, P, and PET). The comparison between observed and predicted streamflow (Q) demonstrates promising results, highlighting the significance of the proposed equation. The highest coefficient of determination (R²) achieved is 21.86% for Model 3, while Model 1 shows an F-ratio exceeding 45. Additionally, the P-value is zero across all models, underscoring the statistical significance and overall effectiveness of the study. The findings of this study offer critical insights for decision-makers by presenting a data-driven methodology for streamflow prediction in data-scarce regions. By showcasing the effectiveness of MLR models, this research facilitates the development of enhanced water resource management strategies, promoting improved planning and the sustainable utilization of basin water resources.
