چکیده مقاله
This research investigates urban pluvial flooding in Qom by integrating over six years 2018 2024 of field measurements, satellite rainfall data, and citizen flood reports The primary aim is to evaluate the reliability of satellite derived rainfall data, develop an effective flood prediction model, and strategically determine optimal rainfall gauge station locations Two Random Forest models were constructed one trained on ground based measurements and the other on satellite data with results indicating a slight performance advantage of satellite data in flood detection Model evaluation using metrics such as accuracy, precision, recall, and F1 score revealed that satellite based predictions achieve over 92% accuracy, with higher precision and recall for flood events, demonstrating the reliability of satellite data in urban flood forecasting Additionally, the study introduces a novel feature importance based method for site selection of rainfall stations, enhancing early flood warning capabilities Compared to traditional hydraulic models and deep learning approaches, the proposed machine learning framework offers a scalable, cost effective, and real time solution suitable for arid, rapidly urbanizing regions The incorporation of citizen flood reports, despite inherent uncertainties, proved valuable in supplementing scarce measurement data, facilitating improved flood prediction and strategic infrastructure planning
کلیدواژهها
نویسندگان
Fereshte Sadat Mousavi Pour
PhD Candidate, Department of Environmental Engineering, Kish international Campus, University of Tehran, Kish, Iran
Mohammad Hossein Niksokhan
Professor, Department of Environmental Engineering, Faculty of Environment, University of Tehran, Tehran, Iran
Jaber Soltani
Associate Professor, Department of Water Engineering, Faculty of Agricultural Technology, University College of Agriculture & Natural Resources, University of Tehran, Tehran, Iran
شیوه ارجاع
Mousavi Pour, Fereshte Sadat and Niksokhan, Mohammad Hossein and Soltani, Jaber,1404,Implementation of Flood Warning Systems in Urban Environments using Satellite and On-field Rainfall Data,14th International Congress on Civil Engineering,Tehran
ارائهشده در
مجموعه مقالات چهاردهمین کنگره بین المللی مهندسی عمران29 مهر 1404 · تهران