Urban areas and their surrounding ecosystems are under increasing strain globally as growing populations, accelerating economic development and increasing agricultural production impact land use, transport networks and air and water quality. An EU-funded project is addressed the challenge of urban sustainability with a novel integrated modelling approach to guide decision-makers towards making cities cleaner, greener and more liveable.
© sdecoret - fotolia.com
Led by Marie SkÅ‚odowska-Curie research fellow Harutyun Shahumyan, the GeoSInPo project has resulted in the creation of a unique open-source toolset, enabling disparate socio-economic and environmental models to be linked as a means to project future trends and evaluate the impact of urban development decisions.
The platform overcomes key model integration challenges, making use of a wide variety of information and indicators from different sources.
“Decision-makers need adequate tools to better understand and evaluate the effects of policy interventions in urban regions,” he says. “This has already led to the development of numerous mathematical and geospatial models focusing on various discipline-specific areas such as land use, transport, emissions and air and water quality. However, the interconnected character of human and natural systems requires integrated decision-making and therefore integrated modelling linking different disciplines.”
Developing an entirely new comprehensive model integrating various disciplines from the ground up would be a costly and time-consuming process. Instead, Shahumyan focused on how to best combine existing discipline-specific tools and independent models to devise a quick and effective solution to generate integrated models for different urban regions.
The work is set to significantly contribute to global understanding of human activity and environmental linkages in urban areas, enabling improved policy development and decision-making focused on ensuring urban sustainability, says Shahumyan.
Loose models and software wrappers
To achieve this, researchers from the US and Ireland needed to overcome numerous interoperability obstacles among systems developed in different programming languages and software environments with various licensing restrictions.
The solution they chose is called a ‘loose model coupling approach’ and involves ‘software wrappers’, a technique used in data mining. The approach enables data to be extracted and exchanged between different models without modifying the underlying systems. This ensures flexibility, portability, enhanced interoperability and performance.
The tools have been applied by the National Center for Smart Growth at the University of Maryland in the US, where Shahumyan spent two years of the Marie Curie fellowship, and at the School of Architecture, Planning and Environmental Policy at University College Dublin in Ireland, where he is continuing to lead research on data analytics, geospatial technologies, integrated modelling and geographic information systems.
“Our coupling toolset has been successfully applied in the Baltimore-Washington region by the University of Maryland to combine five independently developed models covering land use, transportation and mobile and building emissions to simulate alternative regional development scenarios through to 2040,”
Shahumyan says. “Translating the effect of socio-economic alterations into nutrient loading in the Chesapeake Bay, for example, will help us explore the changes in flow and nutrient loads and inform decision-makers to design more effective public policies and restoration plans.”
Meanwhile in Ireland, the GeoSInPo tools are being applied to model urban, industrial and agricultural impacts on water quality in the Greater Dublin Region, enabling the estimation of annual nutrient losses under different regional development scenarios through to 2026.
Outputs from the integrated modelling suite also include many useful socio-economic indicators, covering population and employment, transport flow, land use, building and mobile emissions and potentially even public health indicators.
“The results of these two case studies are very promising,” Shahumyan says. “The independently developed models smoothly exchange data in a single modelling platform, allowing non-technical users to focus on analysis and results. Moreover, this approach allows new models to be added relatively easily: we are planning to enhance our suite with water quality and habitat models as well as health impact models.”
The source codes and supporting documentation generated by GeoSInPo are published publicly and available for other researchers, enabling the platform to easily be applied to other models for other regions.