Assessing CO2 Absorption of Urban Trees Using NDVI, SAVI, and MSARVI in Salatiga, Indonesia

Abstract

Urban areas are influential in contributing to CO₂ emissions because, in urban areas, there can be an increase in land use and higher community mobility. Trees can reduce CO₂ emissions through photosynthesis, which absorbs CO₂ emissions. Nowadays, remote sensing technology can estimate the number and presence of trees and their CO₂ absorption capacity.  Therefore, this study aims to find the best vegetation index approach method and assess the CO₂ absorption capacity of trees in Kalicacing and Mangunsari Urban Villages in Salatiga City using remote sensing. The method used to obtain data on the number of trees and tree species is crown delineation using orthophoto data. Meanwhile, for the estimation of CO₂ absorption, data on Diameter at Breast Height (DBH) and tree height were measured using Worldview-2 imagery with vegetation indices. The estimation model used three vegetation indices: Normalized Difference Vegetation Index (NDVI), Soil Adjusted Vegetation Index (SAVI), and Modified Soil and Atmospheric Resistant Vegetation Index (MSARVI). Statistical analysis is used to generate regression equations, and RMSE is used to determine the model's error rate. This study found that orthophoto data, Worldview-2 image data, and measurement and observation data in the field can map the ability of each tree species to estimate CO₂ absorption with a reasonable prediction model. Approximately 10,102 trees with 54 tree species are identified in Kalicacing and Mangunsari villages. NDVI, SAVI, and MSARVI have error rates of around 0.2. The MSARVI index had the smallest RMSE of 0.285. The range of ability of trees in Kalicacing and Mangunsari Urban Villages to absorb CO₂ emissions is 0.32 to 2.26 tons/tree. Trees that have the highest CO₂ absorption capacity are rudraksha trees (Elaeocarpus sphaericus) and rubber banyan trees or rubber plants (Ficus elastica).