Evaluating urbanization effects on biomass density using a hybrid AI model: a case study

Urbanisation profoundly influences both socioeconomic structures and ecological systems worldwide, highlighting the critical need for integrated approaches to sustainable urban planning and environmental management. This study investigates the effects of population growth and urban expansion on natural vegetation cover in Fethiye, Muğla, Turkey-a coastal region exemplifying the delicate balance between urban development and ecological preservation. To model vegetation dynamics, Normalised Difference Vegetation Index (NDVI) values are projected up to the year 2032 using Long Short-Term Memory (LSTM) networks applied to time series data. Land Surface Temperature (LST) and Normalised Difference Built-up Index (NDBI), derived from MODIS and Landsat satellite imagery spanning 2013-2023, are used as input parameters. The predictions are interpreted within the framework of anticipated urban growth, demographic trends, and stable climate assumptions. Furthermore, a hybrid model combining Discrete Wavelet Transform with LSTM (DWT-LSTM) is proposed to enhance predictive performance, resulting in a 9.1% improvement in NDVI accuracy. This improvement has been validated against CORINE land cover data, demonstrating the model's robustness and generalisability. The findings offer crucial insights for urban planners and environmental policymakers. By quantitatively linking urbanisation indicators with ecological degradation, this study provides a data-driven foundation for mitigating the adverse environmental impacts of rapid urban development. The results underscore the utility of advanced AI-based models in forecasting ecological change and supporting informed, sustainable land-use strategies.