Impacts of urban development on urban water management – Limits of predictability

Abstract This study evaluates how random variations in urban development affect projections of urban water management over long time horizons, based on combined simulations of urban development and water management. The urban development was simulated based on existing procedural tools; however, novel procedures were implemented to allow the simulation of re-development in existing areas. The urban layout was parameterised based on the characteristics of existing building stock. A state-of-the-art hydraulic simulation tool was applied to simulate urban water systems. Experiments were conducted based on different degrees of random variation in the projections of urban development. These ranged from simply shuffling the sequences of urban developments foreseen in planning documents, to simulating urban growth with fixed growth rates but without zoning regulations, to simulations where both the growth rates and urban development locations were selected based on past development characteristics. Odense (a city in Denmark) was analysed using this paradigm. The impacts on the annual flow volumes, combined sewer overflows, exceedances of hydraulic pipe capacity, and surcharge volumes were evaluated. In general, the projections of the urban water management parameters were strongly affected by the random variations in the locations of urban development. In addition, most trends resulting from changing population growth rates could only be identified on spatial scales above 10 km2. Projections on longer time horizons were less affected by the random variations in urban development locations, as bounded simulation areas were considered. In conclusion, conceptual models are considered sufficient for performing combined simulations of urban development and urban water management in exploratory assessments.