Climate-resilient urban planning with KLIMASCANNER: an AI-powered QGIS plugin

PANDORA 3.0 plugin: A new biodiversity ecosystem service assessment tool for urban green infrastructure connectivity planning

Network science and graph theory offer attractive models to describe urban phenomena and examine cases of urban planning and design. However, none of the existing network analysis tools consider simultaneously human visual perception and locations of urban activities. In this paper we introduce POI VizNet, a new QGIS plug-in that constructs various graphs of unobstructed lines of sight. The plug-in integrates an increasing amount of available GIS-based data of point of interest (POI) and visibility into one readily accessible analytical framework. Graphs are created between two types of decision points during urban travelstreet intersections and POIs, origins, and destinations of travel, by connecting these potential observer's decision locations in an open space between buildings. Therefore, the created graphs illustrate hypothetical visual trajectories of a person looking for a particular POI in the city. The graph is termed Integrative Visibility Graph (IVG) as it incorporates both navigational and functional aspects of the city (Figure 1a ). IVG examines connectivity of the particular location, i.e., predefined POI within the street network. In addition to IVG, the current version of the toolbox offers two separate modules of analysis corresponding to two additional types of visibility: Street Network Visibility Graph (SNVG) -creates visual connections between decision points of the street network, i.e., street intersections (Figure 1b ), and Point of Interest Visibility Graph (POIVG) -creates visual connections between POI that are visible from each other in a given building arrangement (Figure 1c ). In addition, POI VizNet provides advanced options to build graphs using a predefined viewing distance and perceptual perspective. Visibility graphs are constructed and visualised as new layers in QGIS and delivered as network files suitable for further exploration, analysis, and visualisation in various network software packages.

Abstract. Network science and graph theory offer attractive models to describe urban phenomena and examine cases of urban planning and design. However, none of the existing network analysis tools consider simultaneously human visual perception and locations of urban activities. In this paper we introduce POI VizNet, a new QGIS plug-in that constructs various 2-dimentianal undirected graphs of unobstructed lines of sight. The plug-in integrates an increasing amount of available GIS-based data of Point Of Interest (POI) and visibility into one readily accessible analytical framework. Graphs are created between two types of decision points during urban travel – street intersections and POIs, origins, and destinations of travel, by connecting these potential observer’s decision locations in an open space between buildings. In addition, POI VizNet provides advanced options to build graphs using a predefined viewing distance and perceptual perspective. Visibility graphs are constructed and visualised as new layers in QGIS and delivered as network files suitable for further exploration, analysis, and visualisation in various network software packages.

As climate change intensifies, ensuring thermal comfort in urban environments becomes a crucial challenge for public health and well-being. Urban planning plays a pivotal role in mitigating the effects of climate change by integrating climate-sensitive design strategies such as tree planting and facade greening. However, effective implementation requires an interdisciplinary understanding of the built environment, involving expertise from urban planning, ecology, and climatology. Additionally, city-based climate services face barriers such as limited data accessibility, communication challenges between stakeholders, and the lack of integrated, user-friendly tools.Microscale RANS (Reynolds Averaged Navier-Stokes) models offer high-resolution urban climate simulations (up to 5 m spatial resolution), incorporating complex interactions between terrain, buildings, land use, and vegetation. However, their computational intensity often makes them impractical for routine planning applications. Simulating a city’s baseline climate state alone can take weeks on commercially available servers, while additional assessments of climate adaptation measures or new urban developments further increase computational demands. Although high-performance computing resources are available in research institutions, their access and costs remain prohibitive for many urban stakeholders.To overcome these limitations, we developed KLIMASCANNER, an AI-powered QGIS plugin that integrates a neural network trained on RANS simulations to predict urban climate parameters such as air temperature (day and night), wind speed, and cold air flow for an autochthonous summer radiation day. By significantly reducing computational time while maintaining a high level of accuracy, the tool enables rapid assessments of urban development impacts on the local climate. KLIMASCANNER is designed to be accessible to urban planners, architects, and municipal decision-makers without requiring expertise in climate modeling. This facilitates informed decision-making and fosters climate-resilient urban design, bridging the gap between urban planning and climate science.

