Potential sediment supply fluxes associated with greenfield residential construction

Residential construction is among the most intensive land disturbance activities in cities, increasing runoff and sediment export. In growing cities, large-scale residential construction areas are ubiquitous on the city fringes, but their relative contribution to city-wide land disturbance patterns is unknown. So is the relative sediment supply potential of different phases of construction. Such insights are required to guide prioritisation of sediment management measures at both city and development scales. A multi-scale assessment was therefore undertaken of land development spatial and temporal patterns. The assessment used remotely-sensed elevation data to characterise earthworks intensity and extent across different zones of a large city (Melbourne, Australia), and fine-scale aerial imagery surveys of bare soil and impervious cover to characterise watershed disturbance over time in new construction areas. Earthworks extent and intensity were greatest in outer growth areas where they were dominated by residential construction. Within residential developments, soil disturbance and drainage alteration proceeded in a consistent temporal sequence, lasting around 4–9 years in total. Bare soil cover approached 100 % during bulk earthworks. A conceptual model of changes in potential runoff and sediment supply over a series of construction phases suggests that potential sediment supply peaks with bare soil cover, but then remains high during subsequent road and house construction as imperviousness and drainage connection increase. Policies to reduce sediment loads to streams and receiving waters should place greater priority on improving construction erosion and sediment control in growth areas, and should target both development-scale and lot-scale activities. The conceptual model developed here provides a generalised hypothesis which could guide monitoring studies to target relevant construction phases at a fine temporal resolution and spatial scale