Carbon Stocks Dynamics of Urban Green Space Ecosystems Using Time-Series Vegetation Indices

The quantification of carbon stocks has emerged as a critical global issue due to its vital role in ecosystem services amid increasing urbanization and the impacts of global climate change. This study assesses carbon stocks in urban green space (UGS) ecosystems using time-series remote sensing data from 2014 to 2022. Carbon stock computation was derived from vegetation indices obtained from Landsat 8 satellite sensors, specifically the red and near infrared (NIR) bands with central wavelengths of 0.665 µm and 0.705 µm, respectively. The results, based on nine years of annual data, indicate a 24% increase in carbon stocks within UGS ecosystems. However, year-to-year transitions showed significant fluctuations, with a 19% decrease in carbon stocks from 2017 to 2019, and notable increases of 25% and 40% during the 2015–2016 and 2019–2020 periods, respectively. Spatially, carbon stock fluctuations were most pronounced in agricultural ecosystems, which are vulnerable to climate change, especially during El Niño-Southern Oscillation (ENSO) and positive Indian Ocean Dipole (IOD) events that influenced vegetation dynamics, particularly in low-density areas. The most substantial contributors to carbon stocks, exhibiting relatively stable and adaptive patterns to climate change, were mangrove and urban forest ecosystems. From a state-of-the-art perspective, this research addresses a gap in the literature where previous studies focused on calculating carbon for specific periods using various model approaches. Our implementation of a new time series analysis demonstrates that carbon stocks are dynamic, as evidenced by our findings. The results underscore the importance of preserving urban forest ecosystems, which play a significant role in climate change mitigation and the reduction of urban greenhouse gas emissions.