Monitoring urban green space using remote sensing derived vegetation indices in Colombo district, Sri Lanka

Abstract

This study investigated the use of Remote Sensing (RS)-derived vegetation indices for monitoring urban green spaces (UGSs) from Sentinel-2 RS multispectral imagery with a spatial resolution of up to 10 meters. Six vegetation indices, including the Normalized Difference Vegetation Index (NDVI), Enhanced Vegetation Index (EVI), Soil Adjusted Vegetation Index (SAVI), Modified Soil Adjusted Vegetation Index (MSAVI), Green Ratio (GR), and Transformed Vegetation Index (TVI) by using Google Earth Engine (GEE) platform. Derived the original and enhanced images were used to compare the aforementioned vegetation indices in order to assess four image quality parameters: Peak Signal to Noise Ratio (PSNR), Root Mean Square Error (RMSE), Standard Deviation (SD), and Correlation Coefficient (CC). The findings demonstrated the range of values for each vegetation index: NDVI (-0.398163 to 0.888742), EVI (-0.287905 to 0.615649), SAVI (-0.597015 to 1.00000), MSAVI (-0.495483 to 0.795898), GR (-0.199505 to 0.444334), and TVI (0.058906 to 2.32316). Among the indices, MSAVI exhibited the best fit with image quality parameters of PSNR, RMSE, SD, and CC measuring at 45.31, 0.0197, 0.06, and 0.9641, respectively. This outcome suggests MSAVI's better performance in estimating green vegetation areas compared to other indices in the study area. The study also discussed the limitations of using vegetation indexes to monitor UGS, including the influence of atmospheric conditions, sensor calibration, and data preprocessing techniques. Overall, this study is insightful which is valuable in terms of the effectiveness of different vegetation indices for UGS. The findings of this study can be used to inform future