Dataset and Code for: A Cognitive Digital Twin Framework for Urban Water Management and Microclimatic Optimization: A Case Study of 'The Spine', Madinaty, Cairo

Overview: This dataset and code package support the research study focused on modeling reference evapotranspiration ($ET_0$) and agricultural water dynamics using machine learning (Random Forest Regressor) based on meteorological variables. File Contents: Cleaned_Climate_Data.xlsx: Cleaned and preprocessed daily meteorological dataset containing 7 primary climate/weather variables (Maximum Temperature, Minimum Temperature, Relative Humidity, Precipitation, Wind Speed, Solar Radiation, and calculated FAO-56 $ET_0$). Rondom Forest The Spine.ipynb: Jupyter Notebook containing the complete Python implementation for data preprocessing, feature importance evaluation, and Random Forest model training and evaluation. Hardware_and_Model_Specifications.ipynb: Jupyter Notebook documenting the exact environment setup, hardware specifications (CPU, RAM, OS), and model hyperparameters. Hardware and Execution Environment Specifications.docx: Supplementary documentation detailing the cloud execution environment and hyperparameter configurations to ensure full research reproducibility.