Mihai V. Herbei ¹ Codruța Bădăluță-Minda ² Cosmin A. Popescu ^1* Adina Horablaga ^1* Lucian Octavian Dragomir ^1* George Popescu ^1* Shuraik Kader ^3* Paul Sestras ⁴

• ¹ University of Life Sciences King Mihai I Timișoara, Timisoara, Romania
• ² Politehnica University of Timișoara, Timișoara, Romania
• ³ School of Engineering and Built Environment, Griffith Sciences, Griffith University, Nathan, Queensland, Australia
• ⁴ Technical University of Cluj-Napoca, Cluj-Napoca, Romania

The Hydrologic Modeling System (HEC-HMS), designed to accurately simulate precipitation-runoff processes in various watershed types, was employed in our study to establish a model for a particular watershed. Consequently, we planned to create a rainfall-runoff model to investigate the specific characteristics of floods, assess some pre-diction models, and issue assumptions about their viability, which could be beneficial in establishing flood warning systems. The model was developed using daily precipitation data collected from different rain gauge stations in the Gilort watershed, located in southern Romania. The study employed the Hec-GeoHMS terrain processing tool, utilizing a digital elevation design to build a hydrological model. The statistical indicators used to evaluate the runoff mechanisms, specifically regression, coefficient of determination, correlation coefficient, index of agreement (Willmott index), and the root mean squared error (RMSE), showed a strong relationship between the simulated and recorded flow of the watershed. The leaking model was assessed using other statistical parameters, including the deviation of runoff volumes (Dv=6.40%), Nash-Sutcliffe efficiency (NSE=0.908), and Kling-Gupta efficiency (KGE=0.901). These parameters confirmed that the simulated data closely matched the observed data, indicating an effective association, and were considered reliable indicators of the model's goodness of fit, ensuring its reliability and efficacy.