Mapping Pervious Surfaces and Canopy Cover Using High-Resolution Airborne Imagery and Digital Elevation Models to Support Urban Planning

Abstract

Urban green infrastructure (UGI) has a key role in improving human and environmental health in cities and contributes to several services related to climate adaptation. Accurate localization and quantification of pervious surfaces and canopy cover are envisaged to implement UGI, address sustainable spatial planning, and include adaptation and mitigation strategies in urban planning practices. This study aims to propose a simple and replicable process to map pervious surfaces and canopy cover and to investigate the reliability and the potential planning uses of UGI maps. The proposed method combines the normalized difference vegetation index (NDVI), extracted from high-resolution airborne imagery (0.20 m), with digital elevation models to map pervious surfaces and canopy cover. The approach is tested in the Municipality of Trento, Italy, and, according to a random sampling validation, has an accuracy exceeding 80%. The paper provides a detailed map of green spaces in the urban areas, describing quantity and distribution, and proposes a synthesis map expressed as a block-level degree of pervious surfaces and canopy cover to drive urban transformations. The proposed approach constitutes a useful tool to geovisualize critical areas and to compare levels of pervious surfaces and canopy cover in the municipal area. Acknowledging the role of green areas in the urban environment, the paper examines the potential applications of the maps in the policy cycle, such as land use management and monitoring, and in climate-related practices, and discusses their integration into the current planning tools to shift towards performative rather than prescriptive planning.

Characterization and Mapping of Public and Private Green Areas in the Municipality of Forlì (NE Italy) Using High-Resolution Images

“…More densely urbanized areas, such as q6 and adjacent areas, are generally covered by artificial surfaces with a high degree of imperviousness. On the other hand, according to [71], bare soil and agricultural land may be confused with impervious surfaces with low NDVI values when the same surfaces are dry or barely vegetated, as observed for vineyards in our case study area.…”

Land-use classification based on high-resolution remote sensing imagery and deep learning models

“…In this study, we choose this dataset for the following two reasons: (1) The data depicts remarkable details of the ground features with sub-meter accuracy. (2) This dataset removes redundant information irrelevant to classification, such as coordinate information, enabling the model to focus on nothing other than the ground objects during the training process. In this dataset, we obtain 5 large remote sensing images of unmanned aerial vehicles (UAVs), each of which is not equal in size; the overall size ranges from 4000×2000 to 8000×8000.…”

“…Land-use mapping based on remote sensing imagery has become increasingly important in providing essential information for many applications, such as urban planning [ 1 , 2 ], ecological management [ 3 , 4 ], regional design of crop cultivation [ 5 , 6 ] and environmental assessment [ 7 ]. With the rapid development of remote sensing technologies, the spatial resolution of imagery has sharply increased over the past decade, providing opportunities to extract more precise land-cover information [ 8 , 9 ].…”

“…Compared with other research papers with comparative models, this study has the following advantages: (1) The spatial resolution of the dataset selected in this study have reached submeter level, and the details of ground objects are complex and diverse, which is in line with the development trend of current remote sensing images. (2) This study uses multiple evaluation indicators to evaluate four types of highly representative models from two perspectives: index evaluation and visual evaluation. Through this study, on the one hand, it can solve the current confusion about the differences between remote sensing image classification models, and on the other hand, it can provide reference for model selection in land-use map, urban functional area identification and natural resource management.…”

Identification of Urban Green Space Types and Estimation of Above-Ground Biomass Using Sentinel-1 and Sentinel-2 Data

“…Traditionally, accurate information on the distribution of UGS was obtained through field surveys at a considerable expense in terms of labor and time [14,15]. Automatic information acquisition and repeatable observations can be used to monitor UGS, by employing remote sensing image classification technology based on pixels or objects [16][17][18]. In addition, the object-oriented method focuses on the relationship between pixels and their neighbors, which can provide higher accuracy in UGS classification results, as an alternative to the traditional pixel-based image classification method, that can result in the "salt and pepper effect" [19][20][21].…”