PhD defence: Hanwen Xu

Hanwen Xu defends his thesis:

"Landscape-driven stormwater management at the meso-level: a perspective from hydrological simulation, evaluation and multiobjective optimization"

Zoom Link

Supervisors:
Associate professor Ole Fryd, IGN
Postdoc Mark Taylor Randall, IGN

Assessment Committee:
Professor Jia Haifeng. Tsinghua University, China
Associate professor Roland Löwe, Technical University of Denmark, Denmark
Professor Lone Søderkvist Kristensen (chair), (IGN)

Abstract:
In the context of urbanization and climate change, the increasing challenges posed by urban flooding have necessitated a shift in the traditional stormwater management paradigm. The overall purpose of the presented PhD study is to explore new frameworks and methodologies to enhance stormwater management at the meso-level. The meso-level, characterized by block, community or subcatchment scales, offers a critical intersection between macro-level management strategies and micro-level locally distributed interventions, making it a vital field for research and practices. This study integrated hydrological simulation, evaluation and multi-objective optimization aspects to provide a multi-dimensional framework, adopting a landscape-driven perspective which emphasizes the role of landscape elements and the built environment in shaping stormwater management outcomes.

In this thesis, a mixed methods approach was employed, integrating both qualitative and quantitative techniques. The study began with a scope-focused literature review by bibliometric and content analysis. Then, simulations were conducted using GIS and SWMM models, developed on the Python-based platform, to analyze hydrological processes and topographic impacts. Multi-objective optimization was applied to manage trade-offs between hydrological benefits, terrain modification, and economic costs. Additionally, post-occupancy evaluation (POE), which included field surveys, questionnaires, and semi-structured interviews, was used to gather public perceptions of Sponge City construction effectiveness. This integrated approach comprehensively examined both technical and social dimensions.

The results revealed the multifaceted nature and the effectiveness of combining simulation, evaluation and optimization techniques in advancing landscape-driven stormwater management at the meso-level. Paper I concluded upward publication trends, leading countries, research elements and common tools. The focus has shifted from theory to specialized technical roadmaps, with spatial optimization and multi-functional integration as emerging priorities. Paper II developed the Terrain Modification Multi-Objective Optimization (TMMOO) framework, combining NSGA-II with DEM-based hydro-morphometric analysis to balance flow velocity, runoff path length, and earthwork costs. Its application in Høje Taastrup, Denmark, demonstrated effectiveness in generating diverse topographic planning solutions across different resolutions. Paper III further expanded the TMMOO framework by integrating SWMM-based hydrological simulation to optimize the layout of stormwater management measures, using bio-retention cells as an example. Paper IV introduced an 10 application workflow that integrated the TMMOO framework with the augmented reality-sandbox to provide assistant feedback for terrain design. Lastly, Paper V evaluated public perceptions in Quyuan park, a Sponge City pilot project, reflecting an overall positive response while lower satisfaction with water management and service management. The mismatch between technical objectives and respondent feedbacks implies a gap between SCC implementation and public awareness.

The key findings of this research demonstrate that integrating hydrological simulation, evaluation, and multi-objective optimization offers a comprehensive perspective to stormwater management at the meso-level. The simulation and optimization framework based on different sensitivity tests verifies the feasibility and generation efficiency of stormwater management strategies and solution sets. The research provides tools and methodologies, such as the TMMOO framework and the AR-Sandbox workflow, which can assist landscape architects and planners in optimizing stormwater management solutions. The emphasis on POE highlights the need for a more comprehensive understanding of the long-term impacts. This thesis discussed the interactional relationship between stormwater management simulation studies and field studies, as well as bottlenecks and future trends. Finally, the thesis shows an example of synthesizing knowledge from landscape architecture, hydrology and operations research to answer specific and generalizable meso-level stormwater management problems, thus contributing a more resilient, sustainable and socially inclusive urban water systems.

A digital version of the PhD thesis can be obtained from the PhD secretary at phd@ign.ku.dk before the defence. After the defence the thesis will become available from the Royal Danish Libary at kb@kb.dk.