Coupling urban expansion models and hydrological models: how important is the prediction of spatial patterns?

It is well known that urban expansion has a severe impact on the surface water balance by transforming
vegetated covers into sealed surfaces. This transformation causes changing fluxes of evapotranspiration,
surface runoff and groundwater recharge. In order to estimate the impact of land cover changes on the
surface hydrology, hydrological models are often coupled with different types of land cover change models.
It is, however, not clear to what extent spatially-explicit urban expansion scenarios provide an added
value in comparison with non-spatial urban expansion models at different scale levels. The objective
of this paper is to acquire a better insight in the importance of scale effects involved in the coupling
of urban expansion scenarios and hydrological models. The relative importance of using different projections
of both (i) quantity and (ii) spatial patterns of urban expansion was analysed at four different
scale levels. The highly urbanised Flanders-Brussels region was taken as an example application. Twelve
different urban expansion scenarios for 2025 and 2050 were developed and subsequently used as an
input in a spatially-distributed water balance model. The results obtained suggest that at the level of the
Flanders-Brussels region, an accurate estimation of the quantity of urban expansion should get priority
over an accurate projection of the spatial patterns. However, the importance of using accurate projections
of the spatial pattern of urban expansion increases systematically at local scale levels. A uniform strategy
for coupling urban expansion models and hydrological models thus seems inappropriate. These findings
are highly relevant for water management and spatial planning policymakers that typically operate at
different administrative levels.