• Subodh Prasad
• Pooja Tamta

Urban vegetation plays a critical role in mitigating environmental challenges such as air pollution, urban heat islands, and water scarcity. In rapidly urbanizing areas like New Delhi, India, understanding the spatio temporal dynamics of vegetation is essential for sustainable urban planning and ecological management. This study investigates vegetation dynamics using satellite derived Normalized Difference Vegetation Index (NDVI) data from the MODIS/Terra sensor over a period spanning 2000 to 2018. Advanced machine learning techniques—specifically, long short term memory (LSTM) neural networks—are employed to model and forecast NDVI time series, capturing both seasonal cycles and interannual variability. Two LSTM architectures (a single layer and a dual layer configuration) are developed and rigorously evaluated using metrics such as root mean square error (RMSE), mean absolute error (MAE), and the coefficient of determination (R²). The results demonstrate that New Delhi’s vegetation exhibits marked seasonal patterns with significant spatial heterogeneity. The dual layer LSTM model consistently outperforms the simpler architecture, with forecasting errors (RMSE) below 0.03. The findings provide valuable insights for urban planners and policymakers, suggesting that predictive remote sensing analytics can guide the design and management of urban green spaces. Future work is recommended to integrate additional environmental variables and to apply the framework in other metropolitan contexts.

NDVI, remote sensing, time series analysis, LSTM, vegetation dynamics, urban ecology, New Delhi

"Vegetation Dynamics of Delhi, India", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 2, page no.87-98, February-2019, Available :http://www.jetir.org/papers/JETIR1902H15.pdf

"Vegetation Dynamics of Delhi, India", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 2, page no. pp87-98, February-2019, Available at : http://www.jetir.org/papers/JETIR1902H15.pdf