Evaluating the sensitivity of the i-Tree Eco pollution model to different pollution data inputs: A case study from Warsaw, Poland

Abstract

Urban trees provide a range of ecosystem services, including air purification. This specific ecosystem service can be quantified using i-Tree Eco software, but the software has some limitations. One limitation is that the pollution model used in the software cannot take into account spatial differences in pollutant concentrations, which occur in every city due to the diversity of polluters. This study aimed to investigate to what extent this limitation of the i-Tree Eco pollution model can influence the results obtained. The study was conducted in such a way that the i-Tree Eco project was broken down into four sub-projects, in which data on the concentration of selected pollutants from different air quality monitoring stations were analyzed using the coefficient of variation (CV). Three stations were designated as "urban background" type stations and one was a "traffic" type station. The study revealed relatively low variation (0.05 < CV < 0.23) in estimated quantities obtained based on concentrations of pollutants retrieved from different air quality monitoring stations of the same type ("urban background" type stations) and relatively high variation (0.24 < CV < 0.57) in estimated quantities obtained on data retrieved from different types of stations (“urban background” versus “traffic”). When data on air pollutant concentrations used in the model is from the “traffic” station type, trees provide air purification ecosystem services with an annual value almost twice as high (EUR 310 000) as when the data used is from “urban background” stations type (EUR 165 000). The results highlight the importance of ensuring that appropriate air pollution data for running a particular type of i-Tree Eco project is used.