Channel–floodplain connectivity during an extreme flood event: implications for sediment erosion, deposition, and delivery

The term connectivity has emerged as a powerful concept in hydrology and geomorphology and is emerging as an innovative component of catchment erosion modeling studies. However, considerable confusion remains regarding its definition and quantification, especially as it relates to fluvial systems. This confusion is exacerbated by a lack of detailed case studies and by the tendency to treat water and sediment separately. Extreme flood events provide a useful framework to assess variability in connectivity, particularly the connection between channels and floodplains. The catastrophic flood of January 2011 in the Lockyer valley, southeast Queensland, Australia provides an opportunity to examine this dimension in some detail and to determine how these dynamics operate under high flow regimes. High resolution aerial photographs and multi-temporal LiDAR digital elevation models (DEMs), coupled with hydrological modeling, are used to assess both the nature of hydrologic and sedimentological connectivity and their dominant controls. Longitudinal variations in flood inundation extent led to the identification of nine reaches which displayed varying channel–floodplain connectivity. The major control on connectivity was significant non-linear changes in channel capacity due to the presence of notable macrochannels which contained a > 3000 average recurrence interval (ARI) event at mid-catchment locations. The spatial pattern of hydrological connectivity was not straight-forward in spite of bankfull discharges for selected reaches exceeding 5600 m3 s–1. Data indicate that the main channel boundary was the dominant source of sediment while the floodplains, where inundated, were the dominant sinks. Spatial variability in channel–floodplain hydrological connectivity leads to dis-connectivity in the downstream transfer of sediments between reaches and affected sediment storage on adjacent floodplains. Consideration of such variability for even the most extreme flood events, highlights the need to carefully consider non-linear changes in key variables such as channel capacity and flood conveyance in the development of a quantitative ‘connectivity index’