Abstract. Land cover classification is critical in various fields, including environmental monitoring, urban planning, and ecological assessment, facilitating informed decision-making processes. Traditional land cover classification methods often involve labor-intensive and time-consuming processes, relying on manual intervention and predefined algorithms. The emergence of deep learning techniques, particularly convolutional neural networks (CNNs), offers a promising solution to automate this process, albeit with complexities in implementation. This study addresses the limitations of existing Geographic Information System (GIS) software and plugins by proposing a novel approach utilizing the Pix2Pix architecture, a type of CNN, for automated land cover classification. The proposed Land Classification Plugin (LCP) integrates seamlessly with QGIS, offering an end-to-end solution for generating classified static maps. The methodology involves preprocessing data, utilizing the Pix2Pix model for image segmentation, and post-processing to produce georeferenced outputs. The development of the LCP involved extensive software and hardware configurations, including essential components like GDAL/OGR, PyTorch, and OpenCV. The plugin's architecture comprises a user-friendly interface for region selection, clipping, and classification aided by the Pix2Pix model. A layout manager feature also allows for the creation of composite maps for enhanced visualization. The accuracy assessment of the LCP demonstrated an overall accuracy of 83.40% across diverse land cover classes, indicating its efficacy in classification tasks. The plugin's capabilities offer significant potential for applications in land management, environmental surveillance, and urban planning, revolutionizing current practices in land cover classification within the realm of GIS software.

In the context of climate change, it is crucial to develop methods for assessing the impact of green infrastructures on urban climates through a co-constructive process with local planners. The study proposes a methodological framework for modelling geoprospective greening scenarios in collaboration with the metropolitan services of Dijon (eastern France, 260 000 inhabitants).A sensitivity analysis confirms the suitability of the Town Energy Balance (TEB) model coupled with Méso-NH for simulating the effects of tall and short vegetation on the UHI. It also highlights the need for detailed vegetation databases to ensure research remains aligned with local conditions, which is especially important in action research.The collaborative development of vegetation scenarios is anchored in the theoretical framework of geoprospective. This approach yields contrasting scenarios tailored to local contexts and leads to realistic greening guidelines derived from urban planning documents. These scenarios address a wide range of concerns raised by local planners, validating geoprospective as an effective tool for informing institutional decision-making. One scenario in particular stands out: it prioritizes the greening of available spaces in the city. It leads to add more than 100 000 trees and the equivalent of 4000 football fields. A dedicated QGIS plug-in was developed to support this scenario, paving the way for broader applications of scenario-based planning tools.This scenario focuses largely on greening private gardens and commercial/industrial areas. Results shows a reduction of the intensity (by a mean of 2°C) and spatial extent of UHI, prompting further consideration of the role of private garden greening.Overall, these findings highlight the need for an interdisciplinary methodological framework that integrates land-use planning, geomatics, and climate modelling. They also underscore the importance of combining action research with the translation of urban planning documents into scenarios, a strategy that could ultimately lead to the development of robust decision-support and urban planning tools.

Urban sprawl is a major cause of environmental change, indirectly affecting climate processes on both the global and local scale and impacting the livelihoods of people who are directly dependent on ecosystem services. In the case of rapidly sprawling cities, land cover monitoring is a spatial planning requirement that must keep pace with urban growth, for the purpose of providing timely responses to environmental change and thus reducing people’s vulnerability. Due to the lack of financial resources, Least Developed Countries need affordable methodology for rapid and effective land cover monitoring, suitable for low cost equipment. This chapter presents a methodology for monitoring land cover changes in Dar es Salaam, Tanzania, developed in the context of a project for the enhancement of local authorities’ capacity to assess vulnerability to climate change and mainstream adaptation objectives into urban development plans. This methodology relies on the classification of free Landsat images and is implementable using open-source software, with the specific purpose of making sustainable the continuous assessment of urban sprawl for Dar City Council’s planning services. The methodology phases are described, from preprocessing to processing. This includes the use of a free open-source plugin for QGIS, developed during the project, which allows for the semi-automatic classification of images. Classification results demonstrate the conspicuous urban growth of Dar es Salaam from 2002 to 2011, and provide insight into the relationship between urban sprawl and population growth.KeywordsLand cover changeRemote sensingLandsat imagerySemi-automatic classificationDar es Salaam