Urban Green Infrastructure Visual Accessibility Study in the City of Brussels
Abstract
The ecological and environmental benefits provided by Urban Green Infrastructure (UGI) are undeniable, however, without a detailed survey of the pre-existing UGI characteristics, and visual accessibility measures, the potential socio-environmental and urban health benefits UGI implementations could bring along may not be achieved. This research proposes a novel methodology to quantify UGI, and its visual accessibility, combining remote sensing and image processing technologies. Urban squares located in the Brussels Capital Region have been studied.
Through an aerial survey, the Normalized Difference Vegetation Index (NDVI) has been computed making use of two methodologies. Method #1 involved data for a single day in April 2020 and at a 0.05m spatial resolution. ii) Method #2 involved the processing of Copernicus Sentinel-2 data using the Google Earth Engine (GEE) at 10m spatial resolution for the entire spring of 2020 at 5-day intervals. A strong Pearson correlation of 0.92 was observed between both methods with a slight larger variability for Aerial method #2, highlightinening the advantage of using a temporally averaged dataset which also captures the weekly NDVI changes. Through a pedestrian-level survey and using 360° Google Street View (GSV) imagery and semantic classification, novel UGI indicators to document the UGI types, their ratios, and their position relative to the camera have been computed. These indicators include greenery position indicators such as Green Vertical angle GVa, the Tree Vertical angle TVa, or the Grass Vertical angle (GRVa), which provide insights about the relative position and scale of the UGI. The results have shown that the Green ratio (Gr) and the NDVI show a strong Pearson correlation of 0.83 validating the results of the proposed GSV approach. Furthermore the green ratio and tree ratio indicators are highly correlated (Pearson correlation of 0.99), explaining that the visual green infrastructure with highest impact on pedestrian level perception is dominated by trees and not by grass patches and bushes. The proposed indicators could be adopted by decision-makers, to render green visual accessibility more equitable.- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2022
- Bibcode:
- 2022AGUFMGH51A..09L