A computational framework for novel UGS design

Rapid urban growth and construction cause environmental and ecological degradation, leading to negative effects on human health and well-being. Urban green systems, such as parks, help to mitigate climate change and heat waves in cities, and contribute to human health and well-being; yet, more research is needed to advance the understanding, planning and design of urban green systems. Such efforts can benefit from studying traditional rural green systems to recover knowledge and to adapt it for designing urban green systems. Historical systems display a rich diversity of traditional knowledge on plant management but require complex and considerable maintenance efforts. This frequently includes plant manipulation methods, such as coppicing, pollarding, pleaching or grafting for specific purposes. For instance: The German Tanzlinden are manipulated trees that provide social gathering points with comfortable micro-climate. Hedge-laying provides field barriers specific to their landscape and the cattle or crops they protect, while providing ecological corridors and wind- protection. These practices have been developed for specific contexts with the aim of achieving diverse but clearly defined functions. In stark contrast, contemporary urban greenery is usually designed with the aim to provide some benefits while minimizing maintenance. Procedures of tree care are standardized and minimized, thereby reducing the possible range and specific control of benefits considerably. Until today the diversity of traditional functional tree manipulation practices and resulting effects on benefits, such as ecosystem service provision, has not been systematically studied and their potential for the development of novel urban green systems remains largely unused. Today novel technologies can facilitate advanced survey, analyse, and better understand traditional green systems. The use of sensor technology, whether terrestrial or airborne by use of drones, computer-aided modelling and simulation, and knowledge engineering can facilitate high-level data acquisition and integration, analysis, and knowledge discovery. More complex forms of maintenance can be provided by robotic technologies. This project will utilise these means and aims at developing computer-aided workflows for designing and managing urban green systems that employ historic plant manipulation techniques. This project will focus on developing new methods to simulate growth reactions of trees to different manipulation practices and to model microclimatic effects and benefits; The aim is to develop an iterative design and management approach that integrates these methods to work towards a decision support system for the planning, design, and maintenance of novel green systems in cities.

The project developed an evidence-based digital workflow for designing and managing urban green systems to mitigate climate-change impacts. Its premise is that trees deliver microclimatic ecosystem services only when long-term growth and maintenance are embedded in design. Inspired by historic "trained" plane-tree canopies that are continually adjusted to meet shading goals, the project translates this iterative practice into a computational design-and-decision framework. Targets are formalized in a voxel model, enabling comparisons between intended and actual canopy development and supporting the choice of interventions. The work produced interoperable tools and validated workflows: Tree Information Modeling (TIM) standardizes digital representations of urban trees; empirical models predict resprouting after pruning; LiDAR-based methods estimate leaf area density distributions from branch scans; and a reinforcement-learning "pruning game" evaluates pruning strategies across tree states to find those best aligned with target canopy configurations. Biophysical evidence improves ecosystem-service assessment through species-specific biomass models, a 15-species leaf area index database capturing size and seasonal effects, and a plane-tree study quantifying the pruning tradeoff - reduced growth but lower water demand as well as lower stem diameter growth in the year after pruning. At the urban-design scale, an ontology and knowledge graph (UGS 4.0) integrates species information, evidence and best practices into transparent recommendations. A Munich public-square case study validated an end-to-end workflow combining knowledge-graph-supported 3D design, growth prediction, and pruning simulation to evaluate multiple ecosystem services while addressing spatial form, user needs, and long-term maintenance